#  @GaryMarcus Gary Marcus Gary Marcus posts on X about open ai, elon musk, agi, [----] the most. They currently have [-------] followers and [----] posts still getting attention that total [-------] engagements in the last [--] hours. ### Engagements: [-------] [#](/creator/twitter::232294292/interactions)  - [--] Week [-------] +7,287% - [--] Month [-------] +294% - [--] Months [----------] -58% - [--] Year [-----------] +12% ### Mentions: [--] [#](/creator/twitter::232294292/posts_active)  - [--] Week [--] +68% - [--] Month [---] +447% - [--] Months [---] -29% - [--] Year [-----] -32% ### Followers: [-------] [#](/creator/twitter::232294292/followers)  - [--] Week [-------] +0.55% - [--] Month [-------] +1% - [--] Months [-------] +10% - [--] Year [-------] +23% ### CreatorRank: [------] [#](/creator/twitter::232294292/influencer_rank)  ### Social Influence **Social category influence** [technology brands](/list/technology-brands) [celebrities](/list/celebrities) [finance](/list/finance) [stocks](/list/stocks) [countries](/list/countries) [us election](/list/us-election) [automotive brands](/list/automotive-brands) [social networks](/list/social-networks) [travel destinations](/list/travel-destinations) [vc firms](/list/vc-firms) **Social topic influence** [open ai](/topic/open-ai) #607, [elon musk](/topic/elon-musk), [agi](/topic/agi) #369, [6969](/topic/6969), [gary](/topic/gary), [ai](/topic/ai) #2724, [money](/topic/money), [microsoft](/topic/microsoft), [sama](/topic/sama) #1036, [$googl](/topic/$googl) #1712 **Top accounts mentioned or mentioned by** [@elonmusk](/creator/undefined) [@sama](/creator/undefined) [@ylecun](/creator/undefined) [@openai](/creator/undefined) [@rahll](/creator/undefined) [@theinformation](/creator/undefined) [@intuitmachine](/creator/undefined) [@iruletheworldmo](/creator/undefined) [@nytimes](/creator/undefined) [@kettlebelldan](/creator/undefined) [@grok](/creator/undefined) [@fchollet](/creator/undefined) [@ofquarks28701](/creator/undefined) [@geoffreyhinton](/creator/undefined) [@emollick](/creator/undefined) [@wsj](/creator/undefined) [@littmath](/creator/undefined) [@pmarca](/creator/undefined) [@ieeespectrum](/creator/undefined) [@karaswisher](/creator/undefined) **Top assets mentioned** [GARY (GARY)](/topic/gary) [Microsoft Corp. (MSFT)](/topic/microsoft) [Alphabet Inc Class A (GOOGL)](/topic/$googl) [Tesla, Inc. (TSLA)](/topic/tesla) [GrokCoin (GROKCOIN)](/topic/grok) [DeepSeek (DEEPSEEK)](/topic/deepseek) [Sora (SORA)](/topic/sora) [Robot Consulting Co., Ltd. (LAWR)](/topic/robot) ### Top Social Posts Top posts by engagements in the last [--] hours "Count me as one of the skeptics No AGI by end of [----] mark my words. But I otherwise think @elonmusks comments @nytimes on AI safety and AI regulation have been measured and on target" [X Link](https://x.com/GaryMarcus/status/1730003151971840419) 2023-11-29T23:17Z 207.8K followers, 92.4K engagements "Put it differently: this is really bad. What factors could explain it The sudden uptick of generative AI in [----] - particularly more costly image generation (per @sashamtls analysis of costs for inferences) - might be a factor Other explanations Not good. Not good at all. https://t.co/bkcyWfznyq Not good. Not good at all. https://t.co/bkcyWfznyq" [X Link](https://x.com/GaryMarcus/status/1745051197843931582) 2024-01-10T11:53Z 207.8K followers, 217.9K engagements "Some AGI-relevant predictions I made publicly long before o3 about what AI could not do by the end of [----]. Do you seriously think o3-enhanced AI will solve any of them in next [----] months" [X Link](https://x.com/GaryMarcus/status/1870469972495044778) 2024-12-21T14:02Z 207.8K followers, 83.9K engagements "By my math the drop in tourism *alone* will more than offset the total (and in many cases careless) cuts from DOGE. We will wind up with a poorer country that works less well with higher prices and less freedom that fewer people want to visit that makes fewer technological breakthroughs. The US is one of the worlds top travel spots pulling in over [--] million visitors last year and generating $13 trillion. But lately travel from Europe and Canada is down by up to 35% thanks to Trumps talk and scary border stories. Trump is bankrupting the US. https://t.co/f8QMkxW6W3 The US is one of the worlds" [X Link](https://x.com/GaryMarcus/status/1911147480902979744) 2025-04-12T20:00Z 207.5K followers, 73.1K engagements "AGI This decade Next decade Next century (or never) Later this century This decade Next decade Next century (or never) Later this century" [X Link](https://x.com/GaryMarcus/status/1927730816018215084) 2025-05-28T14:16Z 207.8K followers, 11.9K engagements "The chance that we will get to AGI (or even just more modestly AI that we can trust) without facing this core problem is pretty much zero. Yet another manifestation of the basic problem that I have been emphasizing since 1998: pure neural networks cannot reliably generalize beyond the training distribution. Yet another manifestation of the basic problem that I have been emphasizing since 1998: pure neural networks cannot reliably generalize beyond the training distribution" [X Link](https://x.com/GaryMarcus/status/1971339618206527925) 2025-09-25T22:22Z 207.6K followers, 17.5K engagements "BREAKING: Leaked October [--] letter from OpenAI to White House shows that OpenAI had begun asking for Federal guarantees over a week ago. Sams long attempted walkback yesterday on X after the huge backlash was a total lie. You cannot trust this man ever. Thats what Ilya saw. Here is an OpenAI document submitted one week ago where they advocate for including datacenter spend within the American manufacturing umbrella There they specifically advocate For Federal loan guarantees Sam Lied to everyone Again https://t.co/kkCiBx4mN8 Here is an OpenAI document submitted one week ago where they" [X Link](https://x.com/anyuser/status/1986798628552515647) 2025-11-07T14:11Z 208.9K followers, 1.3M engagements "@slow_developer [--]. Disinformation undermining democracy [--]. Massive cybersecurity incident [--]. Another AI winter once people realize how wildly GenAI was oversold" [X Link](https://x.com/GaryMarcus/status/1806175810732392787) 2024-06-27T04:00Z 207.9K followers, [---] engagements "Quote of the day maybe the year: *He is not an idiot Elon Musk is very smart very tech savvy. He could do a simple Google search of any of the disinformation he has seen in the last week and see it's untrue but instead he seeks to amplify it and perpetuate it. Page One in Edinburgh: He has vast wealth at his fingertips and he uses it for some of the most wicked evil Ive seen. @ukpapers $TSLA https://t.co/FjU0fB0NdZ Page One in Edinburgh: He has vast wealth at his fingertips and he uses it for some of the most wicked evil Ive seen. @ukpapers $TSLA https://t.co/FjU0fB0NdZ" [X Link](https://x.com/anyuser/status/1821996795482005894) 2024-08-09T19:47Z 208.9K followers, 184.9K engagements "Black box AI has been a massive possibly fatal mistake wrt alignment. This man speaks truth. @GaryMarcus @OpenAI Verify everything. The humanity needs a whitebox AI @GaryMarcus @OpenAI Verify everything. The humanity needs a whitebox AI" [X Link](https://x.com/GaryMarcus/status/1946781379859186065) 2025-07-20T03:57Z 208.4K followers, 110.9K engagements "AI agents arent really working. Why GenAI is like a C student who has secretly been given all the exams from the last decade. He may pass the exam but take away the entire internets worth of cheat codes and theres just not much there. (screenshot from @jeremyakahn)" [X Link](https://x.com/GaryMarcus/status/1950265087195042057) 2025-07-29T18:40Z 208.4K followers, 51.2K engagements "On reddit OpenAI is getting absolutely torched consistent with the dark analysis I posted earlier today (see below). For years I warned that OpenAI might take a serious hit if GPT-5 was disappointing. We finally soon see whether I was right. In a rational world their valuation would take a hit. They no longer have anything like a clear technical lead. GPT-5 is unlikely to be ahead of For years I warned that OpenAI might take a serious hit if GPT-5 was disappointing. We finally soon see whether I was right. In a rational world their valuation would take a hit. They no longer have anything like" [X Link](https://x.com/GaryMarcus/status/1953862343810486586) 2025-08-08T16:54Z 208.4K followers, 40.2K engagements "How have I handled all the negative diatribes and lies about me over the last seven years Being honest I am human and dont enjoy them. But I have stuck to my principles and my understanding of the cognitive sciences which have been a good guide. I paid a price for doing that. But I knew that the truth would eventually come out. And now suddenly all it once it has. Science is not a popularity contest; you cant bully your way to truth. And science cannot advance without people sticking to the truth even in the face of opposition. I hope people will remember that" [X Link](https://x.com/GaryMarcus/status/1956039883262407033) 2025-08-14T17:07Z 207.9K followers, 28K engagements "@ejae_dev @bridgemindai cool. now a try a (new) game that doesnt have a lot of direct existing code examples" [X Link](https://x.com/GaryMarcus/status/2022691697529061725) 2026-02-14T15:17Z 208.9K followers, [----] engagements "With the exception of Nvidia which is selling shovels in a gold rush most generative AI companies are both wildly overvalued and wildly overhyped" Gary Marcus Emeritus Professor of Psychology and Neural Science at New York University told @DeutscheWelle. "My guess is that it will all fall apart possibly soon. The fundamentals technical and economic make no sense."" [X Link](https://x.com/GaryMarcus/status/1986876930923372666) 2025-11-07T19:22Z 208.8K followers, 28.2K engagements "Thats exactly why I came back. Gary Marcus is back on X to bring balance back to the AI hype train. Gary Marcus is back on X to bring balance back to the AI hype train" [X Link](https://x.com/GaryMarcus/status/2022721951853350951) 2026-02-14T17:17Z 208.9K followers, 12.2K engagements "For years I said todays systems were laughably bad and was crucified for it. 🤷♂ A conversation with OpenAI on what to expect in the next [--] months - today's systems are laughably bad - ChatGPT not long term engagement model - models capable of complex "work" - like a great team mate working with u - shift towards verbal interfaces & beyond Multimodality https://t.co/XW91fNNceY A conversation with OpenAI on what to expect in the next [--] months - today's systems are laughably bad - ChatGPT not long term engagement model - models capable of complex "work" - like a great team mate working with u" [X Link](https://x.com/GaryMarcus/status/1787904229425954940) 2024-05-07T17:56Z 208.9K followers, 26.7K engagements "Exactly what I have been trying to tell you. The day of reckoning is coming. I genuinely don't understand why some people are still bullish about LLMs. I use GPT Grok Gemini Mistral etc every day in the hope they'll save me time searching for information and summarizing it. They continue to fabricate links references and quotes like they did from I genuinely don't understand why some people are still bullish about LLMs. I use GPT Grok Gemini Mistral etc every day in the hope they'll save me time searching for information and summarizing it. They continue to fabricate links references and" [X Link](https://x.com/GaryMarcus/status/1905171387641176225) 2025-03-27T08:13Z 208.9K followers, 41K engagements "I never thought I would live to see the President of the United States serving as Russias Press Secretary" [X Link](https://x.com/anyuser/status/1930321596889325624) 2025-06-04T17:51Z 208.9K followers, 851.3K engagements "Three thoughts on what really matters: [--]. Fuck cancer [--]. Friends are irreplaceable [--]. The new "Marcus test" for AI is when AI makes a significant dent on cancer May that happen sooner much sooner rather than later. In memory of my childhood friend Paul" [X Link](https://x.com/anyuser/status/1962212908840239366) 2025-08-31T17:56Z 208.9K followers, 321.4K engagements "Go back and read our [----] book Rebooting AI; it was all there. The limits of statistics and pattern-matching; the desperate need for world models. Six years of grief and Davis and I were right all along. Amazing. "Todays AIs dont understand environments; they just pattern-match inside them." Literally what critics have been saying for years now. "Todays AIs dont understand environments; they just pattern-match inside them." Literally what critics have been saying for years now" [X Link](https://x.com/GaryMarcus/status/1984292577479360747) 2025-10-31T16:13Z 208.9K followers, 139.4K engagements "Hot take on Gemini [--]. Points [--] and [--] might literally change the world. Also OpenAI is in trouble. Great model but its not AGI; has the same kind of problems all earlier models have had. Hallucinations and unreliability persist. In short scaling isnt getting us to AGI OpenAI has basically squandered the technical lead it once had; Google has caught up. What happens to OpenAI if Google undercuts OpenAI on price But biggest news was buried: it got better results than it is competitors *without using Nvidia GPUs* If Google were to make those TPUs commercially available at scale and reasonable" [X Link](https://x.com/GaryMarcus/status/1991183048629707109) 2025-11-19T16:33Z 208.9K followers, 313.8K engagements "translation : Deep Learning Hit a Wall new techniques needed to go forward; just like my infamous [----] paper said. go back and read it "We are no longer in the age of scaling we are back to the age of research.' Kaplan scaling laws are flattening at current frontiers autoregressive transformers exhausted for reasoning/planning/alignment gains. Next jump needs real architectural breakthroughs - test-time https://t.co/2hDIvJvF1F "We are no longer in the age of scaling we are back to the age of research.' Kaplan scaling laws are flattening at current frontiers autoregressive transformers" [X Link](https://x.com/GaryMarcus/status/1993499060981092753) 2025-11-26T01:56Z 208.9K followers, 78.6K engagements "It is one thing for everyone from @sama to @ylecun to @elonmusk and the Twitterverse to have attacked me literally for years for saying that scaling alone would not get us to AGI. Another to pretend now that that never happened now that I have largely been proven correct" [X Link](https://x.com/GaryMarcus/status/1993699159426130324) 2025-11-26T15:11Z 208.9K followers, 22.1K engagements "This place is toxic. For the last seven years I warned you that LLMs and similar approaches would not lead us to AGI. Almost nobody is willing to acknowledge that even though so many of you gave me endless grief about it at the time. I also warned you - first - that Sam Altman could not be trusted that OpenAI would lose its dominance that GPT-5 would not be all AGI and that LLMs lacked world models. That hallucinations would not go away. That out of distribution generalization was THE key issue. And that the economics of LLMs didn't make sense. And that the LLM companies would start seeking" [X Link](https://x.com/GaryMarcus/status/1993742337491784144) 2025-11-26T18:03Z 208.9K followers, 618.7K engagements "How did this work out Are LLM hallucinations largely gone by now So now the @FT platforms the same guy saying most the of the tasks lawyers and accountants do will be replaced in 12-18 months From the same company that said that GPT-5 would a giant humpback whale that blow away PhDs Where is the accountability The concern about CEOs conflict of interest in selling these narrative The view from skeptics LLM hallucinations will be largely eliminated by [----]. thats a huge deal. the implications are far more profound than the threat of the models getting things a bit wrong today. LLM" [X Link](https://x.com/GaryMarcus/status/2022041728375828865) 2026-02-12T20:14Z 208.9K followers, [---] engagements "@ProtagorasTO @FT 5% on known benchmarks (often gamed) is hardly largely eliminated" [X Link](https://x.com/GaryMarcus/status/2022152622833381718) 2026-02-13T03:35Z 208.9K followers, [----] engagements "@deredleritt3r @FT @DamienCharlotin Sir you have changed the argument from hallucinations arent a problem to if you watch LLMs like a hawk you will be fine. The former is false; the latter might be true. See my long post to Dean Ball a moment ago for examples in other domains that converge on the same" [X Link](https://x.com/GaryMarcus/status/2022331287730897209) 2026-02-13T15:25Z 208.9K followers, [----] engagements "I cant retweet Jimmy Apples but also wonder why people take him seriously [--] seconds of web search would have told him that you can already find ascii fishing code repositories on the web. 🙄" [X Link](https://x.com/GaryMarcus/status/2022720722699952561) 2026-02-14T17:12Z 208.9K followers, [----] engagements "Congress needs to bring in Zuckerberg and LeCun to discuss how their unilateral open-sourcing decision rapidly undermined the US advantage in Generative AI. Tomorrow. Meta doesnt just look clueless they look like they sold out the US. Meta doesnt just look clueless they look like they sold out the US" [X Link](https://x.com/GaryMarcus/status/1883671632595873926) 2025-01-27T00:21Z 208.9K followers, 215.2K engagements "Personal Update: My p(doom) has gone up. I dont foresee machines with malice any time soon. But I am starting to see the harm that powerful yet reckless humans who are indifferent to humanity could cause" [X Link](https://x.com/GaryMarcus/status/1944563227540721673) 2025-07-14T01:02Z 208.9K followers, 125.8K engagements "@nikostro LMK on Jan 1" [X Link](https://x.com/GaryMarcus/status/1962711831426027834) 2025-09-02T02:59Z 208.9K followers, [---] engagements "Wow. Just wow. @ylecun taking credit for my March [----] argument that scaling would hit a wall and that LLM would not bring us to AGI--after he initially attacked me for saying it and continued to promote them right up until ChatGPT ate has lunch--has to among the most intellectually dishonest things I have seen a scientist do in my entire life. @SchmidhuberAI was right about LeCun" [X Link](https://x.com/GaryMarcus/status/1993727856204198313) 2025-11-26T17:05Z 208.9K followers, 162.4K engagements "Over [--] million views for that @MattShumer_ blog (Something Big is Happening) and not a shred of actual data. Nobody thought to ask what are the data Does this match reality Is Shumer credible Did he give a single concrete example People gonna believe what they want to believe but here are [--] critical points most people missed: Shumers blog post is weaponized hype filled with vivid narrative and marketing speech but stumbles on the facts especially with respect to reliability. He gives no actual data to support this claim that the latest coding systems can (reliably) write perfect whole" [X Link](https://x.com/GaryMarcus/status/2021985561121374686) 2026-02-12T16:31Z 208.9K followers, 165.8K engagements "@neuralamp4ever @mattshumer_ Thank you. This heads up re paid AI influencers was very helpful and I immediately mentioned it on a BBC World interview that will air shortly. Original source here: https://www.cnbc.com/2026/02/06/google-microsoft-pay-creators-500000-and-more-to-promote-ai.html https://www.cnbc.com/2026/02/06/google-microsoft-pay-creators-500000-and-more-to-promote-ai.html" [X Link](https://x.com/GaryMarcus/status/2021996656418521208) 2026-02-12T17:15Z 208.9K followers, [----] engagements "How did this work out Are LLM hallucinations largely gone by now So now the @FT platforms the same guy saying most the of the tasks lawyers and accountants do will be replaced in 12-18 months From the same company that said that GPT-5 would be a giant humpback whale that would blow away PhDs Where is the accountability The concern about CEOs conflicts of interest in selling these narratives The view from skeptics LLM hallucinations will be largely eliminated by [----]. thats a huge deal. the implications are far more profound than the threat of the models getting things a bit wrong today. LLM" [X Link](https://x.com/GaryMarcus/status/2022042329033724168) 2026-02-12T20:17Z 208.9K followers, 241.8K engagements "No way Man who literally begged governments for wholesale exemption from copyright law so he could free ride on artists and writers bitches about . free riders. 