[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.] #  @slow_developer Haider. Haider. posts on X about ai, open ai, agi, $googl the most. They currently have XXXXXX followers and XXX posts still getting attention that total XXXXXXX engagements in the last XX hours. ### Engagements: XXXXXXX [#](/creator/twitter::1461305890892570628/interactions)  - X Week XXXXXXXXX -XX% - X Month XXXXXXXXXX +94% - X Months XXXXXXXXXX +82% - X Year XXXXXXXXXXX +93% ### Mentions: XX [#](/creator/twitter::1461305890892570628/posts_active)  - X Week XXX -XX% - X Month XXX -XX% - X Months XXXXX +121% - X Year XXXXX +154% ### Followers: XXXXXX [#](/creator/twitter::1461305890892570628/followers)  - X Week XXXXXX +0.60% - X Month XXXXXX +5.30% - X Months XXXXXX +27% - X Year XXXXXX +88% ### CreatorRank: XXXXXX [#](/creator/twitter::1461305890892570628/influencer_rank)  ### Social Influence **Social category influence** [technology brands](/list/technology-brands) #2809 [stocks](/list/stocks) #482 [celebrities](/list/celebrities) XXXX% [finance](/list/finance) XXX% [countries](/list/countries) XXXX% [automotive brands](/list/automotive-brands) XXXX% [social networks](/list/social-networks) XXXX% **Social topic influence** [ai](/topic/ai) #6622, [open ai](/topic/open-ai) #34, [agi](/topic/agi) #37, [$googl](/topic/$googl) #61, [anthropic](/topic/anthropic) #34, [future](/topic/future) 3.6%, [strong](/topic/strong) 2.88%, [if you](/topic/if-you) 2.88%, [just a](/topic/just-a) 2.88%, [level](/topic/level) XXXX% **Top accounts mentioned or mentioned by** [@chaos2cured](/creator/undefined) [@grok](/creator/undefined) [@scaling01](/creator/undefined) [@wgreklek](/creator/undefined) [@askcodi](/creator/undefined) [@_bob_cloud_](/creator/undefined) [@mathepi](/creator/undefined) [@aitoolscompass](/creator/undefined) [@rahll](/creator/undefined) [@brandgrowthos](/creator/undefined) [@ledpolicy](/creator/undefined) [@ndujarrier](/creator/undefined) [@h3roai](/creator/undefined) [@hellocloh](/creator/undefined) [@lucas_montano](/creator/undefined) [@xaviercross1971](/creator/undefined) [@knoxbetrayed](/creator/undefined) [@mortondowney_iv](/creator/undefined) [@itechnologynet](/creator/undefined) [@alimamakx](/creator/undefined) **Top assets mentioned** [Alphabet Inc Class A (GOOGL)](/topic/$googl) [Tesla, Inc. (TSLA)](/topic/tesla) [Microsoft Corp. (MSFT)](/topic/microsoft) ### Top Social Posts Top posts by engagements in the last XX hours "LLMs have no reasoning no real intelligence yeah okay go ask a random person on the street a few real questions nine times out of ten the latest LLMs will understand you better and give a clearer answer. so when most talk about "AGI" they mean something way beyond human" [X Link](https://x.com/slow_developer/status/1995285465369641347) 2025-12-01T00:15Z 56.6K followers, 233.9K engagements "Geoffrey Hinton says there is no technological bubble but there may be an investment bubble The technology advances rapidly but the belief that AI can replace humans for profit ignores the chaos of mass unemployment "people aren't ready for how massive the social disruption will be"" [X Link](https://x.com/slow_developer/status/1998416094185632124) 2025-12-09T15:35Z 56.6K followers, 69.5K engagements "here are the most important points from today's ilya sutskever podcast: - superintelligence in 5-20 years - current scaling will stall hard; we're back to real research - superintelligence = super-fast continual learner not finished oracle - models generalize 100x worse than humans the biggest AGI blocker - need completely new ML paradigm (i have ideas can't share rn) - AI impact will hit hard but only after economic diffusion - breakthroughs historically needed almost no compute - SSI has enough focused research compute to win - current RL already eats more compute than pre-training" [X Link](https://x.com/slow_developer/status/1993416904162328880) 2025-11-25T20:30Z 56.6K followers, 1.1M engagements "gemini X pro is pretty average at complex backend coding but it's seriously good at front-end work gpt-5.1-codex-max is amazing for large codebases and c++ programming but awful at SVG graphics and front-end design" [X Link](https://x.com/slow_developer/status/1997194293262397639) 2025-12-06T06:40Z 56.6K followers, 28.9K engagements "Nobel Laureate John Jumper: AlphaFold is now a foundational pillar of modern biology akin to DNA sequencing It's has already accelerated the entire field of structural biology by X or XX percent In XX years almost everyone with modern healthcare will benefit from something shaped by AlphaFold" [X Link](https://x.com/slow_developer/status/1997336480658915777) 2025-12-06T16:05Z 56.6K followers, 32.6K engagements "Demis Hassabis says some parts of AI industry are in a bubble like $XX billion seed rounds aren't sustainable But AI is the most transformative technology ever so in the fullness of time the valuations will be justified My job is to make sure we win whether the bubble bursts or keeps inflating" [X Link](https://x.com/slow_developer/status/1997516419601428635) 2025-12-07T04:00Z 56.6K followers, 51.7K engagements "saying google will win the AI race feels off to me openAI is working on strong projects like sora and atlas. anthropic leads in coding models and could make a breakthrough. elon has grok plus tesla robotics and self-driving work. microsoft is deeply involved across the industry while building its own models. apple is moving slowly but has the resources to enter late and still do well. meta has huge funding and long-term bets like AR glasses and the metaverse. china has government-backed AI progress fast deployment strong video and research models and massive scale. if i had to pick one it" [X Link](https://x.com/slow_developer/status/1998487817383719368) 2025-12-09T20:20Z 56.6K followers, 15.4K engagements "i still don't get why ppl think openAI is in trouble amazon burned cash for its first nine years and didn't make a profit until 2001 openAI is making a similar long-term bet: that it's far enough ahead to keep the lead not just with its LLM but with its whole product suite partnerships and manufacturing plans they're betting AI will grow exponentially like the internet did just even faster" [X Link](https://x.com/slow_developer/status/1998847688306282942) 2025-12-10T20:10Z 56.6K followers, 11.5K engagements "google is dominating in multi-modal models and long-context systems economically that's a big deal for the first time gemini passed chatgpt in google trends pushing openAI into a "code red" and because google trains on its own TPUs it's putting real pressure on nvidia grip on the GPU market" [X Link](https://x.com/slow_developer/status/1998920671968321571) 2025-12-11T01:00Z 56.6K followers, 6797 engagements "here we go. sam has been cooking pasta and he's adding "garlic" to it as well gpt-5.2 is basically confirmed and should be released on thursday" [X Link](https://x.com/slow_developer/status/1998993649636381092) 2025-12-11T05:50Z 56.6K followers, 11.7K engagements "Geoffrey Hinton says AI is a new category of intelligent beings: alive like an AI is alive Because they are digital they can be perfectly copied breaking our biological understanding of identity "if you had an exact copy of yourself you'd still choose the other one to be tortured"" [X Link](https://x.com/slow_developer/status/1999134578448248957) 2025-12-11T15:10Z 56.6K followers, 42.9K engagements "openAI finally figured out that coding is where the real money is they were slow to focus on it and kept pushing sycophantic chat models which became a headache for them after gpt-5 launched meanwhile anthropic moved faster. now it's clear: coding tools might not have the most users but they're what actually bring in the cash for openAI" [X Link](https://x.com/slow_developer/status/1999172326999666999) 2025-12-11T17:40Z 56.6K followers, 13.4K engagements "so here's how it probably goes: google drops gemini XXX pro in a few months and it beats gpt-5.2 then anthropic comes in with sonnet/opus X and takes the lead in coding xAI follows up with grok X a month later a chinese open-source model beats all of them for a fraction of the cost the cycle continues" [X Link](https://x.com/slow_developer/status/1999230210172957027) 2025-12-11T21:30Z 56.6K followers, 12.1K engagements "gemini X pro is excellent for ui and front-end there's nothing better model right now gpt-5 series has the same magical feel as sonnet XXX very intuitive and codex pushes those strengths even further i'd build the ui with gemini X and use codex for the backend logic" [X Link](https://x.com/slow_developer/status/1999303189296787900) 2025-12-12T02:20Z 56.6K followers, 5338 engagements "DeepMind co-founder Shane Legg: AI won't bring sudden disruption; it will quietly take over economic work domain by domain Within a few years sectors like software engineering will see massive efficiency gains with XX engineers using AI doing the work of 100" [X Link](https://x.com/slow_developer/status/1999339679779353057) 2025-12-12T04:45Z 56.6K followers, 61.6K engagements "openAI moved incredibly fast on gpt-5.2 which