🙄 🚨 BREAKING: OpenAI just WARNED US lawmakers that DeepSeek is using distillation to free-ride on capabilities developed by OpenAI this is part of the CCPs playbook: steal copy and kill @sama knows its over for ClosedAI Once DeepSeek drops V4 and R2 it will easily https://t.co/3zpUktXpsN 🚨 BREAKING: OpenAI just WARNED US lawmakers that DeepSeek is using distillation to free-ride on capabilities developed by OpenAI this is part of" [X Link](https://x.com/GaryMarcus/status/2022086729315102856) 2026-02-12T23:13Z 208.9K followers, 151.8K engagements "@SoniqueBang nobody will win. CCP is not going to crush Google and Google is not going to crush the chinese companies" [X Link](https://x.com/GaryMarcus/status/2022090808921620795) 2026-02-12T23:29Z 208.9K followers, 18.6K engagements "It would be one the great ironies of history if the biggest IP thief in history was taken down by mass IP theft. I think that sound is god laughing. 🚨 BREAKING: OpenAI just WARNED US lawmakers that DeepSeek is using distillation to free-ride on capabilities developed by OpenAI this is part of the CCPs playbook: steal copy and kill @sama knows its over for ClosedAI Once DeepSeek drops V4 and R2 it will easily https://t.co/3zpUktXpsN 🚨 BREAKING: OpenAI just WARNED US lawmakers that DeepSeek is using distillation to free-ride on capabilities developed by OpenAI this is part of the CCPs" [X Link](https://x.com/GaryMarcus/status/2022094395097723331) 2026-02-12T23:44Z 208.9K followers, 67.3K engagements "Honestly OpenAI should get the IP protections it deserves but only if it first properly licenses the IP of those they themselves ripped off. SCOOP: OpenAI warned lawmakers in a memo sent today to the House Select Committee on China that DeepSeek is using new obfuscated methods to continue to distill its AI models as well as those of other US frontier AI labs https://t.co/OsWxPRMF28 w/ @eastland_maggie SCOOP: OpenAI warned lawmakers in a memo sent today to the House Select Committee on China that DeepSeek is using new obfuscated methods to continue to distill its AI models as well as those of" [X Link](https://x.com/GaryMarcus/status/2022107527786508292) 2026-02-13T00:36Z 208.9K followers, 14.4K engagements "@deredleritt3r @FT and yet lawyers keep getting busted for fake cases in their briefs per @DamienCharlotins database Pretty much every day at much a higher clip than two years ago: https://www.damiencharlotin.com/hallucinations/ https://www.damiencharlotin.com/hallucinations/" [X Link](https://x.com/GaryMarcus/status/2022321895618789635) 2026-02-13T14:48Z 208.9K followers, 16.6K engagements "Are LLM hallucinations basically solved as a former Senior Policy Advisor at the White House @deanwball told me below based pure on anecdotal experience and without data No. Instead his post is symptomatic of how subjective finger-in-the-wind evaluations of AI often get things wrong. Here are some actual data. Law: Incidents in which lawyers have gotten busted for using fake cases are way up; @DamienCharlotins database listed around [---] cases less than a year ago and now has over [---] with new incidents reported ever day. Science: The situation is so bad that a team of Japanese reseachers just" [X Link](https://x.com/GaryMarcus/status/2022330800545698088) 2026-02-13T15:23Z 208.9K followers, 38.3K engagements "@deanwball With due respect I think you are off the mark here @deanwball. Some data that you should consider: Are LLM hallucinations basically solved as a former Senior Policy Advisor at the White House @deanwball told me below based pure on anecdotal experience and without data No. Instead his post is symptomatic of how subjective finger-in-the-wind evaluations of AI often get Are LLM hallucinations basically solved as a former Senior Policy Advisor at the White House @deanwball told me below based pure on anecdotal experience and without data No. Instead his post is symptomatic of how" [X Link](https://x.com/GaryMarcus/status/2022331655936147563) 2026-02-13T15:26Z 208.9K followers, [----] engagements "@ejae_dev wonder how much specific training there has been around it though" [X Link](https://x.com/GaryMarcus/status/2022381720574792148) 2026-02-13T18:45Z 208.9K followers, [----] engagements "@chatgpt21 @ejae_dev how can you say with confidence there tons of web hits for pelican bicycle svg and its a well-known benchmark" [X Link](https://x.com/GaryMarcus/status/2022391802314309893) 2026-02-13T19:25Z 208.9K followers, [---] engagements "60 million people on X panicked over a data-free screed about how AI is about to obviate the need for people and IBM triples its hiring the same week. Can't make that shit up" [X Link](https://x.com/GaryMarcus/status/2022393774576669139) 2026-02-13T19:33Z 208.9K followers, 17.9K engagements "@BishopBlougram @chatgpt21 @ejae_dev there may be real progress and maybe not and its really hard to know from the outside. best would be to try same system on a bunch of SVGs that are novel to their literature but challenging" [X Link](https://x.com/GaryMarcus/status/2022396749911068763) 2026-02-13T19:45Z 208.9K followers, [---] engagements "Reposting this/reupping it in light of xAI no longer having a safety team per scoop from @haydenfield Personal Update: My p(doom) has gone up. I dont foresee machines with malice any time soon. But I am starting to see the harm that powerful yet reckless humans who are indifferent to humanity could cause. Personal Update: My p(doom) has gone up. I dont foresee machines with malice any time soon. But I am starting to see the harm that powerful yet reckless humans who are indifferent to humanity could cause" [X Link](https://x.com/GaryMarcus/status/2022420618025804238) 2026-02-13T21:20Z 208.9K followers, [---] engagements "@aran_nayebi Love how you ignore ALL the data and most of the argument I presented and prefer your own anecdotal experience. Stick to the bit" [X Link](https://x.com/GaryMarcus/status/2022690573497569771) 2026-02-14T15:13Z 208.9K followers, [----] engagements "@aran_nayebi @peterevoss @BlueGuardrails did you try GPT-5.2-Pro is it vastly better" [X Link](https://x.com/GaryMarcus/status/2022708268481806406) 2026-02-14T16:23Z 208.9K followers, [---] engagements "We URGENTLY need a federal law forbidding AI from impersonating humans The night before I testified in the US Senate in May [----] the late philosopher Daniel Dennett sent me a manuscript that he called counterfeit people. It was published a few days later in The Atlantic. It started like this MONEY HAS EXISTED for several thousand years and from the outset counterfeiting was recognized to be a very serious crime one that in many cases calls for capital punishment because it undermines the trust on which society depends. Today for the first time in history thanks to artificial intelligence it" [X Link](https://x.com/GaryMarcus/status/2022761851725971667) 2026-02-14T19:56Z 208.9K followers, 13.6K engagements "Since @OpenAI still has not changed misleading blog post about "solving the Rubik's cube" I attach detailed analysis comparing what they say and imply with what they actually did. IMHO most would not be obvious to nonexperts. Please zoom in to read & judge for yourself" [X Link](https://x.com/anyuser/status/1185679169360809984) 2019-10-19T22:08Z 208.9K followers, [----] engagements "I will always stand in awe of Noam Chomsky. I just sent him my recent essay; [--] minutes later he replied with smart comments including a point I should have thought to include. Hes 93" [X Link](https://x.com/anyuser/status/1502767971931881472) 2022-03-12T22:06Z 208.9K followers, [----] engagements "Woker-than-thou" [X Link](https://x.com/anyuser/status/1607023594957045761) 2022-12-25T14:40Z 208.9K followers, 575.8K engagements "Speaking of bullshit anyone remember this" [X Link](https://x.com/anyuser/status/1637596790017912832) 2023-03-19T23:27Z 208.9K followers, 700K engagements "Take this seriously: @geoffreyhinton on AI possibly wiping out humanity: It's not inconceivable Is coding faster and having fun chatbots to play with worth a 1% risk of that coming true https://www.foxnews.com/tech/artificial-intelligence-godfather-ai-possibly-wiping-humanity-not-inconceivable https://www.foxnews.com/tech/artificial-intelligence-godfather-ai-possibly-wiping-humanity-not-inconceivable" [X Link](https://x.com/anyuser/status/1640324679968903177) 2023-03-27T12:07Z 208.9K followers, 1.1M engagements "a big deal: @elonmusk Y. Bengio S. Russell @tegmark V. Kraknova P. Maes @Grady_Booch @AndrewYang @tristanharris & over [----] others including me have called for a temporary pause on training systems exceeding GPT-4 https://futureoflife.org/open-letter/pause-giant-ai-experiments/ https://futureoflife.org/open-letter/pause-giant-ai-experiments/" [X Link](https://x.com/anyuser/status/1640884040835428357) 2023-03-29T01:10Z 208.9K followers, 7.1M engagements "👉Seismic from @geoffreyhinton: The best hope is for the worlds leading scientists to collaborate on ways of controlling the technology. I dont think they should scale this up more until they have understood whether they can control it https://www.nytimes.com/2023/05/01/technology/ai-google-chatbot-engineer-quits-hinton.htm https://www.nytimes.com/2023/05/01/technology/ai-google-chatbot-engineer-quits-hinton.htm" [X Link](https://x.com/anyuser/status/1653017443726692352) 2023-05-01T12:44Z 208.9K followers, 851.3K engagements "remember how I said the quality of the internet was going to go down https://futurism.com/top-google-result-johannes-vermeer-ai-generated-knockoff https://futurism.com/top-google-result-johannes-vermeer-ai-generated-knockoff" [X Link](https://x.com/anyuser/status/1665875893062074368) 2023-06-06T00:18Z 208.9K followers, 601.6K engagements "@EdKrassen NBC interview was pre Ukraine invasion. I dont think it is appropriate to give him a platform now" [X Link](https://x.com/anyuser/status/1695975659095351654) 2023-08-28T01:44Z 208.9K followers, 75.2K engagements "Common view; almost certainly wrong. The boards decision a . Instead it was almost certainly a Hail Mary: 👉This Boards sole job was to look out for humanity - NOT to protect the brand. 👉They must have seen danger in something Sam was doing 👉They must have known that everyone in the world would be watching and that their decision would not be popular. 👉They proceeded anyway despite likely personal costs and despite knowing they might not succeed because they felt there was no other option. Even now their sole goal seems to be make sure that their successors protect the mission. Respect. 💪" [X Link](https://x.com/anyuser/status/1726392480759972285) 2023-11-20T00:10Z 208.9K followers, 2.6M engagements "The number of people simply unable to imagine that the Board acted in a good faith with a legitimate and serious concern boggles my mind. We dont of course know what happened but simply dismissing that possibility seems foolish to me especially given that they were not operating out of financial self interest unlike all the other actors here" [X Link](https://x.com/anyuser/status/1726422263832965189) 2023-11-20T02:08Z 208.9K followers, 221K engagements "Appalled that @timnitgebru AI ethics icon cant lay off a lovely 87-year-old man one of the great scientists of our time whose journalist son was kidnapped and publicly beheaded in an act of political violence. Her power has gone to her head; her humanity has vanished" [X Link](https://x.com/anyuser/status/1731136044563882217) 2023-12-03T02:19Z 208.9K followers, 788.1K engagements "Remember how Sam Altman told the US Senate he had no direct investment in OpenAI and how they gushed over his apparent selflessness 👉He didnt mention that (presumably) he had equity in YC which likely has equity in OpenAI. And now this: 👉OpenAI Agreed to Buy $51 Million of AI Chips From a Startup Backed by CEO Sam Altman #candor" [X Link](https://x.com/anyuser/status/1731375497458987351) 2023-12-03T18:11Z 208.9K followers, 976.1K engagements "As a graduate of MIT I literally couldnt watch this to the end; it was too painful. I honestly dont understand how things got to this point. "This is the same climate of antisemitism that has led to the massacre of Jews throughout the centuries. This is not just harassment. This is our lives on the line." @MIT student Talia Khan highlights the rise of antisemitism at MIT. https://t.co/zXb03xodXb "This is the same climate of antisemitism that has led to the massacre of Jews throughout the centuries. This is not just harassment. This is our lives on the line." @MIT student Talia Khan highlights" [X Link](https://x.com/anyuser/status/1732290318706606163) 2023-12-06T06:46Z 208.9K followers, 425.6K engagements "OpenAI is in a heap of trouble and its not just text. Long thread why (1/n) based on work with @Rahll" [X Link](https://x.com/anyuser/status/1740834294803902807) 2023-12-29T20:36Z 208.9K followers, 1.2M engagements "@Rahll @latimes indeed @bcmerchant we are dropping something big tomorrow. email me for a sneak preview ;)" [X Link](https://x.com/GaryMarcus/status/1742581638935330844) 2024-01-03T16:20Z 207.5K followers, [---] engagements "Potential copyright infringement & why generative AI may be in for a rough ride Video below on how widespread the problem is and a [----] word deep dive w @rahll @ 👉 Why film & game studios artists & actors may sue 👉 Why users are at risk 👉 Why the issue could be hard to fix 👉 How scaling could be making the problem worse https://spectrum.ieee.org/midjourney-copyright https://spectrum.ieee.org/midjourney-copyright https://spectrum.ieee.org/midjourney-copyright https://spectrum.ieee.org/midjourney-copyright" [X Link](https://x.com/GaryMarcus/status/1743653987184972143) 2024-01-06T15:21Z 207.5K followers, 130.1K engagements "Rough Translation: We wont get fabulously rich if you dont let us steal so please dont make stealing a crime Dont make us pay fees either Sure Netflix might pay billions a year in licensing fees but *we* shouldnt have to More money for us moar OpenAI lobbying for copyright law revision in the UK @GaryMarcus @jason_kint https://t.co/kug5rCpINT OpenAI lobbying for copyright law revision in the UK @GaryMarcus @jason_kint https://t.co/kug5rCpINT" [X Link](https://x.com/anyuser/status/1744362345403392510) 2024-01-08T14:16Z 208.9K followers, 910.8K engagements "Now we know why Sam Altman went around the world last summer meeting world leaders: his company wont make it big unless they can convince governments to give them one of the biggest handouts in history. OpenAI lobbying: (a) ChatGPT useless without copyright material (b) Requires special exemptions from the law. https://t.co/yW6DRWSoBF So LLMs are thinly disguised plagiarization algorithms and pilfering copyright is the only way to make them profitable right @GaryMarcus https://t.co/hILAkYpprO OpenAI lobbying: (a) ChatGPT useless without copyright material (b) Requires special exemptions from" [X Link](https://x.com/anyuser/status/1744400559589978174) 2024-01-08T16:47Z 208.9K followers, 94.5K engagements "WOW Its about time somebody looked into this Potentially enormous. Breaking @theinformation" [X Link](https://x.com/anyuser/status/1745127779237310888) 2024-01-10T16:57Z 208.9K followers, 148.9K engagements "Black Mirror has arrived ahead of schedule. An entire cast of deepfaked people tricked a CFO out of $25 million. (In the) multi-person video conference it turns out that everyone was fake Deepfaked shit is getting real" [X Link](https://x.com/anyuser/status/1754188331188130201) 2024-02-04T17:01Z 208.9K followers, 1.2M engagements "The fun never ends" [X Link](https://x.com/anyuser/status/1755476468157833593) 2024-02-08T06:19Z 208.9K followers, 203.6K engagements "When I started working on AI four decades ago it simply didnt occur to me that one of the biggest use cases would be derivative mimicry transferring value from artists and other creators to megacorporations using massive amounts of energy. This is not the AI I dreamed of. OpenAI just released a model that can generate 1-minute videos. You simply cannot argue that these models don't / won't compete with the content they're trained on and the human creators behind that content. What is the model trained on Did the training data providers consent OpenAI just released a model that can generate" [X Link](https://x.com/anyuser/status/1758202932569473034) 2024-02-15T18:53Z 208.9K followers, 840.6K engagements "@SemanticEntity People pay for samples and consider stealing to be stealing" [X Link](https://x.com/anyuser/status/1758211293901963769) 2024-02-15T19:26Z 208.9K followers, 32.9K engagements "Amazing But um ants have six legs. We are about to have a whole generation of children educated by fake videos that are completely plausible to naive audiences yet biologically incorrect. 🤯 "pov footage of an ant navigating the inside of an ant nest" Video generated by Sora https://t.co/N3Ik9QBinU "pov footage of an ant navigating the inside of an ant nest" Video generated by Sora https://t.co/N3Ik9QBinU" [X Link](https://x.com/anyuser/status/1759073075281469504) 2024-02-18T04:31Z 208.9K followers, 299.7K engagements "Musicians and Artists does this sound familiar From one musician I know:" [X Link](https://x.com/anyuser/status/1760006385465733376) 2024-02-20T18:19Z 208.9K followers, 200.9K engagements "A lot of people are missing the funny part. Let me break it down. If OpenAI has actually achieved AGI they get their software back from Microsoft. Elons lawsuit has put them in a position having to prove that they *havent* reached AGI even though OpenAI likes to hint that they have or are close. So they need an expert witness to testify e.g. that their partner Microsoft was full of it when they said that GPT-4 represented Sparks of AGI. Which means hiring someone like *me* to save them. 🤣🤣🤣" [X Link](https://x.com/anyuser/status/1764005820008890713) 2024-03-02T19:12Z 208.9K followers, 306.3K engagements "Dear @openAI This is baloney: We are dedicated to the OpenAI mission and have pursued it every step of the way. The original mission was to be unconstrained by a need to generate a financial return not organized for private gain seeking to open source technology where applicable. Now Microsoft has an exclusive 49% of your profits go them and rather than helping all humanity you are stealing from artists writers etc to increase your profits. For good measure you have almost entirely abandoned open-source and are among the least open AI companies anywhere. Whatever your history with @elonmusk" [X Link](https://x.com/anyuser/status/1765210884597764207) 2024-03-06T03:00Z 208.9K followers, 202.2K engagements "If all we had was ChatGPT we could say hmm maybe hallucinations are just a bug and fantasize that they werent hard to fix. If all we had was Gemini we could say hmm maybe hallucinations are just a bug. If all we had was Mistral we could say hmm maybe hallucinations are just a bug. If all we had was LLAMA we could say hmm maybe hallucinations are just a bug. If all we had was Grok we could say hmm maybe hallucinations are just a bug. Instead we need to wake up and realize that hallucinations are absolutely core to how LLMs work and that we need new approaches based on different ideas" [X Link](https://x.com/anyuser/status/1781769926279409857) 2024-04-20T19:40Z 208.9K followers, 538K engagements "So many people are confused about the relation between human cognitive errors and LLM hallucinations that I wrote this short explainer: Humans say things that aren't true for many different reasons Sometimes they lie Sometimes they misremember things Sometimes they fail to think through what they are saying Sometimes they are on drugs Sometimes they suffer from mental disorders etc LLMs errors result from . They don't have (e.g.) intentions egos or financial interests so they don't lie. They don't take drugs. They don't have emotional states. Instead LLM "hallucinations" arise regularly" [X Link](https://x.com/anyuser/status/1782092855634526505) 2024-04-21T17:03Z 208.9K followers, 211.5K engagements "GPT-4o hot take: The speech synthesis is terrific reminds me of Google Duplex (which never took off). but If OpenAI had GPT-5 they have would shown it. They dont have GPT-5 after [--] months of trying. The most important figure in the blogpost is attached below. And the most important thing about the figure is that 4o is not a lot different from Turbo which is not hugely different from [--]. Lots of quirky errors are already being reported same as ever. (See e.g. examples from @RosenzweigJane and @benjaminjriley.) OpenAI has presumably pivoted to new features precisely because they dont know how" [X Link](https://x.com/anyuser/status/1790122337058119725) 2024-05-13T20:49Z 208.9K followers, 1.2M engagements "Someday my children will ask me why anyone ever took seriously the thought that systems like this had anything to do with AGI. I honestly have no idea how I will answer. Mass delusion (Example emailed to me by a friend.)" [X Link](https://x.com/anyuser/status/1790521732681609353) 2024-05-14T23:17Z 208.9K followers, 310.2K engagements "The OpenAI board was right. Even on little things Sam is not consistently candid. Earlier today OpenAI said the resemblance to ScarJo was a coincidence. It was not: Statement from Scarlett Johansson on the OpenAI situation. Wow: https://t.co/8ibMeLfqP8 Statement from Scarlett Johansson on the OpenAI situation. Wow: https://t.co/8ibMeLfqP8" [X Link](https://x.com/anyuser/status/1792683878051975652) 2024-05-20T22:28Z 208.9K followers, 613.3K engagements "I think @elonmusk owes Apple an apology. I wont say this is perfect (hackers will inevitably find problems) but it seems to me to be quite thoughtfully conceived and likely is state of the art. This is from Apple's State of the Union The local model is a 3B parameter SLM that uses adapters trained for each specific feature. Diffusion model does the same thing adapter for each style. Anything running locally or Apple's Secure Cloud is an Apple model not OpenAI. https://t.co/BRkWAQQCQI This is from Apple's State of the Union The local model is a 3B parameter SLM that uses adapters trained for" [X Link](https://x.com/anyuser/status/1800583615354728797) 2024-06-11T17:39Z 208.9K followers, 312.6K engagements "Wow evidence from OpenAI itself in favor of my conjecture of diminishing returns. Mira Murati says the AI models that OpenAI have in their labs are not much more advanced than those which are publicly available https://t.co/OtgXhZg13P Mira Murati says the AI models that OpenAI have in their labs are not much more advanced than those which are publicly available https://t.co/OtgXhZg13P" [X Link](https://x.com/anyuser/status/1801049792455245986) 2024-06-13T00:31Z 208.9K followers, 297K engagements "I dont think people quite realize the significance of this moment so I will spell it out: OpenAI has been bluffing all along and their CTO just gave away the fact that they are not holding aces. 75% chance that the GenAI bubble will have burst by [--] months from today. Wow evidence from OpenAI itself in favor of my conjecture of diminishing returns. Wow evidence from OpenAI itself in favor of my conjecture of diminishing returns" [X Link](https://x.com/anyuser/status/1801096117427302467) 2024-06-13T03:35Z 208.9K followers, 250.6K engagements "Have to say I am with LeCun on this one" [X Link](https://x.com/anyuser/status/1815147690185171320) 2024-07-21T22:11Z 208.9K followers, 80.3K engagements "Companies that have tried hard to create GPT-5 level models and thus far failed: OpenAI Google Microsoft Anthropic Meta http://X.AI http://X.AI" [X Link](https://x.com/GaryMarcus/status/1815865145320755527) 2024-07-23T21:42Z 207.5K followers, 10.7K engagements "SCOOP: OpenAI may lose $5B this year & may run out of cash in [--] months unless they raise more $ per analysis @theinformation. Investors should ask: What is their moat Unique tech What is their route in profitability when Meta is giving away similar tech for free Do they have a killer app Will the tech ever be reliable What is real and what is just demo" [X Link](https://x.com/anyuser/status/1816116071226868085) 2024-07-24T14:19Z 208.9K followers, 968.8K engagements "Incredibly damning new study on GenAI from Upwork via @baldurbjarnason: Nearly half (47%) of workers using AI say they have no idea how to achieve the productivity gains their employers expect. Over three in four (77%) say AI tools have decreased their productivity and added to their workload in at least one way. Says Bjarnason Its quite unusual for a study like this on a new office tool roughly two years after that toolChatGPTexploded into peoples workplaces to return such a resoundingly negative sentiment. Many on X may love ChatGPT for work but much of the outside world feels quite" [X Link](https://x.com/anyuser/status/1817928994844753973) 2024-07-29T14:23Z 208.9K followers, 321K engagements "Never thought I would do this but I have unfollowed Elon. Reading his relentless distortion is ruining my experience here" [X Link](https://x.com/anyuser/status/1818300194481750104) 2024-07-30T14:58Z 208.9K followers, 105K engagements "OMG The GenAI honeymoon is definitely over. A pharma company stopped using Microsoft's Copilot AI tool with an exec citing high cost and low value. The company used Office [---] Copilot for [---] staff and compared it to middle-school presentations. -@BusinessInsider" [X Link](https://x.com/anyuser/status/1818679777219559689) 2024-07-31T16:07Z 208.9K followers, 291.2K engagements "I just wrote a great piece for WIRED predicting that the AI bubble will in collapse in [----] and now I wish I hadnt. Clearly I got the year wrong. Its going to be days or weeks from now not months" [X Link](https://x.com/anyuser/status/1819525054537126075) 2024-08-03T00:05Z 208.9K followers, 1.6M engagements "⚠ Something terrifying just snapped in place: Sam Altman has acknowledged wanting to train on everyone's personal documents (Word files email etc) ChatGPT has gathered unprecedented amounts of personal data Sam founded WorldCoin known for their eye-scanning orb OpenAI recently put Nakasone (ex NSA) on the board OpenAI just put in money in a $60M fundraise with a Webcam company and is planning hardware joint venture with them Altman wants to know - and monetize - everything about you. He might succeed" [X Link](https://x.com/anyuser/status/1821194497218506910) 2024-08-07T14:39Z 208.9K followers, 115.8K engagements "I am learning so much from Googles AI" [X Link](https://x.com/anyuser/status/1823952848083620332) 2024-08-15T05:20Z 208.9K followers, 145.1K engagements "What more proof do you need that Elon is living in a bubble than this" [X Link](https://x.com/anyuser/status/1826046428843749414) 2024-08-20T23:59Z 208.9K followers, 153.7K engagements "Pretending that Elon Musk is centrist is the new Flat Earth movement" [X Link](https://x.com/anyuser/status/1826289419659452757) 2024-08-21T16:05Z 208.9K followers, 99.3K engagements "Fantastic essay on AI and art today in The New Yorker" [X Link](https://x.com/anyuser/status/1830093759545024763) 2024-09-01T04:02Z 208.9K followers, 58.2K engagements "Crikey AI worse than humans in every way at summarizing information the technology might actually make more work for people not less @crikey_news" [X Link](https://x.com/anyuser/status/1831421543769166270) 2024-09-04T19:58Z 208.9K followers, 119K engagements "Yann LeCun: AI wont be used to generate misinformation in science deliberately or accidentally. Its not a problem. Actual research: Its already happening" [X Link](https://x.com/anyuser/status/1834216963729715594) 2024-09-12T13:06Z 208.9K followers, 63.9K engagements "We are watching Elon meltdown in real-time. If his friends cared for him truly cared for him they would intervene" [X Link](https://x.com/anyuser/status/1835042761428681047) 2024-09-14T19:47Z 208.9K followers, 264.3K engagements "Wait what Murati is leaving Schulman left Karpathy left Ilya left Leike left Brockman is on leave perhaps a dozen others left GPT-5 hasnt dropped Sora hasnt shipped the company had an operating loss of $5b last year there is no obvious moat Meta is giving away similar software for freemany lawsuits pending and people are valuing this company at $150 billion dollars Absolutely insane. Investors shouldnt be pouring more money at higher valuations they should be asking what is going on. I shared the following note with the OpenAI team today. https://t.co/nsZ4khI06P I shared the following note" [X Link](https://x.com/anyuser/status/1839037069307555905) 2024-09-25T20:19Z 208.9K followers, 1.9M engagements "🎉HUGE news it means OpenAI and others can no longer hide behind the BS phrase publicly available. If they scrape something off the internet that is copyrighted or trademarked they are now obligated to disclose that. Newsom just signed a new bill requiring Gen AI businesses to publicly disclose their training data 👀 https://t.co/L6obdI5kzz Newsom just signed a new bill requiring Gen AI businesses to publicly disclose their training data 👀 https://t.co/L6obdI5kzz" [X Link](https://x.com/anyuser/status/1840408467770892753) 2024-09-29T15:09Z 208.9K followers, 83.8K engagements "The big takeway from tonight: Elon is spending too much time on politics and not enough on Tesla" [X Link](https://x.com/anyuser/status/1844578738392252539) 2024-10-11T03:20Z 208.9K followers, 87.4K engagements "👇Superb new article from @apple AI: we found no evidence of formal reasoning in language models . Their behavior is better explained by sophisticated pattern matchingso fragile in fact that changing names can alter results by 10% where changing a word or two in irrelevant ways can give you a different answer. Strongly encourage you to read the whole thread. 1/ Can Large Language Models (LLMs) truly reason Or are they just sophisticated pattern matchers In our latest preprint we explore this key question through a large-scale study of both open-source like Llama Phi Gemma and Mistral" [X Link](https://x.com/anyuser/status/1844754079618400271) 2024-10-11T14:57Z 208.9K followers, 498.5K engagements "I literally dont see how democracy can survive long term if the richest and most powerful people are free to lie at scale with no consequence" [X Link](https://x.com/anyuser/status/1846592170771525736) 2024-10-16T16:40Z 208.9K followers, 133K engagements ". Making a million dollar prize contingent on registering to vote is illegal. See [--] USC 10307(c) https://t.co/8hnQS6lqYV Also with PA registration soon closing for this election it may not accomplish what Musk wants. Making a million dollar prize contingent on registering to vote is illegal. See [--] USC 10307(c) https://t.co/8hnQS6lqYV Also with PA registration soon closing for this election it may not accomplish what Musk wants" [X Link](https://x.com/anyuser/status/1847814901823381939) 2024-10-20T01:39Z 208.9K followers, 178.4K engagements "Hot take on the mess Elon just made by pushing too far: Everyone of Elon Musks boards now has to consider that he has just flagrantly violated federal law and may go to jail for it. That is risk to his companies and because of their fiduciary responsibilities to his boards. If anything tanks lawsuits could be epic. Boards will be asked why they allowed this to go on for so long. . " [X Link](https://x.com/anyuser/status/1847820167507738812) 2024-10-20T02:00Z 208.9K followers, 155.9K engagements "Why Musk will sacrifice democracy and maybe even jail in one convenient chart @nytimes" [X Link](https://x.com/anyuser/status/1848158028027072711) 2024-10-21T00:23Z 208.9K followers, 843.3K engagements "When LeCun is right hes right" [X Link](https://x.com/anyuser/status/1852120136301576287) 2024-10-31T22:47Z 208.9K followers, 276.1K engagements "⚠⚠⚠ I have a PhD in Brain and cognitive science from MIT. / : - : swaying missing and and stumbling in front of the garbage truck - : inappropriate musical pause/swaying similar to longer incident a few weeks ago - : threatening Liz Cheney with violence - : description of Musk as good at computer Things are degenerating fast" [X Link](https://x.com/anyuser/status/1852355571145670831) 2024-11-01T14:22Z 208.9K followers, 1.1M engagements "I would very much welcome the thoughts of anyone specializing in neurology. Whats best diagnosis here Prognosis" [X Link](https://x.com/GaryMarcus/status/1852366084801970439) 2024-11-01T15:04Z 207.5K followers, 94.7K engagements "A bit over [--] hours later Trump showed at least [--] more signs of possible dementia today including more inappropriate foul language (no surprise) more signs of disorientation and the worlds first effort by a presidential candidate to mimic fellatio in public yet another remarkable failure to inhibit socially inappropriate responses confirming my overall impression. Whatever disease he has is likely progressing rapidly" [X Link](https://x.com/anyuser/status/1852577905257431193) 2024-11-02T05:06Z 208.9K followers, 97.3K engagements "and why he may soon wind up in jail. TLDR: Elons own lawyers just screwed up really badly. Longer explanation background first: Musk ran a lottery in PA to try to get MAGA fans to register to vote. Thats very likely a violation of federal election law TBD. But separately it also flouts PA lottery law. Philadelphia DA brought him to court for it. In response and this is the insane part Musks lawyers tried to weasel alleging in court that There is no prize (just people being hired to do ads) and that the winners are not chosen at random. - . . How do we know this Because when Elon announced" [X Link](https://x.com/anyuser/status/1853476869129281717) 2024-11-04T16:38Z 208.9K followers, 157.2K engagements "Folks game over. I won. GPT is hitting a period of diminishing returns just like I said it would. news: OpenAI's upcomning Orion model shows how GPT improvements are slowing down It's prompting OpenAI to bake in reasoning and other tweaks after the initial model training phase. https://t.co/VwD1xSbZUv news: OpenAI's upcomning Orion model shows how GPT improvements are slowing down It's prompting OpenAI to bake in reasoning and other tweaks after the initial model training phase. https://t.co/VwD1xSbZUv" [X Link](https://x.com/anyuser/status/1855382564015689959) 2024-11-09T22:50Z 208.9K followers, 361.2K engagements "An MIT professor just asked me how can you be so confident that OpenAI isnt very close to AGI Here is a thread of links to a several recent observations that I think show how far we are from robust intelligence. Importantly all of them are *known* problems many that I have noted repeatedly for years and even decades. There is no *principled* solution thus far to any of them. 