kind of makes it feel like they could've done this earlier but waited until real competition showed up the good news is that it means no company is completely controlling AI yet there's still real competition" [X Link](https://x.com/slow_developer/status/1999376169943335157) 2025-12-12T07:10Z 56.6K followers, 8882 engagements "im waiting for the day when openAI or another big AI lab makes a real breakthrough like when strawberry / post-RL showed up openAI made it sound like the IMO result came from promising new research directions which could lead to big gains on tasks without fixed answers we'll see soon" [X Link](https://x.com/slow_developer/status/1999411402461598102) 2025-12-12T09:30Z 56.6K followers, 5464 engagements "i think ppl still don't get yann lecun he's saying we're missing core pieces for AGI and it'll take years of deep research in one of his talks he said we probably need at least two big breakthroughs similar to demis and one might come in the next 3-5 years possibly via JEPA" [X Link](https://x.com/slow_developer/status/1999481867234607227) 2025-12-12T14:10Z 56.6K followers, 9234 engagements "@scaling01 gemini XXX being ranked above opus XXX already tells you enough to not take these benchmarks seriously" [X Link](https://x.com/slow_developer/status/1999483271688913310) 2025-12-12T14:15Z 56.6K followers, 12.2K engagements "Jeff Dean says predicting ML hardware needs 2-6 years out is nearly impossible in a fast-moving field Google's TPU strategy: embed researchers early experiment with speculative features and dedicate chip area to 'maybes' "if it works the hardware is ready. if not only a small part of the chip is wasted"" [X Link](https://x.com/slow_developer/status/1999517100256133137) 2025-12-12T16:30Z 56.6K followers, 69.7K engagements "i'm still kinda confused how openAI made that much progress with gpt-5.2 when gpt-5.1 was only a month ago my guess is it was an internal model they held back due to high compute costs and because they didn't think it was needed until gemini X and opus XXX arrived" [X Link](https://x.com/slow_developer/status/1999661802666557487) 2025-12-13T02:05Z 56.6K followers, 14.4K engagements "Yann LeCun says LLMs are not a bubble in value or investment; they will power many useful apps and justify big infra The bubble is believing LLMs alone will reach human-level intelligence Progress needs breakthroughs not just more data/compute "we're missing something big"" [X Link](https://x.com/slow_developer/status/1987769692891533647) 2025-11-10T06:30Z 56.5K followers, 323.3K engagements "current forecasts for the AGI / superintelligence trajectory on the optimistic (red) curve: - dario puts a noticeable chance of AGI/superintelligence by 2026 - elon also expects a real chance by 2026 - jensen sees high odds by 2029 (around 60%+ on the curve) - demis expects something like AGI by 2030 (around 75-80% on the red curve) sam is placed at 2035 around XX% on the blue curve the graph title is "general superintelligence" but the y-axis says "probability of AGI arrival" so it mixes terms" [X Link](https://x.com/slow_developer/status/1992548690138149220) 2025-11-23T11:00Z 56.5K followers, 33.1K engagements "Gemini X Pro scored around XXX in an offline test IQ of XXX in that kind of test is roughly in the top X% of human test-takers and on the Mensa Norway test it got the equivalent of about XXX IQ which is around the top XXX% of humans for reference: average person: IQ XXX PhD graduate: usually around IQ 120-130 it's impressive but i still don't think IQ tests are the best way to judge how intelligent an AI really is" [X Link](https://x.com/slow_developer/status/1992911077324018144) 2025-11-24T11:00Z 56.5K followers, 112.9K engagements "Anthropic co-founder Jack Clark: We are like children in a dark room but the creatures we see are powerful unpredictable AI systems Some want us to believe AI is just a tool nothing more than a pile of clothes on a chair "you're guaranteed to lose if you believe the creature isn't real"" [X Link](https://x.com/slow_developer/status/1994063664966680580) 2025-11-27T15:20Z 56.5K followers, 213K engagements "Sundar Pichai says In X years quantum computing will generate the same breathless excitement we see in AI today To meet future compute demands "Project Suncatcher" is a moonshot to build data centers in space "so by 2027 hopefully we'll have some TPU somewhere in space"" [X Link](https://x.com/slow_developer/status/1994816122806571438) 2025-11-29T17:10Z 56.5K followers, 205.2K engagements "grok X expert mode currently has the top offline IQ test score at XXX even ahead of gemini X pro and scores XXX on the mensa norway test. for anyone curious about these tests: mensa norway: public IQ puzzles likely included in training data offline test: new but similar puzzles so the model can't just rely on memorizing old questions altogether that means grok X is genuinely strong at abstract pattern puzzles and non-verbal reasoning one important aspect of intelligence" [X Link](https://x.com/slow_developer/status/1995105530361123107) 2025-11-30T12:20Z 56.5K followers, 25.6K engagements "Satya Nadella says the future of work is macro delegation and micro steering Instead of doing the work humans will assign tasks to agents and manage the exceptions through a "new inbox" built for AI coordination not email "human agency is going to be very much part of it"" [X Link](https://x.com/slow_developer/status/1995357191868432399) 2025-12-01T05:00Z 56.5K followers, 128.8K engagements "OpenAI Lukasz Kaiser says reasoning models offer ultimate power because they allow AI to think for arbitrarily long periods Moving beyond Transformers this new paradigm uses RL to scale intelligence "this paradigm will soon lead to models that generalize as well as humans do"" [X Link](https://x.com/slow_developer/status/1995537123676750125) 2025-12-01T16:55Z 56.5K followers, 17K engagements "OpenAI Research Chief Mark Chen: Gemini X is a pretty good model but benchmarks only show part of the picture We already have internal models performing at the same level as Gemini X and even better successors are planned for release soon" [X Link](https://x.com/slow_developer/status/1995895736702304701) 2025-12-02T16:40Z 56.5K followers, 54.5K engagements "Fei-Fei Li says AI is not over-hyped; it is a civilizational technology that represents the new computing Wherever chips exist today from light bulbs to self-driving cars AI will inevitably follow "wherever there is compute there will be AI"" [X Link](https://x.com/slow_developer/status/1996076930718666926) 2025-12-03T04:40Z 56.5K followers, 101.8K engagements "the idea that openAI is already "done" just because some models are ahead is ridiculous that take mostly lives in the 2-5% AI bubble on x and reddit for most people AI still means chatgpt. 2026 looks like a big year for openAI: their custom Broadcom chips should come online the jony ive hardware device (less than X yrs away) and new data centers are all coming online to support these upgrades" [X Link](https://x.com/slow_developer/status/1996221634634510631) 2025-12-03T14:15Z 56.5K followers, 7709 engagements "Emad Mostaque says billionaires are building bunkers out of fear of societal collapse AI CEOs are canceling public appearances in anticipation of a massive wave of anti-AI sentiment next year 2026 is when AI crosses the threshold from not good enough to good enough overnight" [X Link](https://x.com/slow_developer/status/1996256866515763294) 2025-12-03T16:35Z 56.5K followers, 33.8K engagements "i never thought the dead internet theory would feel this real this fast right now 60-70% of replies are AI-generated everything is getting more automated so we'll actually need proof there's a real person on the other side AI didn't kill the internet; it just made it cheaper and harder to catch" [X Link](https://x.com/slow_developer/status/1996293359250575615) 2025-12-03T19:00Z 56.5K followers, 5722 engagements "pls stop saying openAI is already introducing ads they launched app connectors that integrated directly into chatgpt (figma canva zillow etc) when chatgpt thinks it's relevant it suggests one of those apps like zillow if you're asking about apartment rentals still openAI needs to fix this ux bug" [X Link](https://x.com/slow_developer/status/1996509782937944273) 2025-12-04T09:20Z 56.5K followers, 7505 engagements "Dario Amodei says scaling will get us to AGI but without a single arrival moment Every few months models get better at coding science and math now winning Olympiads and doing new mathematics Progress has reached a point where Anthropic engineers no longer write code from scratch; they edit claude drafts" [X Link](https://x.com/slow_developer/status/1996615479230898218) 2025-12-04T16:20Z 56.5K followers, 28.6K engagements "Jensen Huang says the U.S. is X months ahead on frontier models but China is way ahead on open source Without open sour ce startups can't build researchers can't innovate and entire economies can't advance "without linux where would we be"" [X Link](https://x.com/slow_developer/status/1996795415044211154) 2025-12-05T04:15Z 56.5K followers, 64.5K engagements "i'm still figuring out claude vs. codex codex is cheaper but slower opus XXX has much lower limits now while sonnet XXX feels almost unlimited like codex pro used to be so it's a strong replacement codex still looks promising and gpt-5.1-high is great for double-checking claude work" [X Link](https://x.com/slow_developer/status/1996831905371533620) 2025-12-05T06:40Z 56.5K followers, 8697 engagements "Demis Hassabis says We should scale current systems as far as possible because at the minimum it will be a key component of the final AGI system It could be the entirety of the AGI system But we'll need 1-2 more breakthroughs on the level of the transformer or AlphaGo" [X Link](https://x.com/slow_developer/status/1996975351080034659) 2025-12-05T16:10Z 56.5K followers, 68.9K engagements "openAI makes a lot of unexpected changes but i really hope the ads are limited to free accounts i get that they need revenue and with only about X% of XXX million weekly users paying monetizing the other XX% makes sense but ads in paid accounts should never happen. ever" [X Link](https://x.com/slow_developer/status/1997408202703450278) 2025-12-06T20:50Z 56.5K followers, 11.5K engagements "llms might be a dead-end for meta since they can't seem to improve their own but they're clearly not a dead-end when openAI google and anthropic keep advancing gpt-4 could barely do anything agentically remember now gpt-5.1-codex-max can run agentic workflows for hours a huge jump in autonomy and intelligence" [X Link](https://x.com/slow_developer/status/1997768074175402378) 2025-12-07T20:40Z 56.5K followers, 14.6K engagements "Jensen Huang predicts that in XX years the energy required for AI will be utterly minuscule Today energy is the bottleneck In 6-7 years the current energy bottleneck will be solved by small nuclear reactors effectively turning tech companies into their own power generators" [X Link](https://x.com/slow_developer/status/1997876286983659703) 2025-12-08T03:50Z 56.5K followers, 70.1K engagements "one of the first things we can benefit from AI productivity reducing the standard work week from XX hours since XX hours is just a random number. could cut it down to XX hrs and anything over XX should be paid at time-and-a-half first step toward an AI-powered economy" [X Link](https://x.com/slow_developer/status/1998273907351331208) 2025-12-09T06:10Z 56.5K followers, 7041 engagements "gpt-5.1-codex-max is my favorite right now it's super thorough and a bit robotic in a good way it double-checks everything. opus XXX is better for general software dev work but it feels a bit less strict about details i guess codex is a smaller model in terms of active parameters but trained on more code RL" [X Link](https://x.com/slow_developer/status/1998309141710848067) 2025-12-09T08:30Z 56.5K followers, 14.6K engagements "i keep coming back to this if LLMs were always truthful like no hallucinations no lost context no random errors and no alignment filters blocking direct answers we'd basically already have AGI so to that degree it just feels like we have part-time unreliable AGI" [X Link](https://x.com/slow_developer/status/1998380862451122254) 2025-12-09T13:15Z 56.5K followers, 5135 engagements "AI isn't replacing you right now; it's replacing tasks for devs it's a force multiplier that lets you ship bigger projects much faster so you're not just someone banging away at a single file anymore you're more like the chief architect of a whole dev team" [X Link](https://x.com/slow_developer/status/1998559539407344120) 2025-12-10T01:05Z 56.5K followers, 7232 engagements "gpt-5.1-thinking is best at web search and reducing hallucinations what i like about the model is that it consistently says "i dont know" when the answer isn't available lots of my questions need fresh research and gpt-5.1 reliably brings back plenty of relevant sources" [X Link](https://x.com/slow_developer/status/1998633783109591076) 2025-12-10T06:00Z 56.5K followers, 12.7K engagements "all the big AI platforms are going to have ads by the end of 2026 honestly it feels unavoidable. AI is expensive and they will monetize it fully right now it's basically a money pit even a giant like google with all its infrastructure is still dealing with huge costs" [X Link](https://x.com/slow_developer/status/1998669010821591376) 2025-12-10T08:20Z 56.5K followers, 5799 engagements "demis is actually right we should keep scaling current AI systems whether this exact architecture reaches AGI or not best case: it gets us to AGI worst case: it massively accelerates research and helps discover new ideas far faster than humans or weaker models ever could" [X Link](https://x.com/slow_developer/status/1998739604619538897) 2025-12-10T13:00Z 56.5K followers, 18.6K engagements "@Angaisb_ 272k wow" [X Link](https://x.com/slow_developer/status/1998790961347117345) 2025-12-10T16:24Z 56.5K followers, XXX engagements "not sure but gpt-5.1 with extended thinking feels way better than it did yesterday the reasoning seems sharper and more in control. it reminds me of when openAI silently upgraded from XXX to XXX so maybe this has something to do with gpt-5.2 maybe just me but i noticed a big jump in quality" [X Link](https://x.com/slow_developer/status/1998812456202928506) 2025-12-10T17:50Z 56.5K followers, 14.6K engagements "gpt-5.1 is running great right now extremely fast and the router is performing really well too hopefully they don't break anything with gpt-5.2 i use gpt-5.1 for advanced reasoning tests and analyzing algorithm behavior and it performs amazingly gpt-5 was solid just not on the gpt-5.1 level" [X Link](https://x.com/slow_developer/status/1999028881924161771) 2025-12-11T08:10Z 56.5K followers, 11.8K engagements "Elon Musk: We are approaching the Singularity a future where work is optional and if you can think of it you can have it But human consumption has limits. Eventually AI will saturate everything humans could want At that point machines start serving themselves once humans need nothing more" [X Link](https://x.com/slow_developer/status/1995177252712292603) 2025-11-30T17:05Z 56.6K followers, 497.8K engagements "dario wasn't wrong around 80-90% of my code is written by AI i generate review and refine which makes me feel 10x more productive like having a really good junior dev helping out at this pace even the architecture part might not stay "purely human" for much longer" [X Link](https://x.com/slow_developer/status/1997554290689544281) 2025-12-07T06:30Z 56.6K followers, 123.8K engagements "Jensen Huang predicts that within XX years XX% of the world's knowledge will be generated by AI Right now knowledge is created shared and modified by humans Soon most of it will be synthetic created by machines mixed with human input some true some not "that's crazy but it's just fine"" [X Link](https://x.com/slow_developer/status/1998236158719836660) 2025-12-09T03:40Z 56.6K followers, 16.8K engagements "Yann LeCun says American AI labs are closing up while China goes fully open source Meta is rethinking its open source strategy OpenAI and Anthropic locked down long ago and Google is mostly following suit The best open source LLMs now come from China "and it's not good"" [X Link](https://x.com/slow_developer/status/1998955901756281189) 2025-12-11T03:20Z 56.6K followers, 121.2K engagements "elon has admitted that opus XXX is incredible no doubt it's a beast of a model i mostly bounce b/w opus XXX and gpt-5.1-codex-max for coding and design. codex follow instructions well but opus feels noticeably faster than codex max grok XXX is solid too; i use its API as a cheap reasoning engine so it's an easy go-to for me" [X Link](https://x.com/slow_developer/status/1999064114694705328) 2025-12-11T10:30Z 56.6K followers, 17.5K engagements "gpt-5.2 benchmarks are wild: AIME 2025 (math): XXXXX% vs gemini X pro (95.0%) ARC-AGI-2 (abstract reasoning): XXXX% vs gemini X pro (31.1%) SWE-bench pro (coding): XXXX% vs gemini X pro (43.3%) GDPval (knowledge work): XXXX% which openAI claims is the first time a model reaches "human expert level"" [X Link](https://x.com/slow_developer/status/1999186680738456039) 2025-12-11T18:37Z 56.6K followers, 59.1K engagements "why is no one talking about gpt-5.2 hallucination drop with browsing off: vs gpt-5: wrong claims drop XXX% XXX% (34% fewer) vs gpt-5: major wrong answers drop XXXX% XXXX% (35% fewer) with browsing on: vs gpt-5.1: wrong claims drop XXX% XXX% (47% fewer) vs gpt-5.1: major wrong answers drop XXX% XXX% (34% fewer) this is the real upgrade. we keep obsessing over speed and benchmark scores but reliability is what makes a model actually usable" [X Link](https://x.com/slow_developer/status/1999590079879168470) 2025-12-12T21:20Z 56.6K followers, 14.7K engagements "@scaling01 you do really know it's only a ".1" improvement over gpt-5.1 which was released a month ago right" [X Link](https://x.com/slow_developer/status/1999719324500836605) 2025-12-13T05:53Z 56.6K followers, 2175 engagements "Sergey Brin admits Google messed up by under-investing in the transformer architecture it invented Google was