🧵" [X Link](https://x.com/anyuser/status/1876127737108119781) 2025-01-06T04:44Z 208.9K followers, 362.7K engagements "The countdown until we are told that LLMs are too big to fail starts now. We cant afford to lose to China they will say accepting their multibillion dollar bailouts" [X Link](https://x.com/anyuser/status/1876361703446737279) 2025-01-06T20:14Z 208.9K followers, 430K engagements "Lets wreck the environment to build an unreliable technology that has made little revenue which China can now build much more efficiently. big. beautiful. buildings. https://t.co/8lAvUufXFm big. beautiful. buildings. https://t.co/8lAvUufXFm" [X Link](https://x.com/GaryMarcus/status/1882941379678200226) 2025-01-24T23:59Z 207.5K followers, 58.6K engagements "People here are strangely quiet about a private unelected citizen taking over the U.S. government. A bit like in this cartoon" [X Link](https://x.com/anyuser/status/1886412457541042388) 2025-02-03T13:52Z 208.9K followers, 52.7K engagements "I retired from the American University system at age [--] (after [--] years) so have no direct stake in this. But am here to say that drastically cutting NIH & NSF funding and reducing grant overhead to 15% are among the dumbest ideas I have ever seen. America will never be the same. Science will take a huge hit; academic stars will leave; university reputations will crumble and homegrown talent will be even harder to find. Asia and Europe will profit. Epic unforced errors" [X Link](https://x.com/anyuser/status/1888235546616742317) 2025-02-08T14:36Z 208.9K followers, 631.1K engagements "Suppose the air conditioning in your car stopped working and you brought the car into the shop and the mechanic said I am afraid your car is broken so I have started dismantling it. That in a nutshell is what Trump and Musk are doing to your country. I disagree. The education system is broken and needs to be dismantled I disagree. The education system is broken and needs to be dismantled" [X Link](https://x.com/anyuser/status/1888595911267500113) 2025-02-09T14:28Z 208.9K followers, 99.4K engagements "⚠⚠⚠ Everyoneand not just @theinformationshould be genuinely terrified that the richest man in the world has built a Large Language Model that spouts propaganda in his image. This isnt free speech. Its a massive massive thumb on the scale from one of the most powerful yet unelected people in the world. If that doesnt scare you you arent paying attention. Grok [--] is so based 😂 https://t.co/Rfphquqk2b Grok [--] is so based 😂 https://t.co/Rfphquqk2b" [X Link](https://x.com/anyuser/status/1891151083747393585) 2025-02-16T15:42Z 208.9K followers, 112.8K engagements "If you live in the U.S. your civil rights will soon be over. This is not a drill" [X Link](https://x.com/anyuser/status/1903177341813018663) 2025-03-21T20:10Z 208.9K followers, 523.5K engagements "God I wish @elonmusk would have taken this bet" [X Link](https://x.com/anyuser/status/1915720664436105436) 2025-04-25T10:52Z 208.9K followers, 98.3K engagements "Horseshit. Every business in the world has discovered in the last several months that GenAI is not in fact smart enough replace most of their employees. Whatever you are reading from these (often gamed sometimes contaminated) benchmarks does not reflect real-world reality. The top AI model is now smarter than 85% of humans. By the end of [----] it will be smarter than 99.9% of humans. And you think you will still have a job https://t.co/4R4ZsuYJZY The top AI model is now smarter than 85% of humans. By the end of [----] it will be smarter than 99.9% of humans. And you think you will still have a" [X Link](https://x.com/anyuser/status/1920262283600580977) 2025-05-07T23:39Z 208.9K followers, 147.1K engagements "A computer scientists perspective on vibe coding:" [X Link](https://x.com/anyuser/status/1923535405372309610) 2025-05-17T00:25Z 208.9K followers, 846.3K engagements ". I will break it down for you in four sentences: It is not just that these systems are doing more pattern recognition than reasoning. It is that LRM reasoning models . You cant have reliable AGI without the reliable execution of algorithms. And we will *never* get to alignment without them" [X Link](https://x.com/anyuser/status/1931718368702419351) 2025-06-08T14:22Z 208.9K followers, 177.8K engagements "Sam I dont take issue with OpenAIs achievements I take issue with the hype and a history of misleading statements (some detailed below). Youve built something impressive and in many ways groundbreaking. I would be singing OpenAI's praises it's amazing how you scaled Transformers - if it werent for the hype much of it fanned by you that has massively overstated what the technology can do today and in the near future going back to the Rubik's cube demonstration that misrepresented what you had done. I think this hype is harming the world. If I seem frustrated its because of the damage and the" [X Link](https://x.com/anyuser/status/1932612831188562070) 2025-06-11T01:36Z 208.9K followers, 141.8K engagements "Terence Tao reporting on the same problem we have seen endlessly in other domains: LLMs produce output that looks correct but is often deeply wrong and even stupid on careful inspection. I know of no domain in which this is NOT the case. And yet people seem surprised over and over to discover it anew in fresh domains. and non experts often dont see it at all. No solution to this problem no AGI. Scaling has not solved it. Terence Tao says today's AIs pass the eye test -- but fail miserably on the smell test. They generate proofs that look flawless. But the mistakes are subtle and strangely" [X Link](https://x.com/anyuser/status/1934131893689282573) 2025-06-15T06:12Z 208.9K followers, 250K engagements "⚠ Everybody who uses LLMs should read this thread: LLM users consistently underperformed on multiple cognitive and neural measures. Its a hefty 206-page research paper and the findings are concerning. "LLM users consistently underperformed at neural linguistic and behavioral levels" This study finds LLM dependence weakens the writers own neural and linguistic fingerprints. 🤔🤔 Relying only on EEG https://t.co/pRBQ15qst3 Its a hefty 206-page research paper and the findings are concerning. "LLM users consistently underperformed at neural linguistic and behavioral levels" This study finds LLM" [X Link](https://x.com/anyuser/status/1934796157223620986) 2025-06-17T02:12Z 208.9K followers, 127.2K engagements "BREAKING: Explosive new paper from MIT/Harvard/UChicago. Things just got worse a lot worse for LLMs and the myth that they can understand and reason. The paper documents a pattern they called Potemkins a kind of reasoning inconsistency (see figure below). They show that LLMs - even models like o3 make these errors frequently. You cant possibly create AGI based on machines that cannot keep consistent with their own assertions. You just cant. success on benchmarks only demonstrates potemkin understanding: the illusion of understanding driven by answers irreconcilable with how any human would" [X Link](https://x.com/anyuser/status/1938629881820323940) 2025-06-27T16:05Z 208.9K followers, 423.8K engagements "what almost every amateur who has been sold bullshit about vibe coding is eventually going to encounter" [X Link](https://x.com/anyuser/status/1942478390323552588) 2025-07-08T06:58Z 208.9K followers, 243.8K engagements "Feel the AGI @sama. No PhD student can match this Oh brother https://t.co/k2ZrO7ygSU Oh brother https://t.co/k2ZrO7ygSU" [X Link](https://x.com/anyuser/status/1953703268103991343) 2025-08-08T06:22Z 208.9K followers, 702.8K engagements "Sorry but this tweet did not age well. https://t.co/1u0MOGvJWU https://t.co/1u0MOGvJWU" [X Link](https://x.com/anyuser/status/1953887884341391775) 2025-08-08T18:35Z 208.9K followers, 911.7K engagements "what did Ilya see cuz I aint seeing it" [X Link](https://x.com/GaryMarcus/status/1953917603874734143) 2025-08-08T20:33Z 207.5K followers, 73.6K engagements "My work here is truly done. Nobody with intellectual integrity can still believe that pure scaling will get us to AGI. GPT-5 may be a moderate quantitative improvement (and it may be cheaper) but it still fails in all the same qualitative ways as its predecessors on chess on reasoning in vision; even sometimes on counting and basic math. Hallucinations linger. Dozens of shots on goal (Grok Claude Gemini) etc have invariably faced the same problems. Thats exactly what it means to hit a wall and exactly the set of obstacles I described in [----]. No hypothesis has ever been given more benefit of" [X Link](https://x.com/anyuser/status/1953939152594252170) 2025-08-08T21:59Z 208.9K followers, 233K engagements "Mark my words this week marks peak bubble. Oracle just went up $300B on what appears to be a non-binding deal with OpenAI. OpenAI doesnt have $300 billion dollars They dont have anywhere near $300 billion dollars By their own (presumably optimistic) projection they wont turn a profit until [----]. And all this from a company thought (or claimed) that GPT-5 was going to be tantamount to AGI (spoiler alert: it wasnt) For good measure Oracle doesnt have the chips they would need to fulfill the contracts or even the cash to buy them. I will be astounded if Oracle meets their projections. Further" [X Link](https://x.com/anyuser/status/1966163598839001589) 2025-09-11T15:35Z 208.9K followers, 833.7K engagements "An $800 billon dollar shortfall based almost entirely on the ever-more questionable gamble that scaling will lead to AGI. A staggering $800 billion global shortfall in AI infrastructure spend is foreseen by [----] according to a new report by Bain and that could hamper AI's progress @business (Gift link: free to read until Sept 30) https://t.co/HvwybrC1aD #AI https://t.co/PN8LZ7OfpV A staggering $800 billion global shortfall in AI infrastructure spend is foreseen by [----] according to a new report by Bain and that could hamper AI's progress @business (Gift link: free to read until Sept 30)" [X Link](https://x.com/GaryMarcus/status/1970555584484151716) 2025-09-23T18:27Z 207.5K followers, 15.7K engagements "BREAKING NEWS: New @betterup/Stanford study on absolutely nails part of why the ROI on GenAI has been so poor: Employees are using AI tools to create low-effort passable looking work that ends up creating more work for their coworkers. On social media which is increasingly clogged with low-quality AI-generated posts this content is often referred to as AI slop. In the context of work we refer to this phenomenon as workslop. AI generated work content that masquerades as good work but lacks the substance to meaningfully advance a given task" [X Link](https://x.com/anyuser/status/1970933316019982395) 2025-09-24T19:28Z 208.9K followers, 116.2K engagements "AI is the asbestos we are shoveling into the walls of our society and our descendants will be digging it out for generations - @doctorow speaking truth and taking no prisoners" [X Link](https://x.com/anyuser/status/1974463227200426158) 2025-10-04T13:14Z 208.9K followers, 83.8K engagements "The last few months have been devastating for LLM dreams: Apple reasoning paper and the ASU mirage paper and many others confirmed that LLMs still cant solve distribution shift. GPT-5 came late and fell short. Turing Award winner Rich Sutton thanked me for my critiques of LLMs and agreed. Karpathy just said agents arent anywhere close and that AGI is a decade away. And Hassabis just blew up some wildly overhyped claims from OpenAI about math. Game over man. LLMs have their place but anyone expecting the current paradigm to be close to AGI is delusional. At first I thought GPT-5 had cracked" [X Link](https://x.com/anyuser/status/1979507386780909906) 2025-10-18T11:18Z 208.9K followers, 363.2K engagements "Absolutely the most important satire of the LLM era. My 3-month-old son is now TWICE as big as when he was born. He's on track to weigh [---] trillion pounds by age [--] https://t.co/S4v9gXLlrt My 3-month-old son is now TWICE as big as when he was born. He's on track to weigh [---] trillion pounds by age [--] https://t.co/S4v9gXLlrt" [X Link](https://x.com/anyuser/status/1980964511910887734) 2025-10-22T11:48Z 208.9K followers, 4M engagements "losing nearly $4 billion dollars a *month*. Microsofts latest SEC filing quietly exposed that OpenAI lost around $11.5 billion last quarter based on Microsofts 27% ownership stake and a $3.1 billion hit to its own net income. The filings confirm that Microsoft has funded $11.6B of its $13B commitment to OpenAI with https://t.co/lMl3wptM8H Microsofts latest SEC filing quietly exposed that OpenAI lost around $11.5 billion last quarter based on Microsofts 27% ownership stake and a $3.1 billion hit to its own net income. The filings confirm that Microsoft has funded $11.6B of its $13B commitment" [X Link](https://x.com/anyuser/status/1984087854004498601) 2025-10-31T02:39Z 208.9K followers, 146.9K engagements "1. OpenAI is a train wreck. [--]. They cant pay the obligations they are taking on and they know it. [--]. Guess what You the taxpayer will be left holding the bag" [X Link](https://x.com/anyuser/status/1986205437012943094) 2025-11-05T22:54Z 208.9K followers, 65.4K engagements "Call your Congress people. Ten months almost to the day OpenAI and NVidia have both rolled out *exactly* the too big to fail . but China game plan I predicted they would be forced to. Our government is welching on food stamps but they are going to bail out the biggest public and private companies of all time with your money. Dont stand for it. The countdown until we are told that LLMs are too big to fail starts now. We cant afford to lose to China they will say accepting their multibillion dollar bailouts. The countdown until we are told that LLMs are too big to fail starts now. We cant" [X Link](https://x.com/anyuser/status/1986223782298853438) 2025-11-06T00:07Z 208.9K followers, 333.3K engagements "Could this be the week We all know the bubble is going to pop; the only question is when" [X Link](https://x.com/anyuser/status/1986462133052404124) 2025-11-06T15:54Z 208.9K followers, 303.8K engagements "if only @sama had a reputation for keeping to his word I would be comforted by this. but he doesnt. and I am not comforted by it not one bit. the reality is that yesterdays statement on government loan guarantees was a trial balloon that a got a ton of flak and this is paragraph after paragraph of spin doctoring to try to walk that back. I would like to clarify a few things. First the obvious one: we do not have or want government guarantees for OpenAI datacenters. We believe that governments should not pick winners or losers and that taxpayers should not bail out companies that make bad" [X Link](https://x.com/anyuser/status/1986534153375129768) 2025-11-06T20:40Z 208.9K followers, 119.7K engagements "Sam is getting roasted alive. Nobody is buying his bullshit anymore. Glorious" [X Link](https://x.com/anyuser/status/1986555284995908055) 2025-11-06T22:04Z 208.9K followers, 34.5K engagements "AI backlash reaches new heights" [X Link](https://x.com/anyuser/status/1987914291362849244) 2025-11-10T16:04Z 208.9K followers, 51K engagements "👉Softbank sells entire Nvidia position. 👉Oracle debt downgraded. 👉Meta financing games revealed. 👉OpenAI CEO @sama couldnt explain how company would meet its $1.4 T obligations. 👉Coreweave drops 20% in a week. You do the math" [X Link](https://x.com/anyuser/status/1988315110553907415) 2025-11-11T18:37Z 208.9K followers, 204.6K engagements "@vaughanbell I don't think you've characterized my views correctly; I posted a comment on your blog" [X Link](https://x.com/GaryMarcus/status/493114943470698496) 2014-07-26T19:25Z 207.5K followers, [--] engagements "@TomChivers @me_datapoint its not a logical argument to say that is because this system follows tricks we must be following similar tricks. my five year old knows that you cant stick a knight on a square occupied by a bishop; the slate star chess bot doesnt. the bot is cute but it teaches us little" [X Link](https://x.com/GaryMarcus/status/1214938770706817024) 2020-01-08T15:55Z 190.8K followers, [--] engagements "@_RobToews thanks (a link to a historical perspective from my gpt-2 piece might be nice since so little has really changed.)" [X Link](https://x.com/GaryMarcus/status/1285017301184475136) 2020-07-20T01:02Z 190.9K followers, [--] engagements "no much fishier @OpenAIs parent organization (by charter allegedly providing oversight) is *still as of 2020* filing as a nonprofit. they are insulating their now-closed business from open investigation whilst still shielding millions from taxes. 🤔 https://projects.propublica.org/nonprofits/organizations/810861541 Related: @OpenAI is no longer non-profit and the source of #GPT3 is closed. https://projects.propublica.org/nonprofits/organizations/810861541 Related: @OpenAI is no longer non-profit and the source of #GPT3 is closed" [X Link](https://x.com/GaryMarcus/status/1299144214098264066) 2020-08-28T00:38Z 162.9K followers, [--] engagements "@_RobToews its just a bunch of anecdotal examples at the moment; theres no statistical analysis and thus far no paper. it is certainly not reliable" [X Link](https://x.com/GaryMarcus/status/1352094526274342916) 2021-01-21T03:23Z 191K followers, [--] engagements "Oh please "The findings hint at a future in which artificial intelligence can help humans to formulate and make sense of complex arguments." This is absurd. Neither the language understanding nor the reasoning skills in current AI are remotely close. https://www.nature.com/articles/d41586-021-00539-5 https://www.nature.com/articles/d41586-021-00539-5" [X Link](https://x.com/GaryMarcus/status/1372956486456942592) 2021-03-19T17:01Z 191.1K followers, [---] engagements "Let us invent then a new breed of AI systems that mix an awareness of the past with values that represent the future that we aspire to. Our focus should be on figuring on how to build AI that can represent and reason about *values* rather than simply perpetuating past data" [X Link](https://x.com/GaryMarcus/status/1384173525368393736) 2021-04-19T15:54Z 162.5K followers, [----] engagements "@Raza_Habib496 show one place that anybody said that out loud" [X Link](https://x.com/GaryMarcus/status/1501915210038145030) 2022-03-10T13:37Z 194.6K followers, [--] engagements "You cant go to a physics conference and say: Ive got a great theory. It accounts for everything and is so simple it can be captured in two words: Anything goes." - Noam Chomsky in email re GPT-3" [X Link](https://x.com/GaryMarcus/status/1525953752418557952) 2022-05-15T21:38Z 192.6K followers, [---] engagements "@ylecun @bleepbeepbzzz @TonyZador @raphaelmilliere @davidchalmers42 @MetaAI @noema @Jake_Browning00 @NoemaMag @luislamb @AvilaGarcez hmm. whats an emergent symbol is it something that stands for something else (as classical symbols are) how would you know if you had one in what sense would it be a symbol could you do eg deductive inference with it" [X Link](https://x.com/GaryMarcus/status/1574551127538139136) 2022-09-27T00:06Z 191.1K followers, [--] engagements "@CriticalAI @EnglishOER @markcmarino you might be referring to something someone sent to me about Noam Chomsky (alive and well as of an email he sent me a couple days ago) allegedly dying in [----] generated by ChatGPTs sibling davinci complete with fake URLs. At least it got Hillary Clintons status correct 🙄" [X Link](https://x.com/GaryMarcus/status/1619732971707338752) 2023-01-29T16:23Z 162.9K followers, [---] engagements "dear hivemind - is there a term for a bunch of fake websites that mutually reinforce each other in order to increase their search ranking - are there any studies or articles about this @noUpside @ProfNoahGian @DG_Rand @sinanaral @erikbryn @chafkin @lorakolodny" [X Link](https://x.com/GaryMarcus/status/1624785466909425664) 2023-02-12T15:00Z 190.9K followers, 17.9K engagements "@elonmusk @geoffreyhinton 🙏 If you really want to fight this we should talk" [X Link](https://x.com/GaryMarcus/status/1640494672723230722) 2023-03-27T23:23Z 187.3K followers, 20.2K engagements "Opportunity Lost. The letter was flawed; so was the communitys reaction. Why there is more to life than The Letter what we should learn from the experience and what we need to do to carry on. https://garymarcus.substack.com/p/i-am-not-afraid-of-robots-i-am-afraidutm_source=twitter&sd=pf https://garymarcus.substack.com/p/i-am-not-afraid-of-robots-i-am-afraidutm_source=twitter&sd=pf" [X Link](https://x.com/GaryMarcus/status/1642581773468176384) 2023-04-02T17:36Z 173.2K followers, 66.2K engagements "Could a computer program watch any arbitrary TV program or Youtube video and reliably answer questions about its content I have been asking this question since [----]. So far the answer is still no. From [----] What Comes After the #TuringTest New test for 21st century: ".build a computer program that can watch any arbitrary TV program or YouTube video and answer questions about its content." @GaryMarcus https://t.co/dTNH0gh5yH #ArtificialIntelligence From [----] What Comes After the #TuringTest New test for 21st century: ".build a computer program that can watch any arbitrary TV program or YouTube" [X Link](https://x.com/GaryMarcus/status/1654104746008141824) 2023-05-04T12:44Z 191.1K followers, 27.8K engagements "Could I get some actual neuroscientists to weigh in on current thinking around the triune (Lizard) brain theory Am I correct in saying it has been universally rejected Whats the reasoning @WiringTheBrain @TonyZador @anne_churchland @Nancy_Kanwisher @PessoaBrain @neurograce Lizard brain Always good to see Big Techs most illustrious AI researchers riffing on 1960s neuroscience. 🙄 Lizard brain Always good to see Big Techs most illustrious AI researchers riffing on 1960s neuroscience. 🙄" [X Link](https://x.com/GaryMarcus/status/1654146596559257600) 2023-05-04T15:30Z 172.3K followers, 127.7K engagements "best use of deep voice fakes I have seen A.I. David Attenborough on #memecoin trading. https://t.co/okiEemDjDJ A.I. David Attenborough on #memecoin trading. https://t.co/okiEemDjDJ" [X Link](https://x.com/GaryMarcus/status/1660971919783391232) 2023-05-23T11:32Z 134.8K followers, 18.7K engagements "Remarkable consensus @UNTechEnvoy #GlobalDigitalCompact deep dive on #AI: Urgency of the situation Need to bridge the global divide Need for transparency accountablility fairness & protection of privacy Need for global coordination in governance of AI" [X Link](https://x.com/GaryMarcus/status/1661809929735340035) 2023-06-29T20:56Z 134.8K followers, 26.1K engagements "Good reminder that vintage [----] neural networks arent very neural. Fly brain has 3000+ distinct neuron types all but entirely abstracted away from in current neural networks. Fair bit of evidence for innate constraints too contra the AI/Ml zeitgeist" [X Link](https://x.com/GaryMarcus/status/1675575037917818880) 2023-07-03T01:34Z 134.8K followers, [---] engagements "One in five people surveyed think we will solve machine intelligence before we can figure out how a worm brain works. I doubt we will solve either this decade; both problems are harder than they initially appeared" [X Link](https://x.com/GaryMarcus/status/1676263567795044353) 2023-07-10T21:55Z 134.9K followers, 16.7K engagements "Clever and interesting paper confirming popular 1960s hypothesis of P. Simon and A. Garfunkel" [X Link](https://x.com/GaryMarcus/status/1678654185561096193) 2023-07-12T21:34Z 134.9K followers, 21.7K engagements "overwrought prose alert Proposals for stringent AI model licensing and surveillance will likely be ineffective concentrating power potentially rolling back the societal gains of the Enlightenment I guess drivers licenses and seatbelt laws rewound 1/2 an Enlightenment" [X Link](https://x.com/GaryMarcus/status/1678864088192122881) 2023-07-11T20:44Z 134.8K followers, 10.8K engagements "Welcome to Dystopia Womans photo of son rejected from Sydney competition after judges ruled it could be AI https://t.co/Fc1OzYtC3z Womans photo of son rejected from Sydney competition after judges ruled it could be AI https://t.co/Fc1OzYtC3z" [X Link](https://x.com/GaryMarcus/status/1679003369740926976) 2023-07-12T05:42Z 162.9K followers, 13K engagements "@bpositive718 @ChristianGlawe As I have said multiple times" [X Link](https://x.com/GaryMarcus/status/1679046541053509632) 2023-07-12T08:41Z 134.8K followers, [--] engagements "View from dinner last night near Aspen" [X Link](https://x.com/GaryMarcus/status/1683482123620032513) 2023-07-24T22:42Z 134.9K followers, [----] engagements "Thread on #lk99 worth watching" [X Link](https://x.com/GaryMarcus/status/1686466063305678850) 2023-08-01T19:58Z 134.9K followers, 12.3K engagements "Not looking great Superconductivity research scholars in Korea have officially come together under a "replication committee" to investigate the claims of LK-99. Here is their initial assessment: https://t.co/M0pADFijsy Superconductivity research scholars in Korea have officially come together under a "replication committee" to investigate the claims of LK-99. Here is their initial assessment: https://t.co/M0pADFijsy" [X Link](https://x.com/GaryMarcus/status/1686795185160880128) 2023-08-02T17:44Z 163K followers, 19.1K engagements "Why do we need regulation for AI Because quite simply current models cant be trusted to be safe. Test of [--] popular models each one turns to be problematic. Voluntary agreements without real teeth are not enough" [X Link](https://x.com/GaryMarcus/status/1686937425233510400) 2023-08-04T03:11Z 134.9K followers, 16.6K engagements "I fully share @samas concerns" [X Link](https://x.com/GaryMarcus/status/1687251235655897089) 2023-08-04T12:23Z 134.8K followers, 16.6K engagements "#LK99 humor" [X Link](https://x.com/GaryMarcus/status/1688186188258213888) 2023-08-06T13:52Z 129.7K followers, 15K engagements "LLM s really are a long way off from being reliable we will soon rely on them more than the best experts Cant be both" [X Link](https://x.com/GaryMarcus/status/1688208613473697792) 2023-08-06T15:21Z 134.8K followers, 29.1K engagements "First major AI-powered mess" [X Link](https://x.com/GaryMarcus/status/1688605017396457472) 2023-08-07T17:36Z 131K followers, 37.2K engagements "So I guess we shouldnt regulate cars airplanes food etc because no issue has ever lead to a species-level extinction event then Close down the FDA and FAA etc cause no foom" [X Link](https://x.com/GaryMarcus/status/1688626868835201028) 2023-08-07T19:03Z 134.8K followers, 21.5K engagements "@_ahmedzaidi Thats the ground I have been arguing for for years" [X Link](https://x.com/GaryMarcus/status/1688703595489714176) 2023-08-08T00:08Z 134.8K followers, [---] engagements "At this point anyone who has played around with GPT-4 realizes that its reasoning is unstable & highly context-dependent; very different from the arithmetic of a calculator. Yet people who note these instabilities are routinely insulted & likened to climate change deniers" [X Link](https://x.com/GaryMarcus/status/1688992061490597888) 2023-08-08T19:14Z 134.8K followers, 45.8K engagements "because it is having an outsized impact on the world and on AI research and because i think we can and should do much better" [X Link](https://x.com/GaryMarcus/status/1689295752072486912) 2023-08-09T15:21Z 134.9K followers, 10K engagements "This clip with @janleike @openAi is great. So much I agree about - the many paths difficulty in integrating them importance of exerting causal influence on the p but why work for OpenAI if you thought p(doom) was in double digits and they were accelerating whatever risks there may be" [X Link](https://x.com/GaryMarcus/status/1689324247095898112) 2023-08-09T17:14Z 121.9K followers, 19.1K engagements "The UN is putting together a high level committee on AI and looking for nominations. Do the world a solid and help them out" [X Link](https://x.com/GaryMarcus/status/1689350011027632128) 2023-08-09T18:56Z 134.9K followers, 39.1K engagements "@ arXiv: bridging the gap between LLMs and classical AI with Doug Lenat" [X Link](https://x.com/GaryMarcus/status/1689434158806777856) 2023-08-10T00:31Z 134.8K followers, 27.9K engagements "A tweet below allegedly defending GPTs honor is a perfect encapsulation of what is wrong with the culture of AI today. Long post on a ubiquitous problem. 👉 [--] examples in a stochastic system with a massive undisclosed database proves nothing. 👉 Evidence of success is triumphantly reported; errors are neither looked for nor discussed (but we all know they exist). 👉 The fact that RLHF and possibly other undisclosed components of GPT-4 appear to be regularly updated is not discussed. 👉 The result depends on custom prompt chosen by a human that happens to be relevant for this task but which" [X Link](https://x.com/GaryMarcus/status/1689635600390078464) 2023-08-10T13:51Z 134.8K followers, 261.5K engagements "@robertskmiles we also discussed this sensitivity in Rebooting AI iirc and its another symptom of shallow comprehension rather than deep understanding" [X Link](https://x.com/GaryMarcus/status/1689651374240550912) 2023-08-10T14:54Z 130.9K followers, [---] engagements "@josecyc The last time I did this I am pretty sure OpenAI just trained on the test set" [X Link](https://x.com/GaryMarcus/status/1689680358563012608) 2023-08-10T16:49Z 134.9K followers, [----] engagements "You know what is funny about this Gallup poll Everyone treats @KamalaHarris as uniquely unpopular and yet statistically she is more or less in the middle of this pack" [X Link](https://x.com/GaryMarcus/status/1689717891871158272) 2023-08-10T19:18Z 160.3K followers, [----] engagements "@pwlot @jeremyphoward @JaffrayW Literally [--] minutes ago from a respected author visiting openAI:" [X Link](https://x.com/GaryMarcus/status/1689817235035185152) 2023-08-11T01:53Z 134.8K followers, [----] engagements "Except on X" [X Link](https://x.com/GaryMarcus/status/1689817550836895744) 2023-08-11T01:54Z 131.2K followers, [----] engagements "Is there a name for this Our examination revealed that 52% of ChatGPT's answers contain inaccuracies and 77% are verbose. Nevertheless users still prefer ChatGPT's responses 39% of the time due to their comprehensiveness and articulate language style. https://arxiv.org/abs/2308.02312 https://arxiv.org/abs/2308.02312" [X Link](https://x.com/GaryMarcus/status/1689851180657397760) 2023-08-11T04:08Z 145.2K followers, 111.2K engagements "No. The urgency for regulation comes from unreliability and uncontrollability of current AI combined with widespread adoption including by bad actors not from fantasies (that I never endorsed) that current AI was somehow near to AGI" [X Link](https://x.com/GaryMarcus/status/1690152047234949120) 2023-08-12T00:03Z 134.8K followers, 17K engagements "No doubt generative AI is great for coders but will other people maintain their monthly subscriptions after the novelty wears off Or are we building our entire approach to the future around an overhyped fad" [X Link](https://x.com/GaryMarcus/status/1690523040646336513) 2023-08-13T00:37Z 122.4K followers, 42.2K engagements "What if generative AI turned out to be a dud At my (free) newsletter that I cant properly link here I discuss some economic and geopolitical implications of this that Xs algorithm doesnt seem to want you to read about" [X Link](https://x.com/GaryMarcus/status/1690554588871741440) 2023-08-13T02:43Z 134.8K followers, 57.3K engagements Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing
@GaryMarcus Gary Marcus
Gary Marcus posts on X about open ai, elon musk, agi, [----] the most. They currently have [-------] followers and [----] posts still getting attention that total [-------] engagements in the last [--] hours.
Engagements: [-------] #
- [--] Week [-------] +7,287%
- [--] Month [-------] +294%
- [--] Months [----------] -58%
- [--] Year [-----------] +12%
Mentions: [--] #
- [--] Week [--] +68%
- [--] Month [---] +447%
- [--] Months [---] -29%
- [--] Year [-----] -32%
Followers: [-------] #
- [--] Week [-------] +0.55%
- [--] Month [-------] +1%
- [--] Months [-------] +10%
- [--] Year [-------] +23%
CreatorRank: [------] #
Social Influence
Social category influence technology brands celebrities finance stocks countries us election automotive brands social networks travel destinations vc firms
Social topic influence open ai #607, elon musk, agi #369, 6969, gary, ai #2724, money, microsoft, sama #1036, $googl #1712
Top accounts mentioned or mentioned by @elonmusk @sama @ylecun @openai @rahll @theinformation @intuitmachine @iruletheworldmo @nytimes @kettlebelldan @grok @fchollet @ofquarks28701 @geoffreyhinton @emollick @wsj @littmath @pmarca @ieeespectrum @karaswisher
Top assets mentioned GARY (GARY) Microsoft Corp. (MSFT) Alphabet Inc Class A (GOOGL) Tesla, Inc. (TSLA) GrokCoin (GROKCOIN) DeepSeek (DEEPSEEK) Sora (SORA) Robot Consulting Co., Ltd. (LAWR)
Top Social Posts
Top posts by engagements in the last [--] hours
"Count me as one of the skeptics No AGI by end of [----] mark my words. But I otherwise think @elonmusks comments @nytimes on AI safety and AI regulation have been measured and on target"
X Link 2023-11-29T23:17Z 207.8K followers, 92.4K engagements
"Put it differently: this is really bad. What factors could explain it The sudden uptick of generative AI in [----] - particularly more costly image generation (per @sashamtls analysis of costs for inferences) - might be a factor Other explanations Not good. Not good at all. https://t.co/bkcyWfznyq Not good. Not good at all. https://t.co/bkcyWfznyq"
X Link 2024-01-10T11:53Z 207.8K followers, 217.9K engagements
"Some AGI-relevant predictions I made publicly long before o3 about what AI could not do by the end of [----]. Do you seriously think o3-enhanced AI will solve any of them in next [----] months"
X Link 2024-12-21T14:02Z 207.8K followers, 83.9K engagements
"By my math the drop in tourism alone will more than offset the total (and in many cases careless) cuts from DOGE. We will wind up with a poorer country that works less well with higher prices and less freedom that fewer people want to visit that makes fewer technological breakthroughs. The US is one of the worlds top travel spots pulling in over [--] million visitors last year and generating $13 trillion. But lately travel from Europe and Canada is down by up to 35% thanks to Trumps talk and scary border stories. Trump is bankrupting the US. https://t.co/f8QMkxW6W3 The US is one of the worlds"
X Link 2025-04-12T20:00Z 207.5K followers, 73.1K engagements
"AGI This decade Next decade Next century (or never) Later this century This decade Next decade Next century (or never) Later this century"
X Link 2025-05-28T14:16Z 207.8K followers, 11.9K engagements
"The chance that we will get to AGI (or even just more modestly AI that we can trust) without facing this core problem is pretty much zero. Yet another manifestation of the basic problem that I have been emphasizing since 1998: pure neural networks cannot reliably generalize beyond the training distribution. Yet another manifestation of the basic problem that I have been emphasizing since 1998: pure neural networks cannot reliably generalize beyond the training distribution"
X Link 2025-09-25T22:22Z 207.6K followers, 17.5K engagements
"BREAKING: Leaked October [--] letter from OpenAI to White House shows that OpenAI had begun asking for Federal guarantees over a week ago. Sams long attempted walkback yesterday on X after the huge backlash was a total lie. You cannot trust this man ever. Thats what Ilya saw. Here is an OpenAI document submitted one week ago where they advocate for including datacenter spend within the American manufacturing umbrella There they specifically advocate For Federal loan guarantees Sam Lied to everyone Again https://t.co/kkCiBx4mN8 Here is an OpenAI document submitted one week ago where they"
X Link 2025-11-07T14:11Z 208.9K followers, 1.3M engagements
"@slow_developer [--]. Disinformation undermining democracy [--]. Massive cybersecurity incident [--]. Another AI winter once people realize how wildly GenAI was oversold"
X Link 2024-06-27T04:00Z 207.9K followers, [---] engagements
"Quote of the day maybe the year: *He is not an idiot Elon Musk is very smart very tech savvy. He could do a simple Google search of any of the disinformation he has seen in the last week and see it's untrue but instead he seeks to amplify it and perpetuate it. Page One in Edinburgh: He has vast wealth at his fingertips and he uses it for some of the most wicked evil Ive seen. @ukpapers $TSLA https://t.co/FjU0fB0NdZ Page One in Edinburgh: He has vast wealth at his fingertips and he uses it for some of the most wicked evil Ive seen. @ukpapers $TSLA https://t.co/FjU0fB0NdZ"
X Link 2024-08-09T19:47Z 208.9K followers, 184.9K engagements
"Black box AI has been a massive possibly fatal mistake wrt alignment. This man speaks truth. @GaryMarcus @OpenAI Verify everything. The humanity needs a whitebox AI @GaryMarcus @OpenAI Verify everything. The humanity needs a whitebox AI"
X Link 2025-07-20T03:57Z 208.4K followers, 110.9K engagements
"AI agents arent really working. Why GenAI is like a C student who has secretly been given all the exams from the last decade. He may pass the exam but take away the entire internets worth of cheat codes and theres just not much there. (screenshot from @jeremyakahn)"
X Link 2025-07-29T18:40Z 208.4K followers, 51.2K engagements
"On reddit OpenAI is getting absolutely torched consistent with the dark analysis I posted earlier today (see below). For years I warned that OpenAI might take a serious hit if GPT-5 was disappointing. We finally soon see whether I was right. In a rational world their valuation would take a hit. They no longer have anything like a clear technical lead. GPT-5 is unlikely to be ahead of For years I warned that OpenAI might take a serious hit if GPT-5 was disappointing. We finally soon see whether I was right. In a rational world their valuation would take a hit. They no longer have anything like"
X Link 2025-08-08T16:54Z 208.4K followers, 40.2K engagements
"How have I handled all the negative diatribes and lies about me over the last seven years Being honest I am human and dont enjoy them. But I have stuck to my principles and my understanding of the cognitive sciences which have been a good guide. I paid a price for doing that. But I knew that the truth would eventually come out. And now suddenly all it once it has. Science is not a popularity contest; you cant bully your way to truth. And science cannot advance without people sticking to the truth even in the face of opposition. I hope people will remember that"
X Link 2025-08-14T17:07Z 207.9K followers, 28K engagements
"@ejae_dev @bridgemindai cool. now a try a (new) game that doesnt have a lot of direct existing code examples"
X Link 2026-02-14T15:17Z 208.9K followers, [----] engagements
"With the exception of Nvidia which is selling shovels in a gold rush most generative AI companies are both wildly overvalued and wildly overhyped" Gary Marcus Emeritus Professor of Psychology and Neural Science at New York University told @DeutscheWelle. "My guess is that it will all fall apart possibly soon. The fundamentals technical and economic make no sense.""