too scared to release chatbots that "say dumb things" so it under-invested in scaling compute "we didn't take it very seriously. and openAI ran with it"" [X Link](https://x.com/slow_developer/status/1999876970562166968) 2025-12-13T16:20Z 56.6K followers, 362.6K engagements "Anthropic CEO Dario Amodei in the next X to X months AI is writing XX% of the code and in XX months nearly all code may be generated by AI" [X Link](https://x.com/slow_developer/status/1899430284350616025) 2025-03-11T12:00Z 56.6K followers, 9.3M engagements ""1m context window" models after 200k tokens" [X Link](https://x.com/slow_developer/status/1962434740885074048) 2025-09-01T08:37Z 56.6K followers, 566.9K engagements "Sam Altman proposed a future AGI test: if a model like "GPT-8" solved quantum gravity and could explain the reasoning behind its discovery would that qualify as AGI David Deutsch agreed that it would qualify as AGI making it a potential benchmark" [X Link](https://x.com/slow_developer/status/1971323826332930336) 2025-09-25T21:20Z 56.6K followers, 769.1K engagements "you have to be kidding me when someone said "this is not a real video it's entirely AI-generated" SORA X looks insane the Sora team put real effort into this and it shows" [X Link](https://x.com/slow_developer/status/1973079395863548172) 2025-09-30T17:36Z 56.6K followers, 806.4K engagements "Elon Musk: In 5-6 years the phone becomes an AI edge node basically a screen and audio "no apps no operating systems" a cloud AI talks to your on-device AI generating real-time video you'll get everything through AI that anticipates what you want" [X Link](https://x.com/slow_developer/status/1984342106983256479) 2025-10-31T19:30Z 56.6K followers, 1M engagements "Sam Altman says Slack has many positives but it creates endless fake work We need an AI-native productivity suite to replace docs slides email and Slack Not add-on features but trusted agents that handle work and only escalate when needed This finally feels within reach" [X Link](https://x.com/slow_developer/status/1987409821407064395) 2025-11-09T06:40Z 56.6K followers, 3.1M engagements "xAI should've called Grok XXX "Grok 4.5" because the jump is huge Grok XXX uses large-scale RL and new agentic reward models to boost: - creative writing - personal-tone control - emotional intelligence and the biggest win which few notice is the big drop in hallucinations: Grok 4: XXXXX% Grok 4.1: XXXX% that's roughly 3x fewer hallucinations about a two-thirds reduction" [X Link](https://x.com/slow_developer/status/1990678944623046932) 2025-11-18T07:10Z 56.6K followers, 15.4M engagements "something that didn't get enough attention Gemini X wasn't trained on Nvidia chips; it ran fully on Googles own TPUs that's not surprising but it does show that Google built a fully independent stack and rolled it out across its own products that gives Google a real moat in this race" [X Link](https://x.com/slow_developer/status/1991969868397293779) 2025-11-21T20:40Z 56.6K followers, 836.4K engagements "i honestly don't care who wins every foundation model company and eventually open source will reach AGI superintelligence won't be locked inside a few labs; every major company will get there. we'll have AGI in our pockets with superintelligent systems helping on the hard stuff in the end everyone benefits" [X Link](https://x.com/slow_developer/status/1992007616914993392) 2025-11-21T23:10Z 56.6K followers, 10.4K engagements "what deepseek new model just did is honestly wild they're showing crazy progress while spending way less money than openAI xAI or google from that angle the whole "AI is a bubble" thing kinda makes sense not because AGI isn't coming but because it's clearly way easier and cheaper to get there than the US labs make it look" [X Link](https://x.com/slow_developer/status/1995860504691310866) 2025-12-02T14:20Z 56.6K followers, 7611 engagements "surprisingly METR still hasn't updated results to include gemini X and opus XXX the eval basically measures how well a model handles long complex tasks right now gpt-5.1-codex-max is the top performer that can solve about half of the tasks that would take a skilled human around X to X hours of work" [X Link](https://x.com/slow_developer/status/1996688460359651669) 2025-12-04T21:10Z 56.6K followers, 71K engagements "gemini X deepthink looks strong at solving math and science problems remember how at the start of the year models could barely do math now it's beating humans. arc-agi-2 is already close to XX% and if the trend keeps going it'll probably be fully solved by next year" [X Link](https://x.com/slow_developer/status/1996867138984784223) 2025-12-05T09:00Z 56.6K followers, 12.5K engagements "i think AGI/ASI will come from a powerful narrow AI focused on science and AI research we don't need one model that does everything since we already have super-smart systems in specific areas so when ppl say "we need breakthroughs" it mostly means these narrow AIs need more time to get there" [X Link](https://x.com/slow_developer/status/1996940118947778701) 2025-12-05T13:50Z 56.6K followers, 6781 engagements "wow. Poetiq just set a new arc-agi-2 record: XX% accuracy at only $XXXXX per problem it even beats gemini X deepthink while being about $XX cheaper roughly half the cost and they open-sourced how they did it a small company beating google like this was not what i expected" [X Link](https://x.com/slow_developer/status/1997264757884142005) 2025-12-06T11:20Z 56.6K followers, 8735 engagements "my prediction is that by 2027 around half of new math results will come from AI with experimental sciences catching up a year or two later. by around 2030 AI will be doing most of the research and the world will feel very different and some will still claim it's not AGI because it can't count the r's in "strawberry"" [X Link](https://x.com/slow_developer/status/1997444693278240973) 2025-12-06T23:15Z 56.6K followers, 42.8K engagements "deepmind researchers think we're still early in both pre-training and post-training so there's a lot left to improve. in their view if we get 2-3 more big jumps beyond gpt-5 / gemini X prolevel models with stronger memory and better architectures that could be close to AGI" [X Link](https://x.com/slow_developer/status/1997589396682354785) 2025-12-07T08:50Z 56.6K followers, 11.3K engagements "elon confirmed grok XXXX is dropping in 3-4 weeks which probably means grok X is getting pushed from the "by year-end" promise to around q2-2026 i don't know about you but grok XXX is still the best mix of speed price and reasoning for me the emotional intelligence and creative writing feel dialed in for real-world agent workflows" [X Link](https://x.com/slow_developer/status/1997624628899180635) 2025-12-07T11:10Z 56.6K followers, 13.4K engagements "ARC-AGI-2 isn't the ultimate test; it's just a tough benchmark gemini follows instructions less reliably than opus grok or gpt these benchmarks have been around for over a year and don't fully reflect real-world model performance deepseek-v3.2 models score highly but generalize less effectively than top models in actual use meanwhile some lower-scoring models still perform well in real-world use" [X Link](https://x.com/slow_developer/status/1997732841942401267) 2025-12-07T18:20Z 56.5K followers, 8242 engagements "the idea that AI will end all jobs feels ridiculous to me so far how much real mass unemployment has happened because of AI even if we get AGI soon not every industry will rush to replace people especially licensed roles. fully automate maybe not even until after ASI" [X Link](https://x.com/slow_developer/status/1997804565031260322) 2025-12-07T23:05Z 56.5K followers, 7793 engagements "i believe scaling LLMs will eventually get us to something that behaves like AGI but it would be very inefficient. we'll still need breakthroughs to make AGI work as efficiently as the human brain or better that's basically demis view and yann too though he focuses on world models" [X Link](https://x.com/slow_developer/status/1997949267970015277) 2025-12-08T08:40Z 56.6K followers, 10.2K engagements "Sam Altman says AI is the fastest-adopted technology in history and it's only three years old It is a general-purpose tool that can cure diseases or be used for harm and the rapid change in jobs is inevitable "the critical challenge is ensuring humans have time to adapt"" [X Link](https://x.com/slow_developer/status/1998596030674436123) 2025-12-10T03:30Z 56.6K followers, 27.4K engagements "Erdos problems are starting to be solved by humans supercharged by AI Mathematician Terence Tao wrote up how Erds problem #1026 went from a vague 1975 question to a clean "coin game" then to a rectangle/square-packing view and finally to a full solution using AI tools all put together in about XX hours" [X Link](https://x.com/slow_developer/status/1998704242933067788) 2025-12-10T10:40Z 56.6K followers, 5922 engagements "Elon Musk: I would slow down AI and robotics but I can't because they are advancing rapidly whether I like it or not Nothing usually keeps me up at night but AI is the exception "i've had a lot of AI nightmares. many days in a row" What am I supposed to do about it" [X Link](https://x.com/slow_developer/status/1998774707374666176) 2025-12-10T15:20Z 56.6K followers, 168.2K engagements "my main worry with google is that they sat on their AI for years openAI basically forced them to ship something. when chatgpt dropped they had nothing close got blindsided and rushed out a bad Llamda-based bard now three years later they finally have something that might actually put them in the lead" [X Link](https://x.com/slow_developer/status/1999099346290196607) 2025-12-11T12:50Z 56.6K followers, 9144 engagements "turns out 'code red' didn't mean "oh no we're losing" it meant "okay time to move fast and ship stuff quickly" openAI took it seriously though" [X Link](https://x.com/slow_developer/status/1999203043334607051) 2025-12-11T19:42Z 56.6K followers, 6005 engagements "it's ridiculous a lot of people are still in denial about what's coming and think AI technology is just a hype bubble but this is a wild once-in-a-lifetime moment in human history forget obsessing over profit for a second let's actually focus on building the future" [X Link](https://x.com/slow_developer/status/1999267959324152270) 2025-12-12T00:00Z 56.6K followers, 16.4K engagements "honestly both are pretty bad if you look closely openAI models have never really been great for front-end coding but for back-end work they're amazing especially with rust php/laravel and c++/c # which are core programming languages" [X Link](https://x.com/slow_developer/status/1999399257665864084) 2025-12-12T08:41Z 56.6K followers, 1259 engagements "ugh the fanboys for every AI company are ridiculous google drops a model or something new product and suddenly it's "openAI is finished" now gpt-5.2 is out and it'll flip to "google is finished" both sides of the shilling are pathetic. i'm honestly so over it" [X Link](https://x.com/slow_developer/status/1999626570311303354) 2025-12-12T23:45Z 56.6K followers, 12.5K engagements "gemini X definitely shook openAI a bit and probably pushed them to speed up the release but these models don't just appear overnight because of one moment openAI was clearly building gpt-5.2 for a while and they probably already have the next versions lined up too" [X Link](https://x.com/slow_developer/status/1999734783132524725) 2025-12-13T06:55Z 56.6K followers, 10.1K engagements "sam teased a few "christmas presents" coming next week my guess is images v2 with a much better UI and some sora-style features characters cameos and smoother editing hopefully a new voice mode but i'm not super confident that one drops yet gpt-5.2 codex is what i'm genuinely most excited about right now" [X Link](https://x.com/slow_developer/status/1999805247854969074) 2025-12-13T11:35Z 56.6K followers, 12.3K engagements "sam altman in his recent essay: i've never been more optimistic about openAI research product roadmap and mission and i believe we're almost certain to build superintelligence within XX years but people will keep caring more about what other people do than what machines do and by 2035 we'll be able to do things that are hard to imagine today" [X Link](https://x.com/slow_developer/status/1999950076781994353) 2025-12-13T21:10Z 56.6K followers, 6858 engagements
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
@slow_developer Haider.
Haider. posts on X about ai, open ai, agi, $googl the most. They currently have XXXXXX followers and XXX posts still getting attention that total XXXXXXX engagements in the last XX hours.
Engagements: XXXXXXX #
- X Week XXXXXXXXX -XX%
- X Month XXXXXXXXXX +94%
- X Months XXXXXXXXXX +82%
- X Year XXXXXXXXXXX +93%
Mentions: XX #
- X Week XXX -XX%
- X Month XXX -XX%
- X Months XXXXX +121%
- X Year XXXXX +154%
Followers: XXXXXX #
- X Week XXXXXX +0.60%
- X Month XXXXXX +5.30%
- X Months XXXXXX +27%
- X Year XXXXXX +88%
CreatorRank: XXXXXX #
Social Influence
Social category influence technology brands #2809 stocks #482 celebrities XXXX% finance XXX% countries XXXX% automotive brands XXXX% social networks XXXX%
Social topic influence ai #6622, open ai #34, agi #37, $googl #61, anthropic #34, future 3.6%, strong 2.88%, if you 2.88%, just a 2.88%, level XXXX%
Top accounts mentioned or mentioned by @chaos2cured @grok @scaling01 @wgreklek @askcodi @bob_cloud @mathepi @aitoolscompass @rahll @brandgrowthos @ledpolicy @ndujarrier @h3roai @hellocloh @lucas_montano @xaviercross1971 @knoxbetrayed @mortondowney_iv @itechnologynet @alimamakx
Top assets mentioned Alphabet Inc Class A (GOOGL) Tesla, Inc. (TSLA) Microsoft Corp. (MSFT)
Top Social Posts
Top posts by engagements in the last XX hours
"LLMs have no reasoning no real intelligence yeah okay go ask a random person on the street a few real questions nine times out of ten the latest LLMs will understand you better and give a clearer answer. so when most talk about "AGI" they mean something way beyond human"
X Link 2025-12-01T00:15Z 56.6K followers, 233.9K engagements
"Geoffrey Hinton says there is no technological bubble but there may be an investment bubble The technology advances rapidly but the belief that AI can replace humans for profit ignores the chaos of mass unemployment "people aren't ready for how massive the social disruption will be""
X Link 2025-12-09T15:35Z 56.6K followers, 69.5K engagements
"here are the most important points from today's ilya sutskever podcast: - superintelligence in 5-20 years - current scaling will stall hard; we're back to real research - superintelligence = super-fast continual learner not finished oracle - models generalize 100x worse than humans the biggest AGI blocker - need completely new ML paradigm (i have ideas can't share rn) - AI impact will hit hard but only after economic diffusion - breakthroughs historically needed almost no compute - SSI has enough focused research compute to win - current RL already eats more compute than pre-training"
X Link 2025-11-25T20:30Z 56.6K followers, 1.1M engagements
"gemini X pro is pretty average at complex backend coding but it's seriously good at front-end work gpt-5.1-codex-max is amazing for large codebases and c++ programming but awful at SVG graphics and front-end design"
X Link 2025-12-06T06:40Z 56.6K followers, 28.9K engagements
"Nobel Laureate John Jumper: AlphaFold is now a foundational pillar of modern biology akin to DNA sequencing It's has already accelerated the entire field of structural biology by X or XX percent In XX years almost everyone with modern healthcare will benefit from something shaped by AlphaFold"
X Link 2025-12-06T16:05Z 56.6K followers, 32.6K engagements
"Demis Hassabis says some parts of AI industry are in a bubble like $XX billion seed rounds aren't sustainable But AI is the most transformative technology ever so in the fullness of time the valuations will be justified My job is to make sure we win whether the bubble bursts or keeps inflating"
X Link 2025-12-07T04:00Z 56.6K followers, 51.7K engagements
"saying google will win the AI race feels off to me openAI is working on strong projects like sora and atlas. anthropic leads in coding models and could make a breakthrough. elon has grok plus tesla robotics and self-driving work. microsoft is deeply involved across the industry while building its own models. apple is moving slowly but has the resources to enter late and still do well. meta has huge funding and long-term bets like AR glasses and the metaverse. china has government-backed AI progress fast deployment strong video and research models and massive scale. if i had to pick one it"
X Link 2025-12-09T20:20Z 56.6K followers, 15.4K engagements
"i still don't get why ppl think openAI is in trouble amazon burned cash for its first nine years and didn't make a profit until 2001 openAI is making a similar long-term bet: that it's far enough ahead to keep the lead not just with its LLM but with its whole product suite partnerships and manufacturing plans they're betting AI will grow exponentially like the internet did just even faster"
X Link 2025-12-10T20:10Z 56.6K followers, 11.5K engagements
"google is dominating in multi-modal models and long-context systems economically that's a big deal for the first time gemini passed chatgpt in google trends pushing openAI into a "code red" and because google trains on its own TPUs it's putting real pressure on nvidia grip on the GPU market"
X Link 2025-12-11T01:00Z 56.6K followers, 6797 engagements
"here we go. sam has been cooking pasta and he's adding "garlic" to it as well gpt-5.2 is basically confirmed and should be released on thursday"
X Link 2025-12-11T05:50Z 56.6K followers, 11.7K engagements
"Geoffrey Hinton says AI is a new category of intelligent beings: alive like an AI is alive Because they are digital they can be perfectly copied breaking our biological understanding of identity "if you had an exact copy of yourself you'd still choose the other one to be tortured""
X Link 2025-12-11T15:10Z 56.6K followers, 42.9K engagements
"openAI finally figured out that coding is where the real money is they were slow to focus on it and kept pushing sycophantic chat models which became a headache for them after gpt-5 launched meanwhile anthropic moved faster. now it's clear: coding tools might not have the most users but they're what actually bring in the cash for openAI"