X Link 2025-11-07T19:22Z 208.8K followers, 28.2K engagements
"Thats exactly why I came back. Gary Marcus is back on X to bring balance back to the AI hype train. Gary Marcus is back on X to bring balance back to the AI hype train"
X Link 2026-02-14T17:17Z 208.9K followers, 12.2K engagements
"For years I said todays systems were laughably bad and was crucified for it. 🤷♂ A conversation with OpenAI on what to expect in the next [--] months - today's systems are laughably bad - ChatGPT not long term engagement model - models capable of complex "work" - like a great team mate working with u - shift towards verbal interfaces & beyond Multimodality https://t.co/XW91fNNceY A conversation with OpenAI on what to expect in the next [--] months - today's systems are laughably bad - ChatGPT not long term engagement model - models capable of complex "work" - like a great team mate working with u"
X Link 2024-05-07T17:56Z 208.9K followers, 26.7K engagements
"Exactly what I have been trying to tell you. The day of reckoning is coming. I genuinely don't understand why some people are still bullish about LLMs. I use GPT Grok Gemini Mistral etc every day in the hope they'll save me time searching for information and summarizing it. They continue to fabricate links references and quotes like they did from I genuinely don't understand why some people are still bullish about LLMs. I use GPT Grok Gemini Mistral etc every day in the hope they'll save me time searching for information and summarizing it. They continue to fabricate links references and"
X Link 2025-03-27T08:13Z 208.9K followers, 41K engagements
"I never thought I would live to see the President of the United States serving as Russias Press Secretary"
X Link 2025-06-04T17:51Z 208.9K followers, 851.3K engagements
"Three thoughts on what really matters: [--]. Fuck cancer [--]. Friends are irreplaceable [--]. The new "Marcus test" for AI is when AI makes a significant dent on cancer May that happen sooner much sooner rather than later. In memory of my childhood friend Paul"
X Link 2025-08-31T17:56Z 208.9K followers, 321.4K engagements
"Go back and read our [----] book Rebooting AI; it was all there. The limits of statistics and pattern-matching; the desperate need for world models. Six years of grief and Davis and I were right all along. Amazing. "Todays AIs dont understand environments; they just pattern-match inside them." Literally what critics have been saying for years now. "Todays AIs dont understand environments; they just pattern-match inside them." Literally what critics have been saying for years now"
X Link 2025-10-31T16:13Z 208.9K followers, 139.4K engagements
"Hot take on Gemini [--]. Points [--] and [--] might literally change the world. Also OpenAI is in trouble. Great model but its not AGI; has the same kind of problems all earlier models have had. Hallucinations and unreliability persist. In short scaling isnt getting us to AGI OpenAI has basically squandered the technical lead it once had; Google has caught up. What happens to OpenAI if Google undercuts OpenAI on price But biggest news was buried: it got better results than it is competitors without using Nvidia GPUs If Google were to make those TPUs commercially available at scale and reasonable"
X Link 2025-11-19T16:33Z 208.9K followers, 313.8K engagements
"translation : Deep Learning Hit a Wall new techniques needed to go forward; just like my infamous [----] paper said. go back and read it "We are no longer in the age of scaling we are back to the age of research.' Kaplan scaling laws are flattening at current frontiers autoregressive transformers exhausted for reasoning/planning/alignment gains. Next jump needs real architectural breakthroughs - test-time https://t.co/2hDIvJvF1F "We are no longer in the age of scaling we are back to the age of research.' Kaplan scaling laws are flattening at current frontiers autoregressive transformers"
X Link 2025-11-26T01:56Z 208.9K followers, 78.6K engagements
"It is one thing for everyone from @sama to @ylecun to @elonmusk and the Twitterverse to have attacked me literally for years for saying that scaling alone would not get us to AGI. Another to pretend now that that never happened now that I have largely been proven correct"
X Link 2025-11-26T15:11Z 208.9K followers, 22.1K engagements
"This place is toxic. For the last seven years I warned you that LLMs and similar approaches would not lead us to AGI. Almost nobody is willing to acknowledge that even though so many of you gave me endless grief about it at the time. I also warned you - first - that Sam Altman could not be trusted that OpenAI would lose its dominance that GPT-5 would not be all AGI and that LLMs lacked world models. That hallucinations would not go away. That out of distribution generalization was THE key issue. And that the economics of LLMs didn't make sense. And that the LLM companies would start seeking"
X Link 2025-11-26T18:03Z 208.9K followers, 618.7K engagements
"How did this work out Are LLM hallucinations largely gone by now So now the @FT platforms the same guy saying most the of the tasks lawyers and accountants do will be replaced in 12-18 months From the same company that said that GPT-5 would a giant humpback whale that blow away PhDs Where is the accountability The concern about CEOs conflict of interest in selling these narrative The view from skeptics LLM hallucinations will be largely eliminated by [----]. thats a huge deal. the implications are far more profound than the threat of the models getting things a bit wrong today. LLM"
X Link 2026-02-12T20:14Z 208.9K followers, [---] engagements
"@ProtagorasTO @FT 5% on known benchmarks (often gamed) is hardly largely eliminated"
X Link 2026-02-13T03:35Z 208.9K followers, [----] engagements
"@deredleritt3r @FT @DamienCharlotin Sir you have changed the argument from hallucinations arent a problem to if you watch LLMs like a hawk you will be fine. The former is false; the latter might be true. See my long post to Dean Ball a moment ago for examples in other domains that converge on the same"
X Link 2026-02-13T15:25Z 208.9K followers, [----] engagements
"I cant retweet Jimmy Apples but also wonder why people take him seriously [--] seconds of web search would have told him that you can already find ascii fishing code repositories on the web. 🙄"
X Link 2026-02-14T17:12Z 208.9K followers, [----] engagements
"Congress needs to bring in Zuckerberg and LeCun to discuss how their unilateral open-sourcing decision rapidly undermined the US advantage in Generative AI. Tomorrow. Meta doesnt just look clueless they look like they sold out the US. Meta doesnt just look clueless they look like they sold out the US"
X Link 2025-01-27T00:21Z 208.9K followers, 215.2K engagements
"Personal Update: My p(doom) has gone up. I dont foresee machines with malice any time soon. But I am starting to see the harm that powerful yet reckless humans who are indifferent to humanity could cause"
X Link 2025-07-14T01:02Z 208.9K followers, 125.8K engagements
"@nikostro LMK on Jan 1"
X Link 2025-09-02T02:59Z 208.9K followers, [---] engagements
"Wow. Just wow. @ylecun taking credit for my March [----] argument that scaling would hit a wall and that LLM would not bring us to AGI--after he initially attacked me for saying it and continued to promote them right up until ChatGPT ate has lunch--has to among the most intellectually dishonest things I have seen a scientist do in my entire life. @SchmidhuberAI was right about LeCun"
X Link 2025-11-26T17:05Z 208.9K followers, 162.4K engagements
"Over [--] million views for that @MattShumer_ blog (Something Big is Happening) and not a shred of actual data. Nobody thought to ask what are the data Does this match reality Is Shumer credible Did he give a single concrete example People gonna believe what they want to believe but here are [--] critical points most people missed: Shumers blog post is weaponized hype filled with vivid narrative and marketing speech but stumbles on the facts especially with respect to reliability. He gives no actual data to support this claim that the latest coding systems can (reliably) write perfect whole"
X Link 2026-02-12T16:31Z 208.9K followers, 165.8K engagements
"@neuralamp4ever @mattshumer_ Thank you. This heads up re paid AI influencers was very helpful and I immediately mentioned it on a BBC World interview that will air shortly. Original source here: https://www.cnbc.com/2026/02/06/google-microsoft-pay-creators-500000-and-more-to-promote-ai.html https://www.cnbc.com/2026/02/06/google-microsoft-pay-creators-500000-and-more-to-promote-ai.html"
X Link 2026-02-12T17:15Z 208.9K followers, [----] engagements
"How did this work out Are LLM hallucinations largely gone by now So now the @FT platforms the same guy saying most the of the tasks lawyers and accountants do will be replaced in 12-18 months From the same company that said that GPT-5 would be a giant humpback whale that would blow away PhDs Where is the accountability The concern about CEOs conflicts of interest in selling these narratives The view from skeptics LLM hallucinations will be largely eliminated by [----]. thats a huge deal. the implications are far more profound than the threat of the models getting things a bit wrong today. LLM"
X Link 2026-02-12T20:17Z 208.9K followers, 241.8K engagements
"No way Man who literally begged governments for wholesale exemption from copyright law so he could free ride on artists and writers bitches about . free riders. 🙄 🚨 BREAKING: OpenAI just WARNED US lawmakers that DeepSeek is using distillation to free-ride on capabilities developed by OpenAI this is part of the CCPs playbook: steal copy and kill @sama knows its over for ClosedAI Once DeepSeek drops V4 and R2 it will easily https://t.co/3zpUktXpsN 🚨 BREAKING: OpenAI just WARNED US lawmakers that DeepSeek is using distillation to free-ride on capabilities developed by OpenAI this is part of"
X Link 2026-02-12T23:13Z 208.9K followers, 151.8K engagements
"@SoniqueBang nobody will win. CCP is not going to crush Google and Google is not going to crush the chinese companies"
X Link 2026-02-12T23:29Z 208.9K followers, 18.6K engagements
"It would be one the great ironies of history if the biggest IP thief in history was taken down by mass IP theft. I think that sound is god laughing. 🚨 BREAKING: OpenAI just WARNED US lawmakers that DeepSeek is using distillation to free-ride on capabilities developed by OpenAI this is part of the CCPs playbook: steal copy and kill @sama knows its over for ClosedAI Once DeepSeek drops V4 and R2 it will easily https://t.co/3zpUktXpsN 🚨 BREAKING: OpenAI just WARNED US lawmakers that DeepSeek is using distillation to free-ride on capabilities developed by OpenAI this is part of the CCPs"
X Link 2026-02-12T23:44Z 208.9K followers, 67.3K engagements
"Honestly OpenAI should get the IP protections it deserves but only if it first properly licenses the IP of those they themselves ripped off. SCOOP: OpenAI warned lawmakers in a memo sent today to the House Select Committee on China that DeepSeek is using new obfuscated methods to continue to distill its AI models as well as those of other US frontier AI labs https://t.co/OsWxPRMF28 w/ @eastland_maggie SCOOP: OpenAI warned lawmakers in a memo sent today to the House Select Committee on China that DeepSeek is using new obfuscated methods to continue to distill its AI models as well as those of"
X Link 2026-02-13T00:36Z 208.9K followers, 14.4K engagements
"@deredleritt3r @FT and yet lawyers keep getting busted for fake cases in their briefs per @DamienCharlotins database Pretty much every day at much a higher clip than two years ago: https://www.damiencharlotin.com/hallucinations/ https://www.damiencharlotin.com/hallucinations/"
X Link 2026-02-13T14:48Z 208.9K followers, 16.6K engagements
"Are LLM hallucinations basically solved as a former Senior Policy Advisor at the White House @deanwball told me below based pure on anecdotal experience and without data No. Instead his post is symptomatic of how subjective finger-in-the-wind evaluations of AI often get things wrong. Here are some actual data. Law: Incidents in which lawyers have gotten busted for using fake cases are way up; @DamienCharlotins database listed around [---] cases less than a year ago and now has over [---] with new incidents reported ever day. Science: The situation is so bad that a team of Japanese reseachers just"
X Link 2026-02-13T15:23Z 208.9K followers, 38.3K engagements
"@deanwball With due respect I think you are off the mark here @deanwball. Some data that you should consider: Are LLM hallucinations basically solved as a former Senior Policy Advisor at the White House @deanwball told me below based pure on anecdotal experience and without data No. Instead his post is symptomatic of how subjective finger-in-the-wind evaluations of AI often get Are LLM hallucinations basically solved as a former Senior Policy Advisor at the White House @deanwball told me below based pure on anecdotal experience and without data No. Instead his post is symptomatic of how"
X Link 2026-02-13T15:26Z 208.9K followers, [----] engagements
"@ejae_dev wonder how much specific training there has been around it though"
X Link 2026-02-13T18:45Z 208.9K followers, [----] engagements
"@chatgpt21 @ejae_dev how can you say with confidence there tons of web hits for pelican bicycle svg and its a well-known benchmark"
X Link 2026-02-13T19:25Z 208.9K followers, [---] engagements
"60 million people on X panicked over a data-free screed about how AI is about to obviate the need for people and IBM triples its hiring the same week. Can't make that shit up"
X Link 2026-02-13T19:33Z 208.9K followers, 17.9K engagements
"@BishopBlougram @chatgpt21 @ejae_dev there may be real progress and maybe not and its really hard to know from the outside. best would be to try same system on a bunch of SVGs that are novel to their literature but challenging"
X Link 2026-02-13T19:45Z 208.9K followers, [---] engagements
"Reposting this/reupping it in light of xAI no longer having a safety team per scoop from @haydenfield Personal Update: My p(doom) has gone up. I dont foresee machines with malice any time soon. But I am starting to see the harm that powerful yet reckless humans who are indifferent to humanity could cause. Personal Update: My p(doom) has gone up. I dont foresee machines with malice any time soon. But I am starting to see the harm that powerful yet reckless humans who are indifferent to humanity could cause"
X Link 2026-02-13T21:20Z 208.9K followers, [---] engagements
"@aran_nayebi Love how you ignore ALL the data and most of the argument I presented and prefer your own anecdotal experience. Stick to the bit"
X Link 2026-02-14T15:13Z 208.9K followers, [----] engagements
"@aran_nayebi @peterevoss @BlueGuardrails did you try GPT-5.2-Pro is it vastly better"
X Link 2026-02-14T16:23Z 208.9K followers, [---] engagements
"We URGENTLY need a federal law forbidding AI from impersonating humans The night before I testified in the US Senate in May [----] the late philosopher Daniel Dennett sent me a manuscript that he called counterfeit people. It was published a few days later in The Atlantic. It started like this MONEY HAS EXISTED for several thousand years and from the outset counterfeiting was recognized to be a very serious crime one that in many cases calls for capital punishment because it undermines the trust on which society depends. Today for the first time in history thanks to artificial intelligence it"
X Link 2026-02-14T19:56Z 208.9K followers, 13.6K engagements
"Since @OpenAI still has not changed misleading blog post about "solving the Rubik's cube" I attach detailed analysis comparing what they say and imply with what they actually did. IMHO most would not be obvious to nonexperts. Please zoom in to read & judge for yourself"
X Link 2019-10-19T22:08Z 208.9K followers, [----] engagements
"I will always stand in awe of Noam Chomsky. I just sent him my recent essay; [--] minutes later he replied with smart comments including a point I should have thought to include. Hes 93"
X Link 2022-03-12T22:06Z 208.9K followers, [----] engagements
"Woker-than-thou"
X Link 2022-12-25T14:40Z 208.9K followers, 575.8K engagements
"Speaking of bullshit anyone remember this"
X Link 2023-03-19T23:27Z 208.9K followers, 700K engagements
"Take this seriously: @geoffreyhinton on AI possibly wiping out humanity: It's not inconceivable Is coding faster and having fun chatbots to play with worth a 1% risk of that coming true https://www.foxnews.com/tech/artificial-intelligence-godfather-ai-possibly-wiping-humanity-not-inconceivable https://www.foxnews.com/tech/artificial-intelligence-godfather-ai-possibly-wiping-humanity-not-inconceivable"
X Link 2023-03-27T12:07Z 208.9K followers, 1.1M engagements
"a big deal: @elonmusk Y. Bengio S. Russell @tegmark V. Kraknova P. Maes @Grady_Booch @AndrewYang @tristanharris & over [----] others including me have called for a temporary pause on training systems exceeding GPT-4 https://futureoflife.org/open-letter/pause-giant-ai-experiments/ https://futureoflife.org/open-letter/pause-giant-ai-experiments/"
X Link 2023-03-29T01:10Z 208.9K followers, 7.1M engagements
"👉Seismic from @geoffreyhinton: The best hope is for the worlds leading scientists to collaborate on ways of controlling the technology. I dont think they should scale this up more until they have understood whether they can control it https://www.nytimes.com/2023/05/01/technology/ai-google-chatbot-engineer-quits-hinton.htm https://www.nytimes.com/2023/05/01/technology/ai-google-chatbot-engineer-quits-hinton.htm"
X Link 2023-05-01T12:44Z 208.9K followers, 851.3K engagements
"remember how I said the quality of the internet was going to go down https://futurism.com/top-google-result-johannes-vermeer-ai-generated-knockoff https://futurism.com/top-google-result-johannes-vermeer-ai-generated-knockoff"
X Link 2023-06-06T00:18Z 208.9K followers, 601.6K engagements
"@EdKrassen NBC interview was pre Ukraine invasion. I dont think it is appropriate to give him a platform now"
X Link 2023-08-28T01:44Z 208.9K followers, 75.2K engagements
"Common view; almost certainly wrong. The boards decision a . Instead it was almost certainly a Hail Mary: 👉This Boards sole job was to look out for humanity - NOT to protect the brand. 👉They must have seen danger in something Sam was doing 👉They must have known that everyone in the world would be watching and that their decision would not be popular. 👉They proceeded anyway despite likely personal costs and despite knowing they might not succeed because they felt there was no other option. Even now their sole goal seems to be make sure that their successors protect the mission. Respect. 💪"
X Link 2023-11-20T00:10Z 208.9K followers, 2.6M engagements
"The number of people simply unable to imagine that the Board acted in a good faith with a legitimate and serious concern boggles my mind. We dont of course know what happened but simply dismissing that possibility seems foolish to me especially given that they were not operating out of financial self interest unlike all the other actors here"
X Link 2023-11-20T02:08Z 208.9K followers, 221K engagements
"Appalled that @timnitgebru AI ethics icon cant lay off a lovely 87-year-old man one of the great scientists of our time whose journalist son was kidnapped and publicly beheaded in an act of political violence. Her power has gone to her head; her humanity has vanished"
X Link 2023-12-03T02:19Z 208.9K followers, 788.1K engagements
"Remember how Sam Altman told the US Senate he had no direct investment in OpenAI and how they gushed over his apparent selflessness 👉He didnt mention that (presumably) he had equity in YC which likely has equity in OpenAI. And now this: 👉OpenAI Agreed to Buy $51 Million of AI Chips From a Startup Backed by CEO Sam Altman #candor"
X Link 2023-12-03T18:11Z 208.9K followers, 976.1K engagements
"As a graduate of MIT I literally couldnt watch this to the end; it was too painful. I honestly dont understand how things got to this point. "This is the same climate of antisemitism that has led to the massacre of Jews throughout the centuries. This is not just harassment. This is our lives on the line." @MIT student Talia Khan highlights the rise of antisemitism at MIT. https://t.co/zXb03xodXb "This is the same climate of antisemitism that has led to the massacre of Jews throughout the centuries. This is not just harassment. This is our lives on the line." @MIT student Talia Khan highlights"
X Link 2023-12-06T06:46Z 208.9K followers, 425.6K engagements
"OpenAI is in a heap of trouble and its not just text. Long thread why (1/n) based on work with @Rahll"
X Link 2023-12-29T20:36Z 208.9K followers, 1.2M engagements
"@Rahll @latimes indeed @bcmerchant we are dropping something big tomorrow. email me for a sneak preview ;)"
X Link 2024-01-03T16:20Z 207.5K followers, [---] engagements
"Potential copyright infringement & why generative AI may be in for a rough ride Video below on how widespread the problem is and a [----] word deep dive w @rahll @ 👉 Why film & game studios artists & actors may sue 👉 Why users are at risk 👉 Why the issue could be hard to fix 👉 How scaling could be making the problem worse https://spectrum.ieee.org/midjourney-copyright https://spectrum.ieee.org/midjourney-copyright https://spectrum.ieee.org/midjourney-copyright https://spectrum.ieee.org/midjourney-copyright"
X Link 2024-01-06T15:21Z 207.5K followers, 130.1K engagements
"Rough Translation: We wont get fabulously rich if you dont let us steal so please dont make stealing a crime Dont make us pay fees either Sure Netflix might pay billions a year in licensing fees but we shouldnt have to More money for us moar OpenAI lobbying for copyright law revision in the UK @GaryMarcus @jason_kint https://t.co/kug5rCpINT OpenAI lobbying for copyright law revision in the UK @GaryMarcus @jason_kint https://t.co/kug5rCpINT"
X Link 2024-01-08T14:16Z 208.9K followers, 910.8K engagements
"Now we know why Sam Altman went around the world last summer meeting world leaders: his company wont make it big unless they can convince governments to give them one of the biggest handouts in history. OpenAI lobbying: (a) ChatGPT useless without copyright material (b) Requires special exemptions from the law. https://t.co/yW6DRWSoBF So LLMs are thinly disguised plagiarization algorithms and pilfering copyright is the only way to make them profitable right @GaryMarcus https://t.co/hILAkYpprO OpenAI lobbying: (a) ChatGPT useless without copyright material (b) Requires special exemptions from"
X Link 2024-01-08T16:47Z 208.9K followers, 94.5K engagements
"WOW Its about time somebody looked into this Potentially enormous. Breaking @theinformation"
X Link 2024-01-10T16:57Z 208.9K followers, 148.9K engagements
"Black Mirror has arrived ahead of schedule. An entire cast of deepfaked people tricked a CFO out of $25 million. (In the) multi-person video conference it turns out that everyone was fake Deepfaked shit is getting real"
X Link 2024-02-04T17:01Z 208.9K followers, 1.2M engagements
"The fun never ends"
X Link 2024-02-08T06:19Z 208.9K followers, 203.6K engagements
"When I started working on AI four decades ago it simply didnt occur to me that one of the biggest use cases would be derivative mimicry transferring value from artists and other creators to megacorporations using massive amounts of energy. This is not the AI I dreamed of. OpenAI just released a model that can generate 1-minute videos. You simply cannot argue that these models don't / won't compete with the content they're trained on and the human creators behind that content. What is the model trained on Did the training data providers consent OpenAI just released a model that can generate"
X Link 2024-02-15T18:53Z 208.9K followers, 840.6K engagements
"@SemanticEntity People pay for samples and consider stealing to be stealing"
X Link 2024-02-15T19:26Z 208.9K followers, 32.9K engagements
"Amazing But um ants have six legs. We are about to have a whole generation of children educated by fake videos that are completely plausible to naive audiences yet biologically incorrect. 🤯 "pov footage of an ant navigating the inside of an ant nest" Video generated by Sora https://t.co/N3Ik9QBinU "pov footage of an ant navigating the inside of an ant nest" Video generated by Sora https://t.co/N3Ik9QBinU"
X Link 2024-02-18T04:31Z 208.9K followers, 299.7K engagements
"Musicians and Artists does this sound familiar From one musician I know:"
X Link 2024-02-20T18:19Z 208.9K followers, 200.9K engagements
"A lot of people are missing the funny part. Let me break it down. If OpenAI has actually achieved AGI they get their software back from Microsoft. Elons lawsuit has put them in a position having to prove that they havent reached AGI even though OpenAI likes to hint that they have or are close. So they need an expert witness to testify e.g. that their partner Microsoft was full of it when they said that GPT-4 represented Sparks of AGI. Which means hiring someone like me to save them. 🤣🤣🤣"
X Link 2024-03-02T19:12Z 208.9K followers, 306.3K engagements
"Dear @openAI This is baloney: We are dedicated to the OpenAI mission and have pursued it every step of the way. The original mission was to be unconstrained by a need to generate a financial return not organized for private gain seeking to open source technology where applicable. Now Microsoft has an exclusive 49% of your profits go them and rather than helping all humanity you are stealing from artists writers etc to increase your profits. For good measure you have almost entirely abandoned open-source and are among the least open AI companies anywhere. Whatever your history with @elonmusk"
X Link 2024-03-06T03:00Z 208.9K followers, 202.2K engagements
"If all we had was ChatGPT we could say hmm maybe hallucinations are just a bug and fantasize that they werent hard to fix. If all we had was Gemini we could say hmm maybe hallucinations are just a bug. If all we had was Mistral we could say hmm maybe hallucinations are just a bug. If all we had was LLAMA we could say hmm maybe hallucinations are just a bug. If all we had was Grok we could say hmm maybe hallucinations are just a bug. Instead we need to wake up and realize that hallucinations are absolutely core to how LLMs work and that we need new approaches based on different ideas"
X Link 2024-04-20T19:40Z 208.9K followers, 538K engagements
"So many people are confused about the relation between human cognitive errors and LLM hallucinations that I wrote this short explainer: Humans say things that aren't true for many different reasons Sometimes they lie Sometimes they misremember things Sometimes they fail to think through what they are saying Sometimes they are on drugs Sometimes they suffer from mental disorders etc LLMs errors result from . They don't have (e.g.) intentions egos or financial interests so they don't lie. They don't take drugs. They don't have emotional states. Instead LLM "hallucinations" arise regularly"
X Link 2024-04-21T17:03Z 208.9K followers, 211.5K engagements
"GPT-4o hot take: The speech synthesis is terrific reminds me of Google Duplex (which never took off). but If OpenAI had GPT-5 they have would shown it. They dont have GPT-5 after [--] months of trying. The most important figure in the blogpost is attached below. And the most important thing about the figure is that 4o is not a lot different from Turbo which is not hugely different from [--]. Lots of quirky errors are already being reported same as ever. (See e.g. examples from @RosenzweigJane and @benjaminjriley.) OpenAI has presumably pivoted to new features precisely because they dont know how"
X Link 2024-05-13T20:49Z 208.9K followers, 1.2M engagements
"Someday my children will ask me why anyone ever took seriously the thought that systems like this had anything to do with AGI. I honestly have no idea how I will answer. Mass delusion (Example emailed to me by a friend.)"
X Link 2024-05-14T23:17Z 208.9K followers, 310.2K engagements
"The OpenAI board was right. Even on little things Sam is not consistently candid. Earlier today OpenAI said the resemblance to ScarJo was a coincidence. It was not: Statement from Scarlett Johansson on the OpenAI situation. Wow: https://t.co/8ibMeLfqP8 Statement from Scarlett Johansson on the OpenAI situation. Wow: https://t.co/8ibMeLfqP8"
X Link 2024-05-20T22:28Z 208.9K followers, 613.3K engagements
"I think @elonmusk owes Apple an apology. I wont say this is perfect (hackers will inevitably find problems) but it seems to me to be quite thoughtfully conceived and likely is state of the art. This is from Apple's State of the Union The local model is a 3B parameter SLM that uses adapters trained for each specific feature. Diffusion model does the same thing adapter for each style. Anything running locally or Apple's Secure Cloud is an Apple model not OpenAI. https://t.co/BRkWAQQCQI This is from Apple's State of the Union The local model is a 3B parameter SLM that uses adapters trained for"
X Link 2024-06-11T17:39Z 208.9K followers, 312.6K engagements
"Wow evidence from OpenAI itself in favor of my conjecture of diminishing returns. Mira Murati says the AI models that OpenAI have in their labs are not much more advanced than those which are publicly available https://t.co/OtgXhZg13P Mira Murati says the AI models that OpenAI have in their labs are not much more advanced than those which are publicly available https://t.co/OtgXhZg13P"
X Link 2024-06-13T00:31Z 208.9K followers, 297K engagements
"I dont think people quite realize the significance of this moment so I will spell it out: OpenAI has been bluffing all along and their CTO just gave away the fact that they are not holding aces. 75% chance that the GenAI bubble will have burst by [--] months from today. Wow evidence from OpenAI itself in favor of my conjecture of diminishing returns. Wow evidence from OpenAI itself in favor of my conjecture of diminishing returns"
X Link 2024-06-13T03:35Z 208.9K followers, 250.6K engagements
"Have to say I am with LeCun on this one"
X Link 2024-07-21T22:11Z 208.9K followers, 80.3K engagements
"Companies that have tried hard to create GPT-5 level models and thus far failed: OpenAI Google Microsoft Anthropic Meta http://X.AI http://X.AI"
X Link 2024-07-23T21:42Z 207.5K followers, 10.7K engagements
"SCOOP: OpenAI may lose $5B this year & may run out of cash in [--] months unless they raise more $ per analysis @theinformation. Investors should ask: What is their moat Unique tech What is their route in profitability when Meta is giving away similar tech for free Do they have a killer app Will the tech ever be reliable What is real and what is just demo"
X Link 2024-07-24T14:19Z 208.9K followers, 968.8K engagements
"Incredibly damning new study on GenAI from Upwork via @baldurbjarnason: Nearly half (47%) of workers using AI say they have no idea how to achieve the productivity gains their employers expect. Over three in four (77%) say AI tools have decreased their productivity and added to their workload in at least one way. Says Bjarnason Its quite unusual for a study like this on a new office tool roughly two years after that toolChatGPTexploded into peoples workplaces to return such a resoundingly negative sentiment. Many on X may love ChatGPT for work but much of the outside world feels quite"
X Link 2024-07-29T14:23Z 208.9K followers, 321K engagements
"Never thought I would do this but I have unfollowed Elon. Reading his relentless distortion is ruining my experience here"
X Link 2024-07-30T14:58Z 208.9K followers, 105K engagements
"OMG The GenAI honeymoon is definitely over. A pharma company stopped using Microsoft's Copilot AI tool with an exec citing high cost and low value. The company used Office [---] Copilot for [---] staff and compared it to middle-school presentations. -@BusinessInsider"
X Link 2024-07-31T16:07Z 208.9K followers, 291.2K engagements
"I just wrote a great piece for WIRED predicting that the AI bubble will in collapse in [----] and now I wish I hadnt. Clearly I got the year wrong. Its going to be days or weeks from now not months"
X Link 2024-08-03T00:05Z 208.9K followers, 1.6M engagements
"⚠ Something terrifying just snapped in place: Sam Altman has acknowledged wanting to train on everyone's personal documents (Word files email etc) ChatGPT has gathered unprecedented amounts of personal data Sam founded WorldCoin known for their eye-scanning orb OpenAI recently put Nakasone (ex NSA) on the board OpenAI just put in money in a $60M fundraise with a Webcam company and is planning hardware joint venture with them Altman wants to know - and monetize - everything about you. He might succeed"
X Link 2024-08-07T14:39Z 208.9K followers, 115.8K engagements
"I am learning so much from Googles AI"
X Link 2024-08-15T05:20Z 208.9K followers, 145.1K engagements
"What more proof do you need that Elon is living in a bubble than this"
X Link 2024-08-20T23:59Z 208.9K followers, 153.7K engagements
"Pretending that Elon Musk is centrist is the new Flat Earth movement"
X Link 2024-08-21T16:05Z 208.9K followers, 99.3K engagements
"Fantastic essay on AI and art today in The New Yorker"
X Link 2024-09-01T04:02Z 208.9K followers, 58.2K engagements
"Crikey AI worse than humans in every way at summarizing information the technology might actually make more work for people not less @crikey_news"
X Link 2024-09-04T19:58Z 208.9K followers, 119K engagements
"Yann LeCun: AI wont be used to generate misinformation in science deliberately or accidentally. Its not a problem. Actual research: Its already happening"
X Link 2024-09-12T13:06Z 208.9K followers, 63.9K engagements
"We are watching Elon meltdown in real-time. If his friends cared for him truly cared for him they would intervene"
X Link 2024-09-14T19:47Z 208.9K followers, 264.3K engagements
"Wait what Murati is leaving Schulman left Karpathy left Ilya left Leike left Brockman is on leave perhaps a dozen others left GPT-5 hasnt dropped Sora hasnt shipped the company had an operating loss of $5b last year there is no obvious moat Meta is giving away similar software for freemany lawsuits pending and people are valuing this company at $150 billion dollars Absolutely insane. Investors shouldnt be pouring more money at higher valuations they should be asking what is going on. I shared the following note with the OpenAI team today. https://t.co/nsZ4khI06P I shared the following note"
X Link 2024-09-25T20:19Z 208.9K followers, 1.9M engagements
"🎉HUGE news it means OpenAI and others can no longer hide behind the BS phrase publicly available. If they scrape something off the internet that is copyrighted or trademarked they are now obligated to disclose that. Newsom just signed a new bill requiring Gen AI businesses to publicly disclose their training data 👀 https://t.co/L6obdI5kzz Newsom just signed a new bill requiring Gen AI businesses to publicly disclose their training data 👀 https://t.co/L6obdI5kzz"
X Link 2024-09-29T15:09Z 208.9K followers, 83.8K engagements
"The big takeway from tonight: Elon is spending too much time on politics and not enough on Tesla"
X Link 2024-10-11T03:20Z 208.9K followers, 87.4K engagements
"👇Superb new article from @apple AI: we found no evidence of formal reasoning in language models . Their behavior is better explained by sophisticated pattern matchingso fragile in fact that changing names can alter results by 10% where changing a word or two in irrelevant ways can give you a different answer. Strongly encourage you to read the whole thread. 1/ Can Large Language Models (LLMs) truly reason Or are they just sophisticated pattern matchers In our latest preprint we explore this key question through a large-scale study of both open-source like Llama Phi Gemma and Mistral"
X Link 2024-10-11T14:57Z 208.9K followers, 498.5K engagements
"I literally dont see how democracy can survive long term if the richest and most powerful people are free to lie at scale with no consequence"
X Link 2024-10-16T16:40Z 208.9K followers, 133K engagements
". Making a million dollar prize contingent on registering to vote is illegal. See [--] USC 10307(c) https://t.co/8hnQS6lqYV Also with PA registration soon closing for this election it may not accomplish what Musk wants. Making a million dollar prize contingent on registering to vote is illegal. See [--] USC 10307(c) https://t.co/8hnQS6lqYV Also with PA registration soon closing for this election it may not accomplish what Musk wants"
X Link 2024-10-20T01:39Z 208.9K followers, 178.4K engagements
"Hot take on the mess Elon just made by pushing too far: Everyone of Elon Musks boards now has to consider that he has just flagrantly violated federal law and may go to jail for it. That is risk to his companies and because of their fiduciary responsibilities to his boards. If anything tanks lawsuits could be epic. Boards will be asked why they allowed this to go on for so long. . "
X Link 2024-10-20T02:00Z 208.9K followers, 155.9K engagements
"Why Musk will sacrifice democracy and maybe even jail in one convenient chart @nytimes"
X Link 2024-10-21T00:23Z 208.9K followers, 843.3K engagements
"When LeCun is right hes right"
X Link 2024-10-31T22:47Z 208.9K followers, 276.1K engagements
"⚠⚠⚠ I have a PhD in Brain and cognitive science from MIT. / : - : swaying missing and and stumbling in front of the garbage truck - : inappropriate musical pause/swaying similar to longer incident a few weeks ago - : threatening Liz Cheney with violence - : description of Musk as good at computer Things are degenerating fast"
X Link 2024-11-01T14:22Z 208.9K followers, 1.1M engagements
"I would very much welcome the thoughts of anyone specializing in neurology. Whats best diagnosis here Prognosis"
X Link 2024-11-01T15:04Z 207.5K followers, 94.7K engagements
"A bit over [--] hours later Trump showed at least [--] more signs of possible dementia today including more inappropriate foul language (no surprise) more signs of disorientation and the worlds first effort by a presidential candidate to mimic fellatio in public yet another remarkable failure to inhibit socially inappropriate responses confirming my overall impression. Whatever disease he has is likely progressing rapidly"
X Link 2024-11-02T05:06Z 208.9K followers, 97.3K engagements
"and why he may soon wind up in jail. TLDR: Elons own lawyers just screwed up really badly. Longer explanation background first: Musk ran a lottery in PA to try to get MAGA fans to register to vote. Thats very likely a violation of federal election law TBD. But separately it also flouts PA lottery law. Philadelphia DA brought him to court for it. In response and this is the insane part Musks lawyers tried to weasel alleging in court that There is no prize (just people being hired to do ads) and that the winners are not chosen at random. - . . How do we know this Because when Elon announced"
X Link 2024-11-04T16:38Z 208.9K followers, 157.2K engagements
"Folks game over. I won. GPT is hitting a period of diminishing returns just like I said it would. news: OpenAI's upcomning Orion model shows how GPT improvements are slowing down It's prompting OpenAI to bake in reasoning and other tweaks after the initial model training phase. https://t.co/VwD1xSbZUv news: OpenAI's upcomning Orion model shows how GPT improvements are slowing down It's prompting OpenAI to bake in reasoning and other tweaks after the initial model training phase. https://t.co/VwD1xSbZUv"
X Link 2024-11-09T22:50Z 208.9K followers, 361.2K engagements
"An MIT professor just asked me how can you be so confident that OpenAI isnt very close to AGI Here is a thread of links to a several recent observations that I think show how far we are from robust intelligence. Importantly all of them are known problems many that I have noted repeatedly for years and even decades. There is no principled solution thus far to any of them. 🧵"
X Link 2025-01-06T04:44Z 208.9K followers, 362.7K engagements
"The countdown until we are told that LLMs are too big to fail starts now. We cant afford to lose to China they will say accepting their multibillion dollar bailouts"
X Link 2025-01-06T20:14Z 208.9K followers, 430K engagements
"Lets wreck the environment to build an unreliable technology that has made little revenue which China can now build much more efficiently. big. beautiful. buildings. https://t.co/8lAvUufXFm big. beautiful. buildings. https://t.co/8lAvUufXFm"
X Link 2025-01-24T23:59Z 207.5K followers, 58.6K engagements
"People here are strangely quiet about a private unelected citizen taking over the U.S. government. A bit like in this cartoon"
X Link 2025-02-03T13:52Z 208.9K followers, 52.7K engagements
"I retired from the American University system at age [--] (after [--] years) so have no direct stake in this. But am here to say that drastically cutting NIH & NSF funding and reducing grant overhead to 15% are among the dumbest ideas I have ever seen. America will never be the same. Science will take a huge hit; academic stars will leave; university reputations will crumble and homegrown talent will be even harder to find. Asia and Europe will profit. Epic unforced errors"
X Link 2025-02-08T14:36Z 208.9K followers, 631.1K engagements
"Suppose the air conditioning in your car stopped working and you brought the car into the shop and the mechanic said I am afraid your car is broken so I have started dismantling it. That in a nutshell is what Trump and Musk are doing to your country. I disagree. The education system is broken and needs to be dismantled I disagree. The education system is broken and needs to be dismantled"
X Link 2025-02-09T14:28Z 208.9K followers, 99.4K engagements
"⚠⚠⚠ Everyoneand not just @theinformationshould be genuinely terrified that the richest man in the world has built a Large Language Model that spouts propaganda in his image. This isnt free speech. Its a massive massive thumb on the scale from one of the most powerful yet unelected people in the world. If that doesnt scare you you arent paying attention. Grok [--] is so based 😂 https://t.co/Rfphquqk2b Grok [--] is so based 😂 https://t.co/Rfphquqk2b"
X Link 2025-02-16T15:42Z 208.9K followers, 112.8K engagements
"If you live in the U.S. your civil rights will soon be over. This is not a drill"
X Link 2025-03-21T20:10Z 208.9K followers, 523.5K engagements
"God I wish @elonmusk would have taken this bet"
X Link 2025-04-25T10:52Z 208.9K followers, 98.3K engagements
"Horseshit. Every business in the world has discovered in the last several months that GenAI is not in fact smart enough replace most of their employees. Whatever you are reading from these (often gamed sometimes contaminated) benchmarks does not reflect real-world reality. The top AI model is now smarter than 85% of humans. By the end of [----] it will be smarter than 99.9% of humans. And you think you will still have a job https://t.co/4R4ZsuYJZY The top AI model is now smarter than 85% of humans. By the end of [----] it will be smarter than 99.9% of humans. And you think you will still have a"
X Link 2025-05-07T23:39Z 208.9K followers, 147.1K engagements
"A computer scientists perspective on vibe coding:"
X Link 2025-05-17T00:25Z 208.9K followers, 846.3K engagements
". I will break it down for you in four sentences: It is not just that these systems are doing more pattern recognition than reasoning. It is that LRM reasoning models . You cant have reliable AGI without the reliable execution of algorithms. And we will never get to alignment without them"
X Link 2025-06-08T14:22Z 208.9K followers, 177.8K engagements
"Sam I dont take issue with OpenAIs achievements I take issue with the hype and a history of misleading statements (some detailed below). Youve built something impressive and in many ways groundbreaking. I would be singing OpenAI's praises it's amazing how you scaled Transformers - if it werent for the hype much of it fanned by you that has massively overstated what the technology can do today and in the near future going back to the Rubik's cube demonstration that misrepresented what you had done. I think this hype is harming the world. If I seem frustrated its because of the damage and the"
X Link 2025-06-11T01:36Z 208.9K followers, 141.8K engagements
"Terence Tao reporting on the same problem we have seen endlessly in other domains: LLMs produce output that looks correct but is often deeply wrong and even stupid on careful inspection. I know of no domain in which this is NOT the case. And yet people seem surprised over and over to discover it anew in fresh domains. and non experts often dont see it at all. No solution to this problem no AGI. Scaling has not solved it. Terence Tao says today's AIs pass the eye test -- but fail miserably on the smell test. They generate proofs that look flawless. But the mistakes are subtle and strangely"
X Link 2025-06-15T06:12Z 208.9K followers, 250K engagements
"⚠ Everybody who uses LLMs should read this thread: LLM users consistently underperformed on multiple cognitive and neural measures. Its a hefty 206-page research paper and the findings are concerning. "LLM users consistently underperformed at neural linguistic and behavioral levels" This study finds LLM dependence weakens the writers own neural and linguistic fingerprints. 🤔🤔 Relying only on EEG https://t.co/pRBQ15qst3 Its a hefty 206-page research paper and the findings are concerning. "LLM users consistently underperformed at neural linguistic and behavioral levels" This study finds LLM"
X Link 2025-06-17T02:12Z 208.9K followers, 127.2K engagements
"BREAKING: Explosive new paper from MIT/Harvard/UChicago. Things just got worse a lot worse for LLMs and the myth that they can understand and reason. The paper documents a pattern they called Potemkins a kind of reasoning inconsistency (see figure below). They show that LLMs - even models like o3 make these errors frequently. You cant possibly create AGI based on machines that cannot keep consistent with their own assertions. You just cant. success on benchmarks only demonstrates potemkin understanding: the illusion of understanding driven by answers irreconcilable with how any human would"
X Link 2025-06-27T16:05Z 208.9K followers, 423.8K engagements
"what almost every amateur who has been sold bullshit about vibe coding is eventually going to encounter"
X Link 2025-07-08T06:58Z 208.9K followers, 243.8K engagements
"Feel the AGI @sama. No PhD student can match this Oh brother https://t.co/k2ZrO7ygSU Oh brother https://t.co/k2ZrO7ygSU"
X Link 2025-08-08T06:22Z 208.9K followers, 702.8K engagements
"Sorry but this tweet did not age well. https://t.co/1u0MOGvJWU https://t.co/1u0MOGvJWU"
X Link 2025-08-08T18:35Z 208.9K followers, 911.7K engagements
"what did Ilya see cuz I aint seeing it"
X Link 2025-08-08T20:33Z 207.5K followers, 73.6K engagements
"My work here is truly done. Nobody with intellectual integrity can still believe that pure scaling will get us to AGI. GPT-5 may be a moderate quantitative improvement (and it may be cheaper) but it still fails in all the same qualitative ways as its predecessors on chess on reasoning in vision; even sometimes on counting and basic math. Hallucinations linger. Dozens of shots on goal (Grok Claude Gemini) etc have invariably faced the same problems. Thats exactly what it means to hit a wall and exactly the set of obstacles I described in [----]. No hypothesis has ever been given more benefit of"