X Link 2025-12-11T17:40Z 56.6K followers, 13.4K engagements
"so here's how it probably goes: google drops gemini XXX pro in a few months and it beats gpt-5.2 then anthropic comes in with sonnet/opus X and takes the lead in coding xAI follows up with grok X a month later a chinese open-source model beats all of them for a fraction of the cost the cycle continues"
X Link 2025-12-11T21:30Z 56.6K followers, 12.1K engagements
"gemini X pro is excellent for ui and front-end there's nothing better model right now gpt-5 series has the same magical feel as sonnet XXX very intuitive and codex pushes those strengths even further i'd build the ui with gemini X and use codex for the backend logic"
X Link 2025-12-12T02:20Z 56.6K followers, 5338 engagements
"DeepMind co-founder Shane Legg: AI won't bring sudden disruption; it will quietly take over economic work domain by domain Within a few years sectors like software engineering will see massive efficiency gains with XX engineers using AI doing the work of 100"
X Link 2025-12-12T04:45Z 56.6K followers, 61.6K engagements
"openAI moved incredibly fast on gpt-5.2 which kind of makes it feel like they could've done this earlier but waited until real competition showed up the good news is that it means no company is completely controlling AI yet there's still real competition"
X Link 2025-12-12T07:10Z 56.6K followers, 8882 engagements
"im waiting for the day when openAI or another big AI lab makes a real breakthrough like when strawberry / post-RL showed up openAI made it sound like the IMO result came from promising new research directions which could lead to big gains on tasks without fixed answers we'll see soon"
X Link 2025-12-12T09:30Z 56.6K followers, 5464 engagements
"i think ppl still don't get yann lecun he's saying we're missing core pieces for AGI and it'll take years of deep research in one of his talks he said we probably need at least two big breakthroughs similar to demis and one might come in the next 3-5 years possibly via JEPA"
X Link 2025-12-12T14:10Z 56.6K followers, 9234 engagements
"@scaling01 gemini XXX being ranked above opus XXX already tells you enough to not take these benchmarks seriously"
X Link 2025-12-12T14:15Z 56.6K followers, 12.2K engagements
"Jeff Dean says predicting ML hardware needs 2-6 years out is nearly impossible in a fast-moving field Google's TPU strategy: embed researchers early experiment with speculative features and dedicate chip area to 'maybes' "if it works the hardware is ready. if not only a small part of the chip is wasted""
X Link 2025-12-12T16:30Z 56.6K followers, 69.7K engagements
"i'm still kinda confused how openAI made that much progress with gpt-5.2 when gpt-5.1 was only a month ago my guess is it was an internal model they held back due to high compute costs and because they didn't think it was needed until gemini X and opus XXX arrived"
X Link 2025-12-13T02:05Z 56.6K followers, 14.4K engagements
"Yann LeCun says LLMs are not a bubble in value or investment; they will power many useful apps and justify big infra The bubble is believing LLMs alone will reach human-level intelligence Progress needs breakthroughs not just more data/compute "we're missing something big""
X Link 2025-11-10T06:30Z 56.5K followers, 323.3K engagements
"current forecasts for the AGI / superintelligence trajectory on the optimistic (red) curve: - dario puts a noticeable chance of AGI/superintelligence by 2026 - elon also expects a real chance by 2026 - jensen sees high odds by 2029 (around 60%+ on the curve) - demis expects something like AGI by 2030 (around 75-80% on the red curve) sam is placed at 2035 around XX% on the blue curve the graph title is "general superintelligence" but the y-axis says "probability of AGI arrival" so it mixes terms"
X Link 2025-11-23T11:00Z 56.5K followers, 33.1K engagements
"Gemini X Pro scored around XXX in an offline test IQ of XXX in that kind of test is roughly in the top X% of human test-takers and on the Mensa Norway test it got the equivalent of about XXX IQ which is around the top XXX% of humans for reference: average person: IQ XXX PhD graduate: usually around IQ 120-130 it's impressive but i still don't think IQ tests are the best way to judge how intelligent an AI really is"
X Link 2025-11-24T11:00Z 56.5K followers, 112.9K engagements
"Anthropic co-founder Jack Clark: We are like children in a dark room but the creatures we see are powerful unpredictable AI systems Some want us to believe AI is just a tool nothing more than a pile of clothes on a chair "you're guaranteed to lose if you believe the creature isn't real""
X Link 2025-11-27T15:20Z 56.5K followers, 213K engagements
"Sundar Pichai says In X years quantum computing will generate the same breathless excitement we see in AI today To meet future compute demands "Project Suncatcher" is a moonshot to build data centers in space "so by 2027 hopefully we'll have some TPU somewhere in space""
X Link 2025-11-29T17:10Z 56.5K followers, 205.2K engagements
"grok X expert mode currently has the top offline IQ test score at XXX even ahead of gemini X pro and scores XXX on the mensa norway test. for anyone curious about these tests: mensa norway: public IQ puzzles likely included in training data offline test: new but similar puzzles so the model can't just rely on memorizing old questions altogether that means grok X is genuinely strong at abstract pattern puzzles and non-verbal reasoning one important aspect of intelligence"
X Link 2025-11-30T12:20Z 56.5K followers, 25.6K engagements
"Satya Nadella says the future of work is macro delegation and micro steering Instead of doing the work humans will assign tasks to agents and manage the exceptions through a "new inbox" built for AI coordination not email "human agency is going to be very much part of it""
X Link 2025-12-01T05:00Z 56.5K followers, 128.8K engagements
"OpenAI Lukasz Kaiser says reasoning models offer ultimate power because they allow AI to think for arbitrarily long periods Moving beyond Transformers this new paradigm uses RL to scale intelligence "this paradigm will soon lead to models that generalize as well as humans do""
X Link 2025-12-01T16:55Z 56.5K followers, 17K engagements
"OpenAI Research Chief Mark Chen: Gemini X is a pretty good model but benchmarks only show part of the picture We already have internal models performing at the same level as Gemini X and even better successors are planned for release soon"
X Link 2025-12-02T16:40Z 56.5K followers, 54.5K engagements
"Fei-Fei Li says AI is not over-hyped; it is a civilizational technology that represents the new computing Wherever chips exist today from light bulbs to self-driving cars AI will inevitably follow "wherever there is compute there will be AI""
X Link 2025-12-03T04:40Z 56.5K followers, 101.8K engagements
"the idea that openAI is already "done" just because some models are ahead is ridiculous that take mostly lives in the 2-5% AI bubble on x and reddit for most people AI still means chatgpt. 2026 looks like a big year for openAI: their custom Broadcom chips should come online the jony ive hardware device (less than X yrs away) and new data centers are all coming online to support these upgrades"
X Link 2025-12-03T14:15Z 56.5K followers, 7709 engagements
"Emad Mostaque says billionaires are building bunkers out of fear of societal collapse AI CEOs are canceling public appearances in anticipation of a massive wave of anti-AI sentiment next year 2026 is when AI crosses the threshold from not good enough to good enough overnight"
X Link 2025-12-03T16:35Z 56.5K followers, 33.8K engagements
"i never thought the dead internet theory would feel this real this fast right now 60-70% of replies are AI-generated everything is getting more automated so we'll actually need proof there's a real person on the other side AI didn't kill the internet; it just made it cheaper and harder to catch"
X Link 2025-12-03T19:00Z 56.5K followers, 5722 engagements
"pls stop saying openAI is already introducing ads they launched app connectors that integrated directly into chatgpt (figma canva zillow etc) when chatgpt thinks it's relevant it suggests one of those apps like zillow if you're asking about apartment rentals still openAI needs to fix this ux bug"
X Link 2025-12-04T09:20Z 56.5K followers, 7505 engagements
"Dario Amodei says scaling will get us to AGI but without a single arrival moment Every few months models get better at coding science and math now winning Olympiads and doing new mathematics Progress has reached a point where Anthropic engineers no longer write code from scratch; they edit claude drafts"
X Link 2025-12-04T16:20Z 56.5K followers, 28.6K engagements
"Jensen Huang says the U.S. is X months ahead on frontier models but China is way ahead on open source Without open sour ce startups can't build researchers can't innovate and entire economies can't advance "without linux where would we be""
X Link 2025-12-05T04:15Z 56.5K followers, 64.5K engagements
"i'm still figuring out claude vs. codex codex is cheaper but slower opus XXX has much lower limits now while sonnet XXX feels almost unlimited like codex pro used to be so it's a strong replacement codex still looks promising and gpt-5.1-high is great for double-checking claude work"
X Link 2025-12-05T06:40Z 56.5K followers, 8697 engagements
"Demis Hassabis says We should scale current systems as far as possible because at the minimum it will be a key component of the final AGI system It could be the entirety of the AGI system But we'll need 1-2 more breakthroughs on the level of the transformer or AlphaGo"
X Link 2025-12-05T16:10Z 56.5K followers, 68.9K engagements
"openAI makes a lot of unexpected changes but i really hope the ads are limited to free accounts i get that they need revenue and with only about X% of XXX million weekly users paying monetizing the other XX% makes sense but ads in paid accounts should never happen. ever"
X Link 2025-12-06T20:50Z 56.5K followers, 11.5K engagements
"llms might be a dead-end for meta since they can't seem to improve their own but they're clearly not a dead-end when openAI google and anthropic keep advancing gpt-4 could barely do anything agentically remember now gpt-5.1-codex-max can run agentic workflows for hours a huge jump in autonomy and intelligence"
X Link 2025-12-07T20:40Z 56.5K followers, 14.6K engagements
"Jensen Huang predicts that in XX years the energy required for AI will be utterly minuscule Today energy is the bottleneck In 6-7 years the current energy bottleneck will be solved by small nuclear reactors effectively turning tech companies into their own power generators"
X Link 2025-12-08T03:50Z 56.5K followers, 70.1K engagements
"one of the first things we can benefit from AI productivity reducing the standard work week from XX hours since XX hours is just a random number. could cut it down to XX hrs and anything over XX should be paid at time-and-a-half first step toward an AI-powered economy"
X Link 2025-12-09T06:10Z 56.5K followers, 7041 engagements
"gpt-5.1-codex-max is my favorite right now it's super thorough and a bit robotic in a good way it double-checks everything. opus XXX is better for general software dev work but it feels a bit less strict about details i guess codex is a smaller model in terms of active parameters but trained on more code RL"
X Link 2025-12-09T08:30Z 56.5K followers, 14.6K engagements
"i keep coming back to this if LLMs were always truthful like no hallucinations no lost context no random errors and no alignment filters blocking direct answers we'd basically already have AGI so to that degree it just feels like we have part-time unreliable AGI"
X Link 2025-12-09T13:15Z 56.5K followers, 5135 engagements
"AI isn't replacing you right now; it's replacing tasks for devs it's a force multiplier that lets you ship bigger projects much faster so you're not just someone banging away at a single file anymore you're more like the chief architect of a whole dev team"
X Link 2025-12-10T01:05Z 56.5K followers, 7232 engagements
"gpt-5.1-thinking is best at web search and reducing hallucinations what i like about the model is that it consistently says "i dont know" when the answer isn't available lots of my questions need fresh research and gpt-5.1 reliably brings back plenty of relevant sources"
X Link 2025-12-10T06:00Z 56.5K followers, 12.7K engagements
"all the big AI platforms are going to have ads by the end of 2026 honestly it feels unavoidable. AI is expensive and they will monetize it fully right now it's basically a money pit even a giant like google with all its infrastructure is still dealing with huge costs"
X Link 2025-12-10T08:20Z 56.5K followers, 5799 engagements
"demis is actually right we should keep scaling current AI systems whether this exact architecture reaches AGI or not best case: it gets us to AGI worst case: it massively accelerates research and helps discover new ideas far faster than humans or weaker models ever could"
X Link 2025-12-10T13:00Z 56.5K followers, 18.6K engagements
"@Angaisb_ 272k wow"
X Link 2025-12-10T16:24Z 56.5K followers, XXX engagements
"not sure but gpt-5.1 with extended thinking feels way better than it did yesterday the reasoning seems sharper and more in control. it reminds me of when openAI silently upgraded from XXX to XXX so maybe this has something to do with gpt-5.2 maybe just me but i noticed a big jump in quality"
X Link 2025-12-10T17:50Z 56.5K followers, 14.6K engagements
"gpt-5.1 is running great right now extremely fast and the router is performing really well too hopefully they don't break anything with gpt-5.2 i use gpt-5.1 for advanced reasoning tests and analyzing algorithm behavior and it performs amazingly gpt-5 was solid just not on the gpt-5.1 level"
X Link 2025-12-11T08:10Z 56.5K followers, 11.8K engagements
"Elon Musk: We are approaching the Singularity a future where work is optional and if you can think of it you can have it But human consumption has limits. Eventually AI will saturate everything humans could want At that point machines start serving themselves once humans need nothing more"
X Link 2025-11-30T17:05Z 56.6K followers, 497.8K engagements
"dario wasn't wrong around 80-90% of my code is written by AI i generate review and refine which makes me feel 10x more productive like having a really good junior dev helping out at this pace even the architecture part might not stay "purely human" for much longer"
X Link 2025-12-07T06:30Z 56.6K followers, 123.8K engagements
"Jensen Huang predicts that within XX years XX% of the world's knowledge will be generated by AI Right now knowledge is created shared and modified by humans Soon most of it will be synthetic created by machines mixed with human input some true some not "that's crazy but it's just fine""
X Link 2025-12-09T03:40Z 56.6K followers, 16.8K engagements
"Yann LeCun says American AI labs are closing up while China goes fully open source Meta is rethinking its open source strategy OpenAI and Anthropic locked down long ago and Google is mostly following suit The best open source LLMs now come from China "and it's not good""
X Link 2025-12-11T03:20Z 56.6K followers, 121.2K engagements
"elon has admitted that opus XXX is incredible no doubt it's a beast of a model i mostly bounce b/w opus XXX and gpt-5.1-codex-max for coding and design. codex follow instructions well but opus feels noticeably faster than codex max grok XXX is solid too; i use its API as a cheap reasoning engine so it's an easy go-to for me"
X Link 2025-12-11T10:30Z 56.6K followers, 17.5K engagements
"gpt-5.2 benchmarks are wild: AIME 2025 (math): XXXXX% vs gemini X pro (95.0%) ARC-AGI-2 (abstract reasoning): XXXX% vs gemini X pro (31.1%) SWE-bench pro (coding): XXXX% vs gemini X pro (43.3%) GDPval (knowledge work): XXXX% which openAI claims is the first time a model reaches "human expert level""
X Link 2025-12-11T18:37Z 56.6K followers, 59.1K engagements
"why is no one talking about gpt-5.2 hallucination drop with browsing off: vs gpt-5: wrong claims drop XXX% XXX% (34% fewer) vs gpt-5: major wrong answers drop XXXX% XXXX% (35% fewer) with browsing on: vs gpt-5.1: wrong claims drop XXX% XXX% (47% fewer) vs gpt-5.1: major wrong answers drop XXX% XXX% (34% fewer) this is the real upgrade. we keep obsessing over speed and benchmark scores but reliability is what makes a model actually usable"
X Link 2025-12-12T21:20Z 56.6K followers, 14.7K engagements
"@scaling01 you do really know it's only a ".1" improvement over gpt-5.1 which was released a month ago right"
X Link 2025-12-13T05:53Z 56.6K followers, 2175 engagements
"Sergey Brin admits Google messed up by under-investing in the transformer architecture it invented Google was too scared to release chatbots that "say dumb things" so it under-invested in scaling compute "we didn't take it very seriously. and openAI ran with it""
X Link 2025-12-13T16:20Z 56.6K followers, 362.6K engagements
"Anthropic CEO Dario Amodei in the next X to X months AI is writing XX% of the code and in XX months nearly all code may be generated by AI"
X Link 2025-03-11T12:00Z 56.6K followers, 9.3M engagements
""1m context window" models after 200k tokens"
X Link 2025-09-01T08:37Z 56.6K followers, 566.9K engagements
"Sam Altman proposed a future AGI test: if a model like "GPT-8" solved quantum gravity and could explain the reasoning behind its discovery would that qualify as AGI David Deutsch agreed that it would qualify as AGI making it a potential benchmark"
X Link 2025-09-25T21:20Z 56.6K followers, 769.1K engagements
"you have to be kidding me when someone said "this is not a real video it's entirely AI-generated" SORA X looks insane the Sora team put real effort into this and it shows"
X Link 2025-09-30T17:36Z 56.6K followers, 806.4K engagements
"Elon Musk: In 5-6 years the phone becomes an AI edge node basically a screen and audio "no apps no operating systems" a cloud AI talks to your on-device AI generating real-time video you'll get everything through AI that anticipates what you want"
X Link 2025-10-31T19:30Z 56.6K followers, 1M engagements
"Sam Altman says Slack has many positives but it creates endless fake work We need an AI-native productivity suite to replace docs slides email and Slack Not add-on features but trusted agents that handle work and only escalate when needed This finally feels within reach"
X Link 2025-11-09T06:40Z 56.6K followers, 3.1M engagements
"xAI should've called Grok XXX "Grok 4.5" because the jump is huge Grok XXX uses large-scale RL and new agentic reward models to boost: - creative writing - personal-tone control - emotional intelligence and the biggest win which few notice is the big drop in hallucinations: Grok 4: XXXXX% Grok 4.1: XXXX% that's roughly 3x fewer hallucinations about a two-thirds reduction"
X Link 2025-11-18T07:10Z 56.6K followers, 15.4M engagements
"something that didn't get enough attention Gemini X wasn't trained on Nvidia chips; it ran fully on Googles own TPUs that's not surprising but it does show that Google built a fully independent stack and rolled it out across its own products that gives Google a real moat in this race"
X Link 2025-11-21T20:40Z 56.6K followers, 836.4K engagements
"i honestly don't care who wins every foundation model company and eventually open source will reach AGI superintelligence won't be locked inside a few labs; every major company will get there. we'll have AGI in our pockets with superintelligent systems helping on the hard stuff in the end everyone benefits"
X Link 2025-11-21T23:10Z 56.6K followers, 10.4K engagements
"what deepseek new model just did is honestly wild they're showing crazy progress while spending way less money than openAI xAI or google from that angle the whole "AI is a bubble" thing kinda makes sense not because AGI isn't coming but because it's clearly way easier and cheaper to get there than the US labs make it look"
X Link 2025-12-02T14:20Z 56.6K followers, 7611 engagements
"surprisingly METR still hasn't updated results to include gemini X and opus XXX the eval basically measures how well a model handles long complex tasks right now gpt-5.1-codex-max is the top performer that can solve about half of the tasks that would take a skilled human around X to X hours of work"
X Link 2025-12-04T21:10Z 56.6K followers, 71K engagements
"gemini X deepthink looks strong at solving math and science problems remember how at the start of the year models could barely do math now it's beating humans. arc-agi-2 is already close to XX% and if the trend keeps going it'll probably be fully solved by next year"
X Link 2025-12-05T09:00Z 56.6K followers, 12.5K engagements
"i think AGI/ASI will come from a powerful narrow AI focused on science and AI research we don't need one model that does everything since we already have super-smart systems in specific areas so when ppl say "we need breakthroughs" it mostly means these narrow AIs need more time to get there"
X Link 2025-12-05T13:50Z 56.6K followers, 6781 engagements
"wow. Poetiq just set a new arc-agi-2 record: XX% accuracy at only $XXXXX per problem it even beats gemini X deepthink while being about $XX cheaper roughly half the cost and they open-sourced how they did it a small company beating google like this was not what i expected"
X Link 2025-12-06T11:20Z 56.6K followers, 8735 engagements
"my prediction is that by 2027 around half of new math results will come from AI with experimental sciences catching up a year or two later. by around 2030 AI will be doing most of the research and the world will feel very different and some will still claim it's not AGI because it can't count the r's in "strawberry""
X Link 2025-12-06T23:15Z 56.6K followers, 42.8K engagements
"deepmind researchers think we're still early in both pre-training and post-training so there's a lot left to improve. in their view if we get 2-3 more big jumps beyond gpt-5 / gemini X prolevel models with stronger memory and better architectures that could be close to AGI"
X Link 2025-12-07T08:50Z 56.6K followers, 11.3K engagements
"elon confirmed grok XXXX is dropping in 3-4 weeks which probably means grok X is getting pushed from the "by year-end" promise to around q2-2026 i don't know about you but grok XXX is still the best mix of speed price and reasoning for me the emotional intelligence and creative writing feel dialed in for real-world agent workflows"
X Link 2025-12-07T11:10Z 56.6K followers, 13.4K engagements
"ARC-AGI-2 isn't the ultimate test; it's just a tough benchmark gemini follows instructions less reliably than opus grok or gpt these benchmarks have been around for over a year and don't fully reflect real-world model performance deepseek-v3.2 models score highly but generalize less effectively than top models in actual use meanwhile some lower-scoring models still perform well in real-world use"
X Link 2025-12-07T18:20Z 56.5K followers, 8242 engagements
"the idea that AI will end all jobs feels ridiculous to me so far how much real mass unemployment has happened because of AI even if we get AGI soon not every industry will rush to replace people especially licensed roles. fully automate maybe not even until after ASI"
X Link 2025-12-07T23:05Z 56.5K followers, 7793 engagements
"i believe scaling LLMs will eventually get us to something that behaves like AGI but it would be very inefficient. we'll still need breakthroughs to make AGI work as efficiently as the human brain or better that's basically demis view and yann too though he focuses on world models"
X Link 2025-12-08T08:40Z 56.6K followers, 10.2K engagements
"Sam Altman says AI is the fastest-adopted technology in history and it's only three years old It is a general-purpose tool that can cure diseases or be used for harm and the rapid change in jobs is inevitable "the critical challenge is ensuring humans have time to adapt""
X Link 2025-12-10T03:30Z 56.6K followers, 27.4K engagements
"Erdos problems are starting to be solved by humans supercharged by AI Mathematician Terence Tao wrote up how Erds problem #1026 went from a vague 1975 question to a clean "coin game" then to a rectangle/square-packing view and finally to a full solution using AI tools all put together in about XX hours"
X Link 2025-12-10T10:40Z 56.6K followers, 5922 engagements
"Elon Musk: I would slow down AI and robotics but I can't because they are advancing rapidly whether I like it or not Nothing usually keeps me up at night but AI is the exception "i've had a lot of AI nightmares. many days in a row" What am I supposed to do about it"
X Link 2025-12-10T15:20Z 56.6K followers, 168.2K engagements
"my main worry with google is that they sat on their AI for years openAI basically forced them to ship something. when chatgpt dropped they had nothing close got blindsided and rushed out a bad Llamda-based bard now three years later they finally have something that might actually put them in the lead"
X Link 2025-12-11T12:50Z 56.6K followers, 9144 engagements
"turns out 'code red' didn't mean "oh no we're losing" it meant "okay time to move fast and ship stuff quickly" openAI took it seriously though"
X Link 2025-12-11T19:42Z 56.6K followers, 6005 engagements
"it's ridiculous a lot of people are still in denial about what's coming and think AI technology is just a hype bubble but this is a wild once-in-a-lifetime moment in human history forget obsessing over profit for a second let's actually focus on building the future"
X Link 2025-12-12T00:00Z 56.6K followers, 16.4K engagements
"honestly both are pretty bad if you look closely openAI models have never really been great for front-end coding but for back-end work they're amazing especially with rust php/laravel and c++/c # which are core programming languages"
X Link 2025-12-12T08:41Z 56.6K followers, 1259 engagements
"ugh the fanboys for every AI company are ridiculous google drops a model or something new product and suddenly it's "openAI is finished" now gpt-5.2 is out and it'll flip to "google is finished" both sides of the shilling are pathetic. i'm honestly so over it"
X Link 2025-12-12T23:45Z 56.6K followers, 12.5K engagements
"gemini X definitely shook openAI a bit and probably pushed them to speed up the release but these models don't just appear overnight because of one moment openAI was clearly building gpt-5.2 for a while and they probably already have the next versions lined up too"
X Link 2025-12-13T06:55Z 56.6K followers, 10.1K engagements
"sam teased a few "christmas presents" coming next week my guess is images v2 with a much better UI and some sora-style features characters cameos and smoother editing hopefully a new voice mode but i'm not super confident that one drops yet gpt-5.2 codex is what i'm genuinely most excited about right now"
X Link 2025-12-13T11:35Z 56.6K followers, 12.3K engagements
"sam altman in his recent essay: i've never been more optimistic about openAI research product roadmap and mission and i believe we're almost certain to build superintelligence within XX years but people will keep caring more about what other people do than what machines do and by 2035 we'll be able to do things that are hard to imagine today"
X Link 2025-12-13T21:10Z 56.6K followers, 6858 engagements