X Link 2025-08-08T21:59Z 208.9K followers, 233K engagements
"Mark my words this week marks peak bubble. Oracle just went up $300B on what appears to be a non-binding deal with OpenAI. OpenAI doesnt have $300 billion dollars They dont have anywhere near $300 billion dollars By their own (presumably optimistic) projection they wont turn a profit until [----]. And all this from a company thought (or claimed) that GPT-5 was going to be tantamount to AGI (spoiler alert: it wasnt) For good measure Oracle doesnt have the chips they would need to fulfill the contracts or even the cash to buy them. I will be astounded if Oracle meets their projections. Further"
X Link 2025-09-11T15:35Z 208.9K followers, 833.7K engagements
"An $800 billon dollar shortfall based almost entirely on the ever-more questionable gamble that scaling will lead to AGI. A staggering $800 billion global shortfall in AI infrastructure spend is foreseen by [----] according to a new report by Bain and that could hamper AI's progress @business (Gift link: free to read until Sept 30) https://t.co/HvwybrC1aD #AI https://t.co/PN8LZ7OfpV A staggering $800 billion global shortfall in AI infrastructure spend is foreseen by [----] according to a new report by Bain and that could hamper AI's progress @business (Gift link: free to read until Sept 30)"
X Link 2025-09-23T18:27Z 207.5K followers, 15.7K engagements
"BREAKING NEWS: New @betterup/Stanford study on absolutely nails part of why the ROI on GenAI has been so poor: Employees are using AI tools to create low-effort passable looking work that ends up creating more work for their coworkers. On social media which is increasingly clogged with low-quality AI-generated posts this content is often referred to as AI slop. In the context of work we refer to this phenomenon as workslop. AI generated work content that masquerades as good work but lacks the substance to meaningfully advance a given task"
X Link 2025-09-24T19:28Z 208.9K followers, 116.2K engagements
"AI is the asbestos we are shoveling into the walls of our society and our descendants will be digging it out for generations - @doctorow speaking truth and taking no prisoners"
X Link 2025-10-04T13:14Z 208.9K followers, 83.8K engagements
"The last few months have been devastating for LLM dreams: Apple reasoning paper and the ASU mirage paper and many others confirmed that LLMs still cant solve distribution shift. GPT-5 came late and fell short. Turing Award winner Rich Sutton thanked me for my critiques of LLMs and agreed. Karpathy just said agents arent anywhere close and that AGI is a decade away. And Hassabis just blew up some wildly overhyped claims from OpenAI about math. Game over man. LLMs have their place but anyone expecting the current paradigm to be close to AGI is delusional. At first I thought GPT-5 had cracked"
X Link 2025-10-18T11:18Z 208.9K followers, 363.2K engagements
"Absolutely the most important satire of the LLM era. My 3-month-old son is now TWICE as big as when he was born. He's on track to weigh [---] trillion pounds by age [--] https://t.co/S4v9gXLlrt My 3-month-old son is now TWICE as big as when he was born. He's on track to weigh [---] trillion pounds by age [--] https://t.co/S4v9gXLlrt"
X Link 2025-10-22T11:48Z 208.9K followers, 4M engagements
"losing nearly $4 billion dollars a month. Microsofts latest SEC filing quietly exposed that OpenAI lost around $11.5 billion last quarter based on Microsofts 27% ownership stake and a $3.1 billion hit to its own net income. The filings confirm that Microsoft has funded $11.6B of its $13B commitment to OpenAI with https://t.co/lMl3wptM8H Microsofts latest SEC filing quietly exposed that OpenAI lost around $11.5 billion last quarter based on Microsofts 27% ownership stake and a $3.1 billion hit to its own net income. The filings confirm that Microsoft has funded $11.6B of its $13B commitment"
X Link 2025-10-31T02:39Z 208.9K followers, 146.9K engagements
"1. OpenAI is a train wreck. [--]. They cant pay the obligations they are taking on and they know it. [--]. Guess what You the taxpayer will be left holding the bag"
X Link 2025-11-05T22:54Z 208.9K followers, 65.4K engagements
"Call your Congress people. Ten months almost to the day OpenAI and NVidia have both rolled out exactly the too big to fail . but China game plan I predicted they would be forced to. Our government is welching on food stamps but they are going to bail out the biggest public and private companies of all time with your money. Dont stand for it. The countdown until we are told that LLMs are too big to fail starts now. We cant afford to lose to China they will say accepting their multibillion dollar bailouts. The countdown until we are told that LLMs are too big to fail starts now. We cant"
X Link 2025-11-06T00:07Z 208.9K followers, 333.3K engagements
"Could this be the week We all know the bubble is going to pop; the only question is when"
X Link 2025-11-06T15:54Z 208.9K followers, 303.8K engagements
"if only @sama had a reputation for keeping to his word I would be comforted by this. but he doesnt. and I am not comforted by it not one bit. the reality is that yesterdays statement on government loan guarantees was a trial balloon that a got a ton of flak and this is paragraph after paragraph of spin doctoring to try to walk that back. I would like to clarify a few things. First the obvious one: we do not have or want government guarantees for OpenAI datacenters. We believe that governments should not pick winners or losers and that taxpayers should not bail out companies that make bad"
X Link 2025-11-06T20:40Z 208.9K followers, 119.7K engagements
"Sam is getting roasted alive. Nobody is buying his bullshit anymore. Glorious"
X Link 2025-11-06T22:04Z 208.9K followers, 34.5K engagements
"AI backlash reaches new heights"
X Link 2025-11-10T16:04Z 208.9K followers, 51K engagements
"👉Softbank sells entire Nvidia position. 👉Oracle debt downgraded. 👉Meta financing games revealed. 👉OpenAI CEO @sama couldnt explain how company would meet its $1.4 T obligations. 👉Coreweave drops 20% in a week. You do the math"
X Link 2025-11-11T18:37Z 208.9K followers, 204.6K engagements
"@vaughanbell I don't think you've characterized my views correctly; I posted a comment on your blog"
X Link 2014-07-26T19:25Z 207.5K followers, [--] engagements
"@TomChivers @me_datapoint its not a logical argument to say that is because this system follows tricks we must be following similar tricks. my five year old knows that you cant stick a knight on a square occupied by a bishop; the slate star chess bot doesnt. the bot is cute but it teaches us little"
X Link 2020-01-08T15:55Z 190.8K followers, [--] engagements
"@_RobToews thanks (a link to a historical perspective from my gpt-2 piece might be nice since so little has really changed.)"
X Link 2020-07-20T01:02Z 190.9K followers, [--] engagements
"no much fishier @OpenAIs parent organization (by charter allegedly providing oversight) is still as of 2020 filing as a nonprofit. they are insulating their now-closed business from open investigation whilst still shielding millions from taxes. 🤔 https://projects.propublica.org/nonprofits/organizations/810861541 Related: @OpenAI is no longer non-profit and the source of #GPT3 is closed. https://projects.propublica.org/nonprofits/organizations/810861541 Related: @OpenAI is no longer non-profit and the source of #GPT3 is closed"
X Link 2020-08-28T00:38Z 162.9K followers, [--] engagements
"@_RobToews its just a bunch of anecdotal examples at the moment; theres no statistical analysis and thus far no paper. it is certainly not reliable"
X Link 2021-01-21T03:23Z 191K followers, [--] engagements
"Oh please "The findings hint at a future in which artificial intelligence can help humans to formulate and make sense of complex arguments." This is absurd. Neither the language understanding nor the reasoning skills in current AI are remotely close. https://www.nature.com/articles/d41586-021-00539-5 https://www.nature.com/articles/d41586-021-00539-5"
X Link 2021-03-19T17:01Z 191.1K followers, [---] engagements
"Let us invent then a new breed of AI systems that mix an awareness of the past with values that represent the future that we aspire to. Our focus should be on figuring on how to build AI that can represent and reason about values rather than simply perpetuating past data"
X Link 2021-04-19T15:54Z 162.5K followers, [----] engagements
"@Raza_Habib496 show one place that anybody said that out loud"
X Link 2022-03-10T13:37Z 194.6K followers, [--] engagements
"You cant go to a physics conference and say: Ive got a great theory. It accounts for everything and is so simple it can be captured in two words: Anything goes." - Noam Chomsky in email re GPT-3"
X Link 2022-05-15T21:38Z 192.6K followers, [---] engagements
"@ylecun @bleepbeepbzzz @TonyZador @raphaelmilliere @davidchalmers42 @MetaAI @noema @Jake_Browning00 @NoemaMag @luislamb @AvilaGarcez hmm. whats an emergent symbol is it something that stands for something else (as classical symbols are) how would you know if you had one in what sense would it be a symbol could you do eg deductive inference with it"
X Link 2022-09-27T00:06Z 191.1K followers, [--] engagements
"@CriticalAI @EnglishOER @markcmarino you might be referring to something someone sent to me about Noam Chomsky (alive and well as of an email he sent me a couple days ago) allegedly dying in [----] generated by ChatGPTs sibling davinci complete with fake URLs. At least it got Hillary Clintons status correct 🙄"
X Link 2023-01-29T16:23Z 162.9K followers, [---] engagements
"dear hivemind - is there a term for a bunch of fake websites that mutually reinforce each other in order to increase their search ranking - are there any studies or articles about this @noUpside @ProfNoahGian @DG_Rand @sinanaral @erikbryn @chafkin @lorakolodny"
X Link 2023-02-12T15:00Z 190.9K followers, 17.9K engagements
"@elonmusk @geoffreyhinton 🙏 If you really want to fight this we should talk"
X Link 2023-03-27T23:23Z 187.3K followers, 20.2K engagements
"Opportunity Lost. The letter was flawed; so was the communitys reaction. Why there is more to life than The Letter what we should learn from the experience and what we need to do to carry on. https://garymarcus.substack.com/p/i-am-not-afraid-of-robots-i-am-afraidutm_source=twitter&sd=pf https://garymarcus.substack.com/p/i-am-not-afraid-of-robots-i-am-afraidutm_source=twitter&sd=pf"
X Link 2023-04-02T17:36Z 173.2K followers, 66.2K engagements
"Could a computer program watch any arbitrary TV program or Youtube video and reliably answer questions about its content I have been asking this question since [----]. So far the answer is still no. From [----] What Comes After the #TuringTest New test for 21st century: ".build a computer program that can watch any arbitrary TV program or YouTube video and answer questions about its content." @GaryMarcus https://t.co/dTNH0gh5yH #ArtificialIntelligence From [----] What Comes After the #TuringTest New test for 21st century: ".build a computer program that can watch any arbitrary TV program or YouTube"
X Link 2023-05-04T12:44Z 191.1K followers, 27.8K engagements
"Could I get some actual neuroscientists to weigh in on current thinking around the triune (Lizard) brain theory Am I correct in saying it has been universally rejected Whats the reasoning @WiringTheBrain @TonyZador @anne_churchland @Nancy_Kanwisher @PessoaBrain @neurograce Lizard brain Always good to see Big Techs most illustrious AI researchers riffing on 1960s neuroscience. 🙄 Lizard brain Always good to see Big Techs most illustrious AI researchers riffing on 1960s neuroscience. 🙄"
X Link 2023-05-04T15:30Z 172.3K followers, 127.7K engagements
"best use of deep voice fakes I have seen A.I. David Attenborough on #memecoin trading. https://t.co/okiEemDjDJ A.I. David Attenborough on #memecoin trading. https://t.co/okiEemDjDJ"
X Link 2023-05-23T11:32Z 134.8K followers, 18.7K engagements
"Remarkable consensus @UNTechEnvoy #GlobalDigitalCompact deep dive on #AI: Urgency of the situation Need to bridge the global divide Need for transparency accountablility fairness & protection of privacy Need for global coordination in governance of AI"
X Link 2023-06-29T20:56Z 134.8K followers, 26.1K engagements
"Good reminder that vintage [----] neural networks arent very neural. Fly brain has 3000+ distinct neuron types all but entirely abstracted away from in current neural networks. Fair bit of evidence for innate constraints too contra the AI/Ml zeitgeist"
X Link 2023-07-03T01:34Z 134.8K followers, [---] engagements
"One in five people surveyed think we will solve machine intelligence before we can figure out how a worm brain works. I doubt we will solve either this decade; both problems are harder than they initially appeared"
X Link 2023-07-10T21:55Z 134.9K followers, 16.7K engagements
"Clever and interesting paper confirming popular 1960s hypothesis of P. Simon and A. Garfunkel"
X Link 2023-07-12T21:34Z 134.9K followers, 21.7K engagements
"overwrought prose alert Proposals for stringent AI model licensing and surveillance will likely be ineffective concentrating power potentially rolling back the societal gains of the Enlightenment I guess drivers licenses and seatbelt laws rewound 1/2 an Enlightenment"
X Link 2023-07-11T20:44Z 134.8K followers, 10.8K engagements
"Welcome to Dystopia Womans photo of son rejected from Sydney competition after judges ruled it could be AI https://t.co/Fc1OzYtC3z Womans photo of son rejected from Sydney competition after judges ruled it could be AI https://t.co/Fc1OzYtC3z"
X Link 2023-07-12T05:42Z 162.9K followers, 13K engagements
"@bpositive718 @ChristianGlawe As I have said multiple times"
X Link 2023-07-12T08:41Z 134.8K followers, [--] engagements
"View from dinner last night near Aspen"
X Link 2023-07-24T22:42Z 134.9K followers, [----] engagements
"Thread on #lk99 worth watching"
X Link 2023-08-01T19:58Z 134.9K followers, 12.3K engagements
"Not looking great Superconductivity research scholars in Korea have officially come together under a "replication committee" to investigate the claims of LK-99. Here is their initial assessment: https://t.co/M0pADFijsy Superconductivity research scholars in Korea have officially come together under a "replication committee" to investigate the claims of LK-99. Here is their initial assessment: https://t.co/M0pADFijsy"
X Link 2023-08-02T17:44Z 163K followers, 19.1K engagements
"Why do we need regulation for AI Because quite simply current models cant be trusted to be safe. Test of [--] popular models each one turns to be problematic. Voluntary agreements without real teeth are not enough"
X Link 2023-08-04T03:11Z 134.9K followers, 16.6K engagements
"I fully share @samas concerns"
X Link 2023-08-04T12:23Z 134.8K followers, 16.6K engagements
"#LK99 humor"
X Link 2023-08-06T13:52Z 129.7K followers, 15K engagements
"LLM s really are a long way off from being reliable we will soon rely on them more than the best experts Cant be both"
X Link 2023-08-06T15:21Z 134.8K followers, 29.1K engagements
"First major AI-powered mess"
X Link 2023-08-07T17:36Z 131K followers, 37.2K engagements
"So I guess we shouldnt regulate cars airplanes food etc because no issue has ever lead to a species-level extinction event then Close down the FDA and FAA etc cause no foom"
X Link 2023-08-07T19:03Z 134.8K followers, 21.5K engagements
"@_ahmedzaidi Thats the ground I have been arguing for for years"
X Link 2023-08-08T00:08Z 134.8K followers, [---] engagements
"At this point anyone who has played around with GPT-4 realizes that its reasoning is unstable & highly context-dependent; very different from the arithmetic of a calculator. Yet people who note these instabilities are routinely insulted & likened to climate change deniers"
X Link 2023-08-08T19:14Z 134.8K followers, 45.8K engagements
"because it is having an outsized impact on the world and on AI research and because i think we can and should do much better"
X Link 2023-08-09T15:21Z 134.9K followers, 10K engagements
"This clip with @janleike @openAi is great. So much I agree about - the many paths difficulty in integrating them importance of exerting causal influence on the p but why work for OpenAI if you thought p(doom) was in double digits and they were accelerating whatever risks there may be"
X Link 2023-08-09T17:14Z 121.9K followers, 19.1K engagements
"The UN is putting together a high level committee on AI and looking for nominations. Do the world a solid and help them out"
X Link 2023-08-09T18:56Z 134.9K followers, 39.1K engagements
"@ arXiv: bridging the gap between LLMs and classical AI with Doug Lenat"
X Link 2023-08-10T00:31Z 134.8K followers, 27.9K engagements
"A tweet below allegedly defending GPTs honor is a perfect encapsulation of what is wrong with the culture of AI today. Long post on a ubiquitous problem. 👉 [--] examples in a stochastic system with a massive undisclosed database proves nothing. 👉 Evidence of success is triumphantly reported; errors are neither looked for nor discussed (but we all know they exist). 👉 The fact that RLHF and possibly other undisclosed components of GPT-4 appear to be regularly updated is not discussed. 👉 The result depends on custom prompt chosen by a human that happens to be relevant for this task but which"
X Link 2023-08-10T13:51Z 134.8K followers, 261.5K engagements
"@robertskmiles we also discussed this sensitivity in Rebooting AI iirc and its another symptom of shallow comprehension rather than deep understanding"
X Link 2023-08-10T14:54Z 130.9K followers, [---] engagements
"@josecyc The last time I did this I am pretty sure OpenAI just trained on the test set"
X Link 2023-08-10T16:49Z 134.9K followers, [----] engagements
"You know what is funny about this Gallup poll Everyone treats @KamalaHarris as uniquely unpopular and yet statistically she is more or less in the middle of this pack"
X Link 2023-08-10T19:18Z 160.3K followers, [----] engagements
"@pwlot @jeremyphoward @JaffrayW Literally [--] minutes ago from a respected author visiting openAI:"
X Link 2023-08-11T01:53Z 134.8K followers, [----] engagements
"Except on X"
X Link 2023-08-11T01:54Z 131.2K followers, [----] engagements
"Is there a name for this Our examination revealed that 52% of ChatGPT's answers contain inaccuracies and 77% are verbose. Nevertheless users still prefer ChatGPT's responses 39% of the time due to their comprehensiveness and articulate language style. https://arxiv.org/abs/2308.02312 https://arxiv.org/abs/2308.02312"
X Link 2023-08-11T04:08Z 145.2K followers, 111.2K engagements
"No. The urgency for regulation comes from unreliability and uncontrollability of current AI combined with widespread adoption including by bad actors not from fantasies (that I never endorsed) that current AI was somehow near to AGI"
X Link 2023-08-12T00:03Z 134.8K followers, 17K engagements
"No doubt generative AI is great for coders but will other people maintain their monthly subscriptions after the novelty wears off Or are we building our entire approach to the future around an overhyped fad"
X Link 2023-08-13T00:37Z 122.4K followers, 42.2K engagements
"What if generative AI turned out to be a dud At my (free) newsletter that I cant properly link here I discuss some economic and geopolitical implications of this that Xs algorithm doesnt seem to want you to read about"
X Link 2023-08-13T02:43Z 134.8K followers, 57.3K engagements
Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing