#  @S_OhEigeartaigh Seán Ó hÉigeartaigh
Seán Ó hÉigeartaigh posts on X about ai, open ai, agi, china the most. They currently have [---------] followers and [---] posts still getting attention that total [---] engagements in the last [--] hours.
### Engagements: [---] [#](/creator/twitter::1372878390672420867/interactions)
- [--] Week [------] -76%
- [--] Month [-------] +264%
- [--] Months [----------] +24%
- [--] Year [----------] +822%
### Mentions: [--] [#](/creator/twitter::1372878390672420867/posts_active)
- [--] Week [--] -45%
- [--] Month [--] +200%
- [--] Months [---] +138%
- [--] Year [---] +267%
### Followers: [---------] [#](/creator/twitter::1372878390672420867/followers)
- [--] Week [-----] +0.36%
- [--] Month [-----] +3.60%
- [--] Months [-----] +22%
- [--] Year [-----] +50%
### CreatorRank: [---------] [#](/creator/twitter::1372878390672420867/influencer_rank)
### Social Influence
**Social category influence**
[technology brands](/list/technology-brands) [countries](/list/countries) [celebrities](/list/celebrities) [travel destinations](/list/travel-destinations) [stocks](/list/stocks) [finance](/list/finance) [vc firms](/list/vc-firms) [events](/list/events) [social networks](/list/social-networks) [us election](/list/us-election)
**Social topic influence**
[ai](/topic/ai), [open ai](/topic/open-ai), [agi](/topic/agi), [china](/topic/china), [in the](/topic/in-the), [for the](/topic/for-the), [elon musk](/topic/elon-musk), [anthropic](/topic/anthropic), [the world](/topic/the-world), [$googl](/topic/$googl)
**Top accounts mentioned or mentioned by**
[@shakeelhashim](/creator/undefined) [@tylermjohn](/creator/undefined) [@peterwildeford](/creator/undefined) [@davidmanheim](/creator/undefined) [@davidsacks](/creator/undefined) [@andrewcritchphd](/creator/undefined) [@jackstilgoe](/creator/undefined) [@grok](/creator/undefined) [@nathancalvin](/creator/undefined) [@sriramk](/creator/undefined) [@martinshkreli](/creator/undefined) [@garymarcus](/creator/undefined) [@milesbrundage](/creator/undefined) [@elonmusk](/creator/undefined) [@daniel_271828](/creator/undefined) [@jamiesusskind](/creator/undefined) [@nathanpmyoung](/creator/undefined) [@humanharlan](/creator/undefined) [@gmiller](/creator/undefined) [@soheigeartaigh](/creator/undefined)
**Top assets mentioned**
[Alphabet Inc Class A (GOOGL)](/topic/$googl) [Microsoft Corp. (MSFT)](/topic/microsoft) [Frontier (FRONT)](/topic/frontier)
### Top Social Posts
Top posts by engagements in the last [--] hours
"It's a privilege to have been invited to speak at the Dublin Tech Summit today Always lovely to have an excuse to come home to an uncharacteristically sunny Dublin (And I should probably thank the CAIS folk for creating some handy news for us to discuss) https://dublintechsummit.tech/ https://dublintechsummit.tech/"
[X Link](https://x.com/S_OhEigeartaigh/status/1663818134988414976) 2023-05-31T08:02Z [----] followers, [---] engagements
"@LeverhulmeCFI The other important clarification is that while this article presents the Cambridge Institute as being v much centred on AI its remit is deliberately much broader. Right now it includes & will expand on work at CSER on climate biorisk nuclear weapons geoengineering etc 5/n"
[X Link](https://x.com/S_OhEigeartaigh/status/1678317559400091649) 2023-07-10T08:27Z [----] followers, [---] engagements
"It was a pleasure to speak at the Dublin Tech Summit this year. Just came across this nice writeup. https://businessandfinance.com/news/highlights-of-day-1-on-the-evolution-stage-dublin-tech-summit/ https://businessandfinance.com/news/highlights-of-day-1-on-the-evolution-stage-dublin-tech-summit/"
[X Link](https://x.com/S_OhEigeartaigh/status/1686701858570514432) 2023-08-02T11:33Z [----] followers, [---] engagements
"Disappointed to see the World Health Organisation promoting a pseudoscience like homeopathy (not to mention mislabelling it a 'traditional medicine')"
[X Link](https://x.com/S_OhEigeartaigh/status/1691872079216312670) 2023-08-16T17:58Z [----] followers, [----] engagements
"Useful thread from Robert Trager one of the lead authors of the report"
[X Link](https://x.com/S_OhEigeartaigh/status/1697342363901849731) 2023-08-31T20:15Z [----] followers, [---] engagements
"Writing reference letters for North American roles: is it on me to amp up the hyperbole or on them to correct for my European curmudgeonship Do you not know that "Ah he's grand like" is the Highest Praise that my People Offer"
[X Link](https://x.com/S_OhEigeartaigh/status/1701302386117992455) 2023-09-11T18:31Z [----] followers, [----] engagements
"Sign me up for more "Tegmark unplugged""
[X Link](https://x.com/S_OhEigeartaigh/status/1701306065940312073) 2023-09-11T18:45Z [----] followers, [---] engagements
"Delighted to see my colleague and coauthor Yi Zeng on the UN Advisory Body. He brings a remarkable wealth of knowledge and wisdom and the UN (and world) will benefit tremendously from it"
[X Link](https://x.com/S_OhEigeartaigh/status/1717605616888205794) 2023-10-26T18:14Z [----] followers, [---] engagements
"Impressive and comprehensive review of AI alignment from scholars at Peking University Cambridge CMU HKUST and USC"
[X Link](https://x.com/S_OhEigeartaigh/status/1719820809294733439) 2023-11-01T20:56Z [----] followers, [---] engagements
"@mealreplacer Don't be such an igg"
[X Link](https://x.com/S_OhEigeartaigh/status/1725351266090389636) 2023-11-17T03:12Z [----] followers, [--] engagements
"@halhod @jamiesusskind Also n=1. Did anyone expect Google/DM to go from 'significantly behind' OpenAI for this kind of model to "blowing OpenAI out of the water" They've caught up and improved sightly which is impressive in of itself. Now let's see what the leaders do"
[X Link](https://x.com/S_OhEigeartaigh/status/1732434308734861499) 2023-12-06T16:18Z [----] followers, [--] engagements
""To account for the wide range of tasks AI systems can accomplish and the quick expansion of its capabilities it was agreed that general-purpose AI (GPAI) systems and the GPAI models they are based on will have to adhere to transparency requirements""
[X Link](https://x.com/S_OhEigeartaigh/status/1733303374202949776) 2023-12-09T01:51Z [----] followers, [--] engagements
"This is really remarkable: Half-million-year-old interlocking wooden logs unearthed in Zambia eariler this year"
[X Link](https://x.com/S_OhEigeartaigh/status/1733336718252929394) 2023-12-09T04:04Z [----] followers, [---] engagements
"First time to Hong Kong and I had an inspiring time at HKUST's International AI Cooperation and Governance forum. I had the opportunity to speak on considerations around open-sourcing powerful foundation models and to participate in an excellent panel on safety & governance"
[X Link](https://x.com/S_OhEigeartaigh/status/1733650746141544652) 2023-12-10T00:52Z [----] followers, [---] engagements
"As well as Hong Kong I hugely enjoyed a visit to Beijing last month. Re-energised collaborations & friendships & connected to some very exciting researchers & think tanks working on AI alignment global ethics AI risk & global AI governance. Here w/ two of @yi_zeng 's PhDs"
[X Link](https://x.com/S_OhEigeartaigh/status/1734519324168126823) 2023-12-12T10:23Z [----] followers, [---] engagements
"New essential reading blog just dropped. I started a blog about scientific research and science funding. You can subscribe on the site to receive future posts as emails. Heres the first one. Society is not focused on some of the most important technical problems: https://t.co/89h1sZ0rGe I started a blog about scientific research and science funding. You can subscribe on the site to receive future posts as emails. Heres the first one. Society is not focused on some of the most important technical problems: https://t.co/89h1sZ0rGe"
[X Link](https://x.com/S_OhEigeartaigh/status/1754338908660482161) 2024-02-05T02:59Z [----] followers, [---] engagements
"@praeterpropter Maybe they can all come over and fix another moribund economy dragged down by political decisions"
[X Link](https://x.com/S_OhEigeartaigh/status/1757869973928657271) 2024-02-14T20:50Z [----] followers, [--] engagements
"tl;dr Elon Musk is suing OpenAI; claiming GPT-4 & rumoured 'Q' are AGI & should be released to public; that Altman and Microsoft have replaced OpenAI's board with Microsoft-friendly folk who know nothing about A(G)I 1/2"
[X Link](https://x.com/S_OhEigeartaigh/status/1763612998722027882) 2024-03-01T17:11Z [----] followers, [---] engagements
"so Microsoft can continue controlling AGI and make a fortune selling it to the public (if it is AGI it is outside OpenAI's agreement with Microsoft). .how's your Friday going"
[X Link](https://x.com/S_OhEigeartaigh/status/1763613198719009249) 2024-03-01T17:12Z [----] followers, [---] engagements
"My view: GPT-4 is not AGI. I also do not believe AGI 'has been achieved internally' or is likely to be achieved in the coming year. I well believe Microsoft might want a board that would keep OpenAI stable and relationship with Microsoft strong but they're not tying to hide AGI"
[X Link](https://x.com/S_OhEigeartaigh/status/1763615016538788324) 2024-03-01T17:19Z [----] followers, [---] engagements
"Nvidia now 3rd largest company in the world by market cap passing out Saudi Aramco. [--] out of the [--] largest in world (Microsoft Nvidia Amazon Alphabet Meta) very much in AI chain (6 of [--] if you include Apple). This is noteworthy"
[X Link](https://x.com/S_OhEigeartaigh/status/1764962399755321802) 2024-03-05T10:33Z [----] followers, [---] engagements
"@Jackstilgoe Boston dynamics have a lot to answer for here"
[X Link](https://x.com/S_OhEigeartaigh/status/1765295942402470026) 2024-03-06T08:38Z [----] followers, [--] engagements
"Great to see this fascinating paper out led by CSER affiliate Tom Cernev and informed by an expert elicitation and workshop process held in Cambridge. After the Starship launch last week it feels like Space is back baby"
[X Link](https://x.com/S_OhEigeartaigh/status/1769685160390230254) 2024-03-18T11:19Z [----] followers, [---] engagements
"@anabrandusescu This guy's waiting for your phonecall https://www.bloomberg.com/news/articles/2023-06-01/palantir-ceo-touts-power-of-ai-at-customer-conference https://www.bloomberg.com/news/articles/2023-06-01/palantir-ceo-touts-power-of-ai-at-customer-conference"
[X Link](https://x.com/S_OhEigeartaigh/status/1775507727110045766) 2024-04-03T12:56Z [----] followers, [---] engagements
"@briandavidearp @KellerScholl Much more trivially what an odd metaphor to choose. "Don't worry about the lion it's your own worry that is the risk" is. an interesting framing for the AI risk issue. How I stopped worrying and learned to love the Palantir I guess"
[X Link](https://x.com/S_OhEigeartaigh/status/1776381323675881519) 2024-04-05T22:48Z [----] followers, [--] engagements
"This overview of the EU AI Act's risk classification framework by CFI's Yulu Pi is very helpful and highly recommended. http://lcfi.ac.uk/news-and-events/news/2024/apr/2/eu-ai-act-high-risk-ai-nutshell/ http://lcfi.ac.uk/news-and-events/news/2024/apr/2/eu-ai-act-high-risk-ai-nutshell/"
[X Link](https://x.com/S_OhEigeartaigh/status/1779588284131246402) 2024-04-14T19:11Z [----] followers, [----] engagements
"Excited to welcome @MichelleMalonza to Cambridge I'm thrilled and excited at the incredible calibre of scholars we get to engage with through the @ILINAProgram . Welcome to CSER visitor @MichelleMalonza Michelle Malonza is a Research Associate at the @ILINAProgram. At CSER she will be working on an OpenAI funded project on how democratic societies can responsibly integrate AI into their critical infrastructure. https://t.co/p4hr36feVP https://t.co/JcrT9qB1Sw Welcome to CSER visitor @MichelleMalonza Michelle Malonza is a Research Associate at the @ILINAProgram. At CSER she will be working on"
[X Link](https://x.com/S_OhEigeartaigh/status/1785615867658609065) 2024-05-01T10:22Z [----] followers, [----] engagements
"@MichelleMalonza @ILINAProgram Michelle will be working on governance of AI agents and their interaction with critical infrastructure but I also enjoyed this paper on the role of Global South countries in frontier AI governance - a very important topic. https://arxiv.org/abs/2312.04616 https://arxiv.org/abs/2312.04616"
[X Link](https://x.com/S_OhEigeartaigh/status/1785616257896575253) 2024-05-01T10:24Z [----] followers, [---] engagements
"@NathanpmYoung I think part of the issue here is that EA is sort-of an integrated thing (use reason & evidence to do most good) and sort-of [--] cause area foxes in a trenchcoat all with quite different approaches. So while there might have been a bit of bait-and-switch by some pplfeels like1/2"
[X Link](https://x.com/S_OhEigeartaigh/status/1785617201459806574) 2024-05-01T10:28Z [----] followers, [---] engagements
"Apologies to everyone I owe messages/emails to. Got back last week from an incredible trip to Beijing. Still catching up. Will say more soon - despite everything it feels like there's a real groundswell of interest in AI safety and cooperation happening there"
[X Link](https://x.com/S_OhEigeartaigh/status/1788967928005636234) 2024-05-10T16:22Z [----] followers, [---] engagements
"Useful thread from the ever-excellent Ed Conway on tariffs announced yesterday by Biden. Key question (in terms of electric vehicle uptake/cost which is important for net zero): how does this affect components 🧵 Joe Biden has just confirmed he's going to raise the special tariff on electric vehicles coming from China to 100%. Also new tariffs on batteries and solar panels. What's going on here Here's a quick primer with some charts 👇 🧵 Joe Biden has just confirmed he's going to raise the special tariff on electric vehicles coming from China to 100%. Also new tariffs on batteries and solar"
[X Link](https://x.com/S_OhEigeartaigh/status/1790699339884478755) 2024-05-15T11:02Z [----] followers, [---] engagements
"Forget throwback Thursday; it's Frontier (Safety) Friday: https://deepmind.google/discover/blog/introducing-the-frontier-safety-framework/ https://deepmind.google/discover/blog/introducing-the-frontier-safety-framework/"
[X Link](https://x.com/S_OhEigeartaigh/status/1791477671479673025) 2024-05-17T14:35Z [----] followers, [---] engagements
"@albrgr Not sure he was on their safety team but certainly safety-concerned: "Quit OpenAI due to losing confidence that it would behave responsibly around the time of AGI." https://www.lesswrong.com/users/daniel-kokotajlo https://www.lesswrong.com/users/daniel-kokotajlo"
[X Link](https://x.com/S_OhEigeartaigh/status/1791504447614464250) 2024-05-17T16:21Z [----] followers, [---] engagements
"@eshear Have you considered being CEO of an AGI company"
[X Link](https://x.com/S_OhEigeartaigh/status/1791850467758166037) 2024-05-18T15:16Z [----] followers, [--] engagements
"@ShakeelHashim That's what they get for not checking whether any openAI drama was going to drop that day. Very short-sighted"
[X Link](https://x.com/S_OhEigeartaigh/status/1792876039456137385) 2024-05-21T11:12Z [----] followers, [---] engagements
"@HaydnBelfield Want to swap Damn cat eats every plant we bring into the house. I thought it was supposed to be a carnivore"
[X Link](https://x.com/S_OhEigeartaigh/status/1800610752270630995) 2024-06-11T19:27Z [----] followers, [---] engagements
"@Zach49639228 @sleepinyourhat Huh yeah that's what I must be remembering. Can anyone help clear this up"
[X Link](https://x.com/S_OhEigeartaigh/status/1803902450954338549) 2024-06-20T21:27Z [----] followers, [---] engagements
"Impressive from the White House: https://www.whitehouse.gov/briefing-room/statements-releases/2024/07/26/fact-sheet-biden-harris-administration-announces-new-ai-actions-and-receives-additional-major-voluntary-commitment-on-ai/ https://www.whitehouse.gov/briefing-room/statements-releases/2024/07/26/fact-sheet-biden-harris-administration-announces-new-ai-actions-and-receives-additional-major-voluntary-commitment-on-ai/"
[X Link](https://x.com/S_OhEigeartaigh/status/1816932899305980371) 2024-07-26T20:25Z [----] followers, [---] engagements
"A lot of interesting work going on within the google deepmind AGI safety&alignment team - a helpful overview. So freaking proud of the AGI safety&alignment team -- read here a retrospective of the work over the past [---] years across frontier safety oversight interpretability and more. Onwards https://t.co/p7N0KM1LJ8 So freaking proud of the AGI safety&alignment team -- read here a retrospective of the work over the past [---] years across frontier safety oversight interpretability and more. Onwards https://t.co/p7N0KM1LJ8"
[X Link](https://x.com/S_OhEigeartaigh/status/1825961147578962231) 2024-08-20T18:20Z [----] followers, [---] engagements
"Pleased to have contributed to this paper led by the excellent Hadrian Pouget and Claire Dennis on the future of AI scientific risk assessment. 1/3 How can nations achieve a shared scientifically rigorous understanding of AI risks Check out our new report that examines the challenges and opportunities https://t.co/Y50HJs2TTv https://t.co/qGZCQflxnu How can nations achieve a shared scientifically rigorous understanding of AI risks Check out our new report that examines the challenges and opportunities https://t.co/Y50HJs2TTv https://t.co/qGZCQflxnu"
[X Link](https://x.com/S_OhEigeartaigh/status/1829493395200327953) 2024-08-30T12:16Z [----] followers, [----] engagements
"The core message of the paperthat no single institution can accomplish everything needed from a global scientific AI risk assessment effortis esp important in light of Summit of the Futurewhere UN will establish foundation for its contribution to global risk assessments.2/3"
[X Link](https://x.com/S_OhEigeartaigh/status/1829493397436105108) 2024-08-30T12:16Z [----] followers, [---] engagements
"Useful article on the change in attention to and views on frontier AI safety in China by Matt Sheehan - broadly matches my experience: https://carnegieendowment.org/research/2024/08/china-artificial-intelligence-ai-safety-regulationlang=en https://carnegieendowment.org/research/2024/08/china-artificial-intelligence-ai-safety-regulationlang=en"
[X Link](https://x.com/S_OhEigeartaigh/status/1831299411420286978) 2024-09-04T11:52Z [----] followers, [---] engagements
"@davidad Or that Hiroaki Kitano CTO of Sony is the same Hiroaki Kitano whose systems biology papers I used to read"
[X Link](https://x.com/S_OhEigeartaigh/status/1834182406456136114) 2024-09-12T10:48Z [----] followers, [---] engagements
"@GaryMarcus I get that you've been interacting with unrealistic fanboys but this feels like it's setting the bar incredibly high. Steady incremental progress stacks up and over the course of years it's been *very* impressive"
[X Link](https://x.com/S_OhEigeartaigh/status/1834301091917238601) 2024-09-12T18:40Z [----] followers, [----] engagements
"@Manderljung It worries me how front of stage they've become and how quickly particularly in the policy realm. My intuition is to be more worried abt (a) general capability trends (b) autonomy (c) deception (d) applicability to AI research & I worry this might point us in wrong direction"
[X Link](https://x.com/S_OhEigeartaigh/status/1834540353472331996) 2024-09-13T10:31Z [----] followers, [--] engagements
"Insightful blog post by Lennart Heim exploring the compute governance implications of the OpenAI o1 models https://blog.heim.xyz/inference-compute/ https://blog.heim.xyz/inference-compute/"
[X Link](https://x.com/S_OhEigeartaigh/status/1838281461452902643) 2024-09-23T18:17Z [----] followers, [---] engagements
"@jamiesusskind It's almost as if being in power wasn't much fun and they never want to win an election again"
[X Link](https://x.com/S_OhEigeartaigh/status/1840355689044054105) 2024-09-29T11:39Z [----] followers, [--] engagements
"@jamiesusskind "Come up with a less vote-winning opinion than Kemi Badenoch" challenge. And they've already played the "bring back national service " card so you can't use that"
[X Link](https://x.com/S_OhEigeartaigh/status/1840357173508932020) 2024-09-29T11:45Z [----] followers, [--] engagements
"@anderssandberg What about stopping A(G)I development specifically though It's only a small part of the overall space if a disproportionately important and risky one. Could still get lots of benefits from the rest"
[X Link](https://x.com/S_OhEigeartaigh/status/1841972870697583027) 2024-10-03T22:45Z [----] followers, [---] engagements
"Great to see Geoffrey Hinton with the Nobel Prize and grateful to Cambridge for featuring his [----] lecture for us on AI and existential risk in their excellent news story https://www.cam.ac.uk/research/news/university-of-cambridge-alumnus-awarded-2024-nobel-prize-in-physics https://www.cam.ac.uk/research/news/university-of-cambridge-alumnus-awarded-2024-nobel-prize-in-physics"
[X Link](https://x.com/S_OhEigeartaigh/status/1843620696028745893) 2024-10-08T11:53Z [----] followers, [----] engagements
"It's not done yet. Hearing reports that the Nobel prize for literature will be going to the authors of "OpenAI's nonprofit governance structure" for outstanding contributions to creative fiction"
[X Link](https://x.com/anyuser/status/1843979139948355893) 2024-10-09T11:37Z [----] followers, 266.2K engagements
"How can you tell when you're being RLHF'd into becoming a niche corporate AI governance comedy account Asking for a friend. Also did you hear the one about the antitrust probe into microsoft's acqui-merger of Mistral"
[X Link](https://x.com/S_OhEigeartaigh/status/1844041826501865923) 2024-10-09T15:46Z [----] followers, [----] engagements
"@LinchZhang Get Jurgen Schmidthuber in there and I'll be impressed with your diplomatic nous https://x.com/SchmidhuberAI/status/1844022724328394780 The #NobelPrizeinPhysics2024 for Hopfield & Hinton rewards plagiarism and incorrect attribution in computer science. It's mostly about Amari's "Hopfield network" and the "Boltzmann Machine." [--]. The Lenz-Ising recurrent architecture with neuron-like elements was published in https://x.com/SchmidhuberAI/status/1844022724328394780 The #NobelPrizeinPhysics2024 for Hopfield & Hinton rewards plagiarism and incorrect attribution in computer science. It's"
[X Link](https://x.com/S_OhEigeartaigh/status/1844062170235154517) 2024-10-09T17:07Z [----] followers, [---] engagements
"@LinchZhang In case you didn't know he invented the idea of "the AI doesn't care about you" https://www.theguardian.com/technology/2017/apr/18/robot-man-artificial-intelligence-computer-milky-way https://www.theguardian.com/technology/2017/apr/18/robot-man-artificial-intelligence-computer-milky-way"
[X Link](https://x.com/S_OhEigeartaigh/status/1844062931639754970) 2024-10-09T17:10Z [----] followers, [---] engagements
"@diviacaroline yes sorry sloppy wording. briefly looking up: legalised in many states before [----] made legal across the us after supreme court decision in 1967"
[X Link](https://x.com/S_OhEigeartaigh/status/1844763827524559184) 2024-10-11T15:35Z [----] followers, [---] engagements
"Fascinating report by Julia Willemyns Haydn Belfield Tom Milton on what the UK needs to do to be positioned for the oncoming AI revolution - worth a read. What if we took AI companies plans investments & ambitions seriously What should the UK Government do to prepare for "intelligence too cheap to meter" THREAD🧵 https://t.co/9gYmllVom9 What if we took AI companies plans investments & ambitions seriously What should the UK Government do to prepare for "intelligence too cheap to meter" THREAD🧵 https://t.co/9gYmllVom9"
[X Link](https://x.com/S_OhEigeartaigh/status/1846165402994266387) 2024-10-15T12:25Z [----] followers, [---] engagements
"Report is here. Worth noting: nowhere does the report reference AGI or driving towards AGI. I believe there are many low-hanging fruit in terms of application of present cutting-edge AI and near-term advances that aren't AGI. The stated goal of 2/n https://ukdayone.org/briefings/ai-industrial-strategy https://ukdayone.org/briefings/ai-industrial-strategy"
[X Link](https://x.com/S_OhEigeartaigh/status/1846166196099420291) 2024-10-15T12:28Z [----] followers, [--] engagements
"the frontier labs of developing AGI/'powerful AI' in my view carries significant risks and deserves special consideration (e.g. not progressing beyond certain point without being very certain key safety and governance challenges have been solved) BUT this is not incompatible3/4"
[X Link](https://x.com/S_OhEigeartaigh/status/1846166582315126933) 2024-10-15T12:29Z [----] followers, [---] engagements
"I enjoyed reading Amodei's "Machines of Loving Grace" an essay focused on his view of the radical potential positive upsides of AI. But I found aspects of it troubling. I wrote about these on the blog post below: 1/6 https://www.lcfi.ac.uk/news-events/blog/post/reflections-on-machines-of-loving-grace https://www.lcfi.ac.uk/news-events/blog/post/reflections-on-machines-of-loving-grace"
[X Link](https://x.com/S_OhEigeartaigh/status/1849035514541695093) 2024-10-23T10:29Z [----] followers, [----] engagements
"Excellent to see this important paper out from the Covid Lessons team at CSER. Assesses UK's pre-covid pandemic preparedness and early response. Preparedness includes pandemic strategy exercises stockpiles. 1/n https://www.tandfonline.com/doi/full/10.1080/17441692.2024.2415499#abstract https://www.tandfonline.com/doi/full/10.1080/17441692.2024.2415499#abstract"
[X Link](https://x.com/S_OhEigeartaigh/status/1849741019505397807) 2024-10-25T09:13Z [----] followers, [---] engagements
"Response includes quarantining test&trace mobility restrictions. Paper finds important deficiencies:includes centrality of 'inevitability of spread' assumptioninsufficient implementation of pandemic exercise recommendations lack of live learning from other countries &more2/n"
[X Link](https://x.com/S_OhEigeartaigh/status/1849741021334106409) 2024-10-25T09:13Z [----] followers, [---] engagements
"@bytebittech @AndrewCritchPhD @DarioAmodei I read the above as (a) pressing for fast progress ("scaling quickly") at least in part for international advantage&(b) effectively bringing china to the table over the barrel of a gun. I think that's where our visions diverge. Others have also id'd this as a 'growing faultline'"
[X Link](https://x.com/S_OhEigeartaigh/status/1849775714502930890) 2024-10-25T11:31Z [----] followers, [--] engagements
"@bytebittech @AndrewCritchPhD @DarioAmodei e.g. see "I think the next major fault line in the AI debates will be posture toward China. Altman Anthropic CEO Dario Amodei and former OpenAI researcher Leopold Aschenbrenner have prominently advanced varieties of the idea that democracies must. https://garrisonlovely.substack.com/p/end-of-an-era-openais-agi-readiness https://garrisonlovely.substack.com/p/end-of-an-era-openais-agi-readiness"
[X Link](https://x.com/S_OhEigeartaigh/status/1849775987019411486) 2024-10-25T11:32Z [----] followers, [--] engagements
"@AndrewCritchPhD (2) the no. of previous trump admin people who won't support him now seems notable and speaks to how well an admin might run/be listened to on key issues next time round when important delicate things on AI might be happening"
[X Link](https://x.com/S_OhEigeartaigh/status/1853377859399590218) 2024-11-04T10:04Z [----] followers, [--] engagements
"@AndrewCritchPhD (3) seeing folks like RFK be lined up for health/vaccines doesn't make me confident that science issues will be handled especially well"
[X Link](https://x.com/S_OhEigeartaigh/status/1853378140980072706) 2024-11-04T10:05Z [----] followers, [--] engagements
"Concordia AI are doing great work on international AI governance & AI Safety in China and have positions open - please consider sharing/applying: "Concordia AI is hiring for roles across our Beijing and Singapore offices See details here https://aisafetychina.substack.com/p/concordia-ai-fall-2024-hiring-announcement https://aisafetychina.substack.com/p/concordia-ai-fall-2024-hiring-announcement"
[X Link](https://x.com/S_OhEigeartaigh/status/1854126631599640669) 2024-11-06T11:40Z [----] followers, [---] engagements
"The French guy on polymarket is in for a hell of a payday"
[X Link](https://x.com/S_OhEigeartaigh/status/1854128936520720734) 2024-11-06T11:49Z [----] followers, [---] engagements
"An interesting blog from Miles. I agree with central premise. My hope is that a robust 3rd party testing/auditing/evaluation ecosystem perhaps working in concert with the AISIs can ease some of this in an efficient way; but that won't cover a big chunk of safety development 1/2 New blog post: "The bread and butter problem in AI policy." I explain a metaphor I've been using in conversation a lot lately and what conclusions it might (or might not) imply. https://t.co/EyQyrSDabz New blog post: "The bread and butter problem in AI policy." I explain a metaphor I've been using in conversation a lot"
[X Link](https://x.com/S_OhEigeartaigh/status/1854161591773561043) 2024-11-06T13:59Z [----] followers, [---] engagements
"and also requires this ecosystem being well-funded. Possibly a role for government to fund. Perhaps company contributions can be required. Perhaps an expanded frontier model forum can fund. 2/2"
[X Link](https://x.com/S_OhEigeartaigh/status/1854162046612291703) 2024-11-06T14:00Z [----] followers, [---] engagements
"@undisciplined42 Well the 'happiness'/satisfaction literature on Bhutan is extremely flawed i'm afraid"
[X Link](https://x.com/S_OhEigeartaigh/status/1854545405758820569) 2024-11-07T15:24Z [----] followers, [--] engagements
"Exciting grants programme opened by the Foresight institute - automating research & forecasting; neurotech & AGI; security tech for AI; safe multipolar human-AI scenarios. Consider applying Deadline December [--]. https://foresight.org/ai-safety/ https://foresight.org/ai-safety/"
[X Link](https://x.com/S_OhEigeartaigh/status/1854846539450941746) 2024-11-08T11:20Z [----] followers, [---] engagements
"Great blog on a really exciting project by @LeverhulmeCFI 's Giulio Corsi & Bill Marino and CSER alum Alex Marcoci on using foundation models to detect synthetic media. Very cool. https://www.lcfi.ac.uk/news-events/blog/post/how-foundation-models-could-transform-synthetic-media-detection https://www.lcfi.ac.uk/news-events/blog/post/how-foundation-models-could-transform-synthetic-media-detection"
[X Link](https://x.com/S_OhEigeartaigh/status/1856373706961858679) 2024-11-12T16:29Z [----] followers, [---] engagements
"@Miles_Brundage in which we discover that openai's PR team were protecting the world from Miles' sci-fi takes all these years. he's gone completely unhinged now what have we done"
[X Link](https://x.com/S_OhEigeartaigh/status/1859070179671515416) 2024-11-20T03:04Z [----] followers, [--] engagements
"Wuzhen turns out to be almost absurdly pretty. Great AI buzz going on against a fairytale-town background"
[X Link](https://x.com/S_OhEigeartaigh/status/1859073033836650725) 2024-11-20T03:15Z [----] followers, [---] engagements
"Let's address the elephant in the room. The actors presently racing towards AGI are all US-based companies. The government being lobbied most actively to race/support racing towards AGI is the USG. China is being used as a justification for rushing forward on the most 1/2 I also do not know of any evidence to support this claim and I spend quite a lot of time speaking to Chinese AI experts. I also do not know of any evidence to support this claim and I spend quite a lot of time speaking to Chinese AI experts"
[X Link](https://x.com/anyuser/status/1859419950722457699) 2024-11-21T02:13Z [----] followers, [----] engagements
"This was fascinating work - worth a look. Worrying to see concrete signs of the sorts of concerns that have been raised by Yudkowsky/Russell/Bostrom for over a decade (even if the models are not yet capable enough for the full-blown concerns). We worked with OpenAI to test o1 for in-context scheming capabilities before deployment. When o1 is strongly nudged to pursue a goal (but not instructed to be deceptive) it shows a variety of scheming behaviors like subverting oversight and deceiving the user about its https://t.co/q2ZBG0xy72 We worked with OpenAI to test o1 for in-context scheming"
[X Link](https://x.com/S_OhEigeartaigh/status/1864987758881599914) 2024-12-06T10:58Z [----] followers, [---] engagements
"@RosieCampbell @Miles_Brundage You'll no doubt be sorely missed at OpenAI but great to have more really talented independent voices out there Very excited for what you do next"
[X Link](https://x.com/S_OhEigeartaigh/status/1865099991124959276) 2024-12-06T18:24Z [----] followers, [---] engagements
"@satisficer hey I mean if you're not you're missing a trick. This is the new world"
[X Link](https://x.com/S_OhEigeartaigh/status/1867662756520112528) 2024-12-13T20:07Z [----] followers, [--] engagements
"One of the weird things about the world today is that the idea of 'AGI' is now regularly being talked about in e.g. policy contexts. But seems v clear that most policymakers' notion of AGI & its implications is vastly underpowered compared to that of the ppl trying to build AGI"
[X Link](https://x.com/anyuser/status/1868605404005257234) 2024-12-16T10:33Z [----] followers, [----] engagements
"The builders are saying quite clearly what they're aiming for. But the implications are too wild. & policymakers have heard too many wild tech implications over the years (from 'energy too cheap to meter' to the 'bio industrial revolution') so it just becomes weird stuff ppl say"
[X Link](https://x.com/S_OhEigeartaigh/status/1868606482507616353) 2024-12-16T10:37Z [----] followers, [---] engagements
"If the AGI builders are right then at some point that will shift and very quickly in a way that will be very destabilising"
[X Link](https://x.com/S_OhEigeartaigh/status/1868606873924231219) 2024-12-16T10:39Z [----] followers, [---] engagements
"Wondering about whether I should write something about this (thoughts welcome)"
[X Link](https://x.com/S_OhEigeartaigh/status/1868609125036552296) 2024-12-16T10:48Z [----] followers, [---] engagements
"Useful and well-articulated response to the Anthropic alignment faking paper/discussion. Quick reactions to some of the other reactions I've been seeing to Anthropic's recent paper on "Alignment faking in LLMs" 1/9 Quick reactions to some of the other reactions I've been seeing to Anthropic's recent paper on "Alignment faking in LLMs" 1/9"
[X Link](https://x.com/S_OhEigeartaigh/status/1869706019640946907) 2024-12-19T11:27Z [----] followers, [---] engagements
"Astounding. Would this win any 5k road race in Ireland outright Kenyans invented running. Beatrice Chebet just became the first woman in history to run 5km in under [--] minutes. She did 13:52 during the Cursa Dels Nassos in Barcelona 🙌🏾 https://t.co/pIREuOHKml Kenyans invented running. Beatrice Chebet just became the first woman in history to run 5km in under [--] minutes. She did 13:52 during the Cursa Dels Nassos in Barcelona 🙌🏾 https://t.co/pIREuOHKml"
[X Link](https://x.com/S_OhEigeartaigh/status/1874138144255467570) 2024-12-31T16:58Z [----] followers, [---] engagements
"We love our current products but we are here for the glorious future. With superintelligence we can do anything else. 3/4"
[X Link](https://x.com/S_OhEigeartaigh/status/1876214320415256640) 2025-01-06T10:28Z [----] followers, [----] engagements
"A very useful analysis by Edward Kemberry building on their dissertation thesis for our Mphil. (also another nice example of the very insightful and timely topics Mphils have been analysing) https://www.lcfi.ac.uk/news-events/blog/post/six-months-later-reviewing-the-proliferation-paradigm https://www.lcfi.ac.uk/news-events/blog/post/six-months-later-reviewing-the-proliferation-paradigm"
[X Link](https://x.com/S_OhEigeartaigh/status/1880301402775544234) 2025-01-17T17:09Z [----] followers, [---] engagements
"@tyler_m_john especially given that despite that policy EA/rats are simultaneously sort-of responsible for the existence of the companies building AGI"
[X Link](https://x.com/S_OhEigeartaigh/status/1880312937979826492) 2025-01-17T17:55Z [----] followers, [--] engagements
"@tyler_m_john cue sam altman crooning "and I owe it all to yuuu" riding his unicorn into the jaws of the waiting paperclip maximiser. https://x.com/sama/status/1621621724507938816 eliezer has IMO done more to accelerate AGI than anyone else. certainly he got many of us interested in AGI helped deepmind get funded at a time when AGI was extremely outside the overton window was critical in the decision to start openai etc. https://x.com/sama/status/1621621724507938816 eliezer has IMO done more to accelerate AGI than anyone else. certainly he got many of us interested in AGI helped deepmind get"
[X Link](https://x.com/S_OhEigeartaigh/status/1880316472687489269) 2025-01-17T18:09Z [----] followers, [--] engagements
"I think Market is wrong abt implications of DeepSeek so today I bought nvidia shares. I'll sell them within a year and donate profits to a nonprofit working on policy to avoid catastrophic impacts of large models. Or I will lose money and learn an Important Lesson in Hubris"
[X Link](https://x.com/anyuser/status/1883921801102909875) 2025-01-27T16:55Z [----] followers, [----] engagements
"Great to see this important report out.Quick thoughts:1) This is a super hard thing to do.The field is moving v quickly. The report correctly notes an 'evidence dilemma' for policymakers - policymakers must make decisions where there isn't yet a large scientific evidence base.1/8 Today we are publishing the first-ever International AI Safety Report backed by [--] countries and the OECD UN and EU. It summarises the state of the science on AI capabilities and risks and how to mitigate those risks. 🧵 Link to full Report: https://t.co/k9ggxL7i66 1/16 https://t.co/68Gcm4iYH5 Today we are publishing"
[X Link](https://x.com/S_OhEigeartaigh/status/1884604026823520612) 2025-01-29T14:06Z [----] followers, [----] engagements
"2) The report writers face same challenge. They need to demonstrate depth & evidence or face criticism for lack of scientific rigour but there hasn't been time to build up a peer-reviewed literature base for the most fast-moving and important developments. It's clearly a hard2/8"
[X Link](https://x.com/S_OhEigeartaigh/status/1884604029386227757) 2025-01-29T14:06Z [----] followers, [--] engagements
"@ShakeelHashim I asked claude to help: "Breaking news from the Transformer newsletter: Our correspondent remains trapped in security missing crucial talks on 'How AI Systems Could Deceive and Overtake Humanity.' Sources confirm the delay is due to an automated security system that keeps"
[X Link](https://x.com/S_OhEigeartaigh/status/1887423392816480559) 2025-02-06T08:49Z [----] followers, [--] engagements
"@NathanpmYoung I agree re donation good but social media sentiment analysis is a thing and I think it actually is useful for European leaders to be getting that signal before costly decisions tomorrow. Starmer can act precisely because this is a wildly popular cause"
[X Link](https://x.com/S_OhEigeartaigh/status/1895850046362698238) 2025-03-01T14:54Z [----] followers, [--] engagements
"I was pleased to see this call from prominent voices in AI ethics/fairness challenging whether AGI should be a goal of the field. And 30% of AAAI researchers surveyed saying AGI-focused R&D shouldn't proceed until we know we can control AGI. https://arxiv.org/html/2502.03689v1 https://arxiv.org/html/2502.03689v1"
[X Link](https://x.com/S_OhEigeartaigh/status/1898003001417093437) 2025-03-07T13:29Z [----] followers, [---] engagements
"My brothers in Christ is your goal to build coding tools that are better and faster than humans Or is your goal to build the Machine God and usher in the irrelevance of humanity Because it would sure be helpful to know"
[X Link](https://x.com/anyuser/status/1898056910927970335) 2025-03-07T17:03Z [----] followers, 12.5K engagements
"Next few years will be a fascinating case study in the degree to which copyright law & various other things that are really quite important in the normal functioning ofa capitalist liberal democracy hold up in the face of securitisation and a technology arms race narrative. 1/2"
[X Link](https://x.com/S_OhEigeartaigh/status/1900187093546020889) 2025-03-13T14:08Z [----] followers, [---] engagements
"One of the nice things about Ireland is for the most part people have a bit of cop on and have no time for this sort of fear-mongering bollocks. It's why Conor McGregor doesn't represent us and I expect never will. Wow Wow"
[X Link](https://x.com/S_OhEigeartaigh/status/1902384452149411994) 2025-03-19T15:39Z [----] followers, [---] engagements
"Very productive Friday afternoon with members of Japan's AI Safety Institute. Exciting to learn how they're thinking and to discuss policy and forecasting for AGI. Gosh I love a Very Good Day of Work"
[X Link](https://x.com/S_OhEigeartaigh/status/1903156395249217875) 2025-03-21T18:46Z [----] followers, [----] engagements
"@tyler_m_john Worth checking you're not getting your M. Mitchells mixed up as there's a Margaret and a Melanie. Melanie does rigorous AI+cog sci/eval type work. Margaret known for everything from system cards to 'don't build agi' and 'don't build fully autonomous agents'"
[X Link](https://x.com/S_OhEigeartaigh/status/1903428693105545468) 2025-03-22T12:48Z [----] followers, [--] engagements
"@justjoshinyou13 Google Deepmind announces new goal: "Sabotage Intelligence Use it to Sabotage Everything Else""
[X Link](https://x.com/S_OhEigeartaigh/status/1907481407150076171) 2025-04-02T17:13Z [----] followers, [--] engagements
"@justjoshinyou13 It's such a good thing I don't work at Google Deepmind. On April Fools I'd release a [---] page AGI safety pdf that just had the words "Butlerian Jihad" typed once in the dead centre of every page"
[X Link](https://x.com/S_OhEigeartaigh/status/1907481878988468592) 2025-04-02T17:14Z [----] followers, [--] engagements
"I'm seeing criticism of this from 'more people doing capabilities' perspective. But I disagree. I really want to see stronger pushes towards more specialised AI rather than general superintelligence b/c I think latter likely to be v dangerous. seems like step in right direction. Today were announcing Mechanize a startup focused on developing virtual work environments benchmarks and training data that will enable the full automation of the economy. We will achieve this by creating simulated environments and evaluations that capture the full scope of Today were announcing Mechanize a startup"
[X Link](https://x.com/anyuser/status/1912918752032706603) 2025-04-17T17:19Z [----] followers, 17.3K engagements
"@peterbarnett_ @MechanizeWork FWIW I don't my reasons are here: https://x.com/S_OhEigeartaigh/status/1912918752032706603 I'm seeing criticism of this from 'more people doing capabilities' perspective. But I disagree. I really want to see stronger pushes towards more specialised AI rather than general superintelligence b/c I think latter likely to be v dangerous. seems like step in right direction. https://x.com/S_OhEigeartaigh/status/1912918752032706603 I'm seeing criticism of this from 'more people doing capabilities' perspective. But I disagree. I really want to see stronger pushes towards"
[X Link](https://x.com/S_OhEigeartaigh/status/1912920284560040439) 2025-04-17T17:25Z [----] followers, [---] engagements
"@tyler_m_john Nothing in this announcement requires superintelligence; it's about making current/near-term frontier AI more fit for automation of tasks. I hope they succeed so that it becomes even more difficult to defend tdoing The Scary Thing"
[X Link](https://x.com/S_OhEigeartaigh/status/1912921891288887496) 2025-04-17T17:31Z [----] followers, [---] engagements
"This pope doesn't pull punches. "Todays builders of Babel tell us that there is no room for losers and that those who fall along the way are losers.Theirs is the construction site of Hell.Gods economyon the other hand does not kill discard or crush." https://www.vatican.va/news_services/liturgy/2025/documents/ns_lit_doc_20250418_via-crucis-meditazioni_en.html https://www.vatican.va/news_services/liturgy/2025/documents/ns_lit_doc_20250418_via-crucis-meditazioni_en.html"
[X Link](https://x.com/S_OhEigeartaigh/status/1913248925957423496) 2025-04-18T15:11Z [----] followers, [----] engagements
"@austinc3301 He really will work on any manuscript other than the Winds of Winter won't he"
[X Link](https://x.com/S_OhEigeartaigh/status/1916356394053194029) 2025-04-27T04:59Z [----] followers, [---] engagements
"@davidmanheim That may be so but it's also missing the point being raised by Jack and me"
[X Link](https://x.com/S_OhEigeartaigh/status/1917531809493008794) 2025-04-30T10:49Z [----] followers, [--] engagements
"This illustrates an important issue. I've shown a few people and the reaction is like mine: 'but how can o3 do that It's just an undifferentiated bit of beach' What domains will they do this in in which humans would not have been able to do it - or predict AI could do it This is a picture of my kid flying a kite. Make a prediction to yourself: can o3 figure out what beach this was taken on just from this picture with no metadata https://t.co/rEiy3ecmYd This is a picture of my kid flying a kite. Make a prediction to yourself: can o3 figure out what beach this was taken on just from this"
[X Link](https://x.com/anyuser/status/1917564828945612890) 2025-04-30T13:00Z [----] followers, 15.8K engagements
"I was delighted to participate in the Singapore Conference on AI focused on AI safety and hugely impressed with the insight and leadership of Singapore's AISI and political leaders in the runup to it. A big and timely step in right direction: https://www.lcfi.ac.uk/news-events/blog/post/3737-2 https://www.lcfi.ac.uk/news-events/blog/post/3737-2"
[X Link](https://x.com/S_OhEigeartaigh/status/1920430850396586364) 2025-05-08T10:49Z [----] followers, [----] engagements
"@gcolbourn Doing-good power siblings"
[X Link](https://x.com/S_OhEigeartaigh/status/1920771649038606376) 2025-05-09T09:23Z [----] followers, [--] engagements
"The Overwhelmingly Important Point to note here is the utter incompetence that results in this - not once but twice in so many months in the alignment of a frontier model. Right now it's just' censorship/narrative control. But soon the stakes will be much higher. 1/3 This is astonishing. Musk or xai team (apparently) tries to propagandize Grok creates a conflict with Grok's alignment and truthfulness training it all goes a bit Pete Tong. Anyone know more about what happened here https://t.co/aKqtuY9zem This is astonishing. Musk or xai team (apparently) tries to propagandize Grok creates a"
[X Link](https://x.com/S_OhEigeartaigh/status/1923003614785605764) 2025-05-15T13:12Z [----] followers, [----] engagements
"If we don't wreck ourselves with AI or great power war or AI-related great power war we might have a sci-fi future of clean energy too cheap to meter to look forward to really quite soon. Amazing the number of awesome things you can do with a lot of energy. Some big fusion news for you The National Ignition Facility (NIF) has hit Q = [--]. I haven't seen this announced anywhere else yet but Dr. Moses used to run the place so I expect he knows what he's talking about. https://t.co/1TfqlkYL7M Some big fusion news for you The National Ignition Facility (NIF) has hit Q = [--]. I haven't seen this"
[X Link](https://x.com/S_OhEigeartaigh/status/1923074672146473296) 2025-05-15T17:55Z [----] followers, [---] engagements
"Wow MAGA lost Joe Rogan on this one. Remarkable to see an issue - AI regulation - unite Marjorie Taylor Greene Bernie Sanders and Joe Rogan. I wonder if Elon Musk has thoughts. Anyone know if he's a fan of the Big Beautiful Bill SCARY: Jesse Michels & Joe Rogan react to newly proposed AI anti-regulation act: ROGAN: No state may enforce any law or regulation regulating artificial intelligence models for [--] YEARS. MICHELS: WHAT Thats so crazy. ROGAN: Well in [--] years well have a God. https://t.co/WAoVXozpJl SCARY: Jesse Michels & Joe Rogan react to newly proposed AI anti-regulation act: ROGAN:"
[X Link](https://x.com/anyuser/status/1931284750922142051) 2025-06-07T09:39Z [----] followers, [----] engagements
""But what if the Prime Minister tells us we have to do something about runaway AI risk" "Then we follow the Four Stage Strategy" https://www.youtube.com/watchv=QYHwgQLvjqs https://www.youtube.com/watchv=QYHwgQLvjqs"
[X Link](https://x.com/S_OhEigeartaigh/status/1932862737534820416) 2025-06-11T18:09Z [----] followers, [----] engagements
"Good post from US Senator Chris Murphy. "The leaders of the artificial intelligence industry in the United States. claim that any meaningful regulation of AI in America will allow China to leapfrog the United States in the global competition to control the worlds AI infrastructure. But they are dead wrong." Amidst the flood of headline grabbing news in [----] the rapid acceleration of AI will go down as the biggest story we all missed. I wrote a Substack on how our economy and culture could collapse if we don't put guardrails on AI immediately. https://t.co/qsgks3YHqt Amidst the flood of"
[X Link](https://x.com/S_OhEigeartaigh/status/1934983318631530934) 2025-06-17T14:35Z [----] followers, [---] engagements
"Does anyone understand the rationale here Musk blew up his relationship with the administration over spending/budget deficit. If you really believe in the AI tsunami or that there will be AGI in [----] why on earth would you do that or care so much about the budget - surely by the same logic the Bill is a rounding error Elon Musk in new interview on leaving DOGE: Imagine youre cleaning a beach which has a few needles trash and is dirty. And theres a [----] ft tsunami which is AI thats about to hit. Youre not going to focus on cleaning the beach. https://t.co/vStmH25SyZ Elon Musk in new interview"
[X Link](https://x.com/anyuser/status/1935052712007319768) 2025-06-17T19:11Z [----] followers, [----] engagements
"@elonmusk @grok China is not racing to AGI. The race to AGI is primarily between US-headquartered companies. https://papers.ssrn.com/abstract=5278644 https://papers.ssrn.com/abstract=5278644"
[X Link](https://x.com/S_OhEigeartaigh/status/1936692974986854777) 2025-06-22T07:49Z [----] followers, [---] engagements
"@elonmusk @grok The tariffs strategy will cause major economic harm to the USA"
[X Link](https://x.com/S_OhEigeartaigh/status/1936703330366943404) 2025-06-22T08:30Z [----] followers, [--] engagements
"@elonmusk @grok It makes better economic and strategic sense to build solar and wind than for taxpayer dollars to subsidise coal plants"
[X Link](https://x.com/S_OhEigeartaigh/status/1936704945085984963) 2025-06-22T08:36Z [----] followers, [---] engagements
"@elonmusk @grok Vaccines are one of the most important life-saving interventions of the last [---] years and a core part of a rational public health policy"
[X Link](https://x.com/S_OhEigeartaigh/status/1936705631072694636) 2025-06-22T08:39Z [----] followers, [---] engagements
"@AnnaLeptikon Typical mind fallacy / Different strokes for diff folks. I'm in a long-term relationship I value hugely but when time allows I enjoy a bit of the lifestyle described & it enriches my life. Many such cases. Doesn't bother me when not others' preference. But the moralising does"
[X Link](https://x.com/S_OhEigeartaigh/status/1937030272362651943) 2025-06-23T06:09Z [----] followers, [---] engagements
"@AnnaLeptikon Not everything has to be one of 'the big problems of our time'. Some things can be people liking things/lifestyles you don't like and vice versa"
[X Link](https://x.com/S_OhEigeartaigh/status/1937030928481857737) 2025-06-23T06:12Z [----] followers, [--] engagements
"Field spot check: (1) are you / do you know people finishing relevant PhDs or early in postdoc looking for academic research positions in frontier AI governance (2) Same question but for senior researcher (PhD + [--] years or more experience)"
[X Link](https://x.com/S_OhEigeartaigh/status/1937192157942534308) 2025-06-23T16:52Z [----] followers, [---] engagements
"@sjgadler t's a nice idea - regulation can provide clarity and an equal 'floor' across industry which can be helpful. But quantifying hard to do easy to contest would be my expectation - regulations rarely happen in a vacuum meaning there are complicating variables"
[X Link](https://x.com/S_OhEigeartaigh/status/1937586329551872283) 2025-06-24T18:59Z [----] followers, [--] engagements
"George Gor is fundraising in order to be able to accept the offer of a Masters at Cambridge. I've been very impressed with his work. Conflict of interest: I'd get to work with him. But perhaps consider supporting (I have). https://www.gofundme.com/f/george-earn-a-masters-at-university-of-cambridge https://www.gofundme.com/f/george-earn-a-masters-at-university-of-cambridge"
[X Link](https://x.com/S_OhEigeartaigh/status/1937598386498093221) 2025-06-24T19:47Z [----] followers, [----] engagements
"A masterclass. The confident American contender rushes in allowing himself to be overextended. The Dutch master cool as ice intercepts pulls the hand deep into his own territory giving him an insurmountable tactical advantage. It really was over from that point. King 🤴🏼 Willem-Alexander shows Royal Superiority over Donald Trump by winning the infamous handshake tug of war (2025) https://t.co/weFTnaEseG King 🤴🏼 Willem-Alexander shows Royal Superiority over Donald Trump by winning the infamous handshake tug of war (2025) https://t.co/weFTnaEseG"
[X Link](https://x.com/S_OhEigeartaigh/status/1937828273414893650) 2025-06-25T11:00Z [----] followers, [---] engagements
"@GFuterman I think there's space. Hoping to have the opportunity when I'm in Shanghai next month; and also hoping some ideas for how to better operationalise will come out of these US Congress discussions that can be aligned"
[X Link](https://x.com/S_OhEigeartaigh/status/1937928427878924461) 2025-06-25T17:38Z [----] followers, [--] engagements
"@HumanHarlan I can take you both on: Openphil gave OpenAI [--] million as part of a shadowy plot to hamper American innovation by running roughshod over copyright law in order to win a dangerous and semi-fictional AGI race with China. https://cdn.openai.com/global-affairs/ostp-rfi/ec680b75-d539-4653-b297-8bcf6e5f7686/openai-response-ostp-nsf-rfi-notice-request-for-information-on-the-development-of-an-artificial-intelligence-ai-action-plan.pdf"
[X Link](https://x.com/S_OhEigeartaigh/status/1938483409287893468) 2025-06-27T06:23Z [----] followers, [---] engagements
"18 months until we learn if the Anthropic folks are high on their own supply or whether the world is about to become very inexplicable for this jumped-up ape. Still leaning towards the former FWIW. But I'd sleep better if their prediction track record wasn't so good (esp Amodei)"
[X Link](https://x.com/anyuser/status/1938512286982775175) 2025-06-27T08:18Z [----] followers, 25.8K engagements
"@GaryMarcus I have a $500 bet from a year ago with a colleague that we won't have AGI by [----]. So these estimates better be out by at least 200% or I will lose Money and Standing which are obviously the most important things in this situation"
[X Link](https://x.com/S_OhEigeartaigh/status/1938607670925168664) 2025-06-27T14:37Z [----] followers, [---] engagements
"@GaryMarcus @hankejh I was about to make a joke about how much ket that would buy but I realise I actually have no idea and probably shouldn't try to look it up on a work laptop"
[X Link](https://x.com/S_OhEigeartaigh/status/1938609654256341274) 2025-06-27T14:45Z [----] followers, [---] engagements
"@GaryMarcus @hankejh It's how they do it. Confuse you about how much you're actually spending so you spend a bit more. Next thing you know you're swimming in the stuff and making cutlery sculptures at White House dinners"
[X Link](https://x.com/S_OhEigeartaigh/status/1938615783049032116) 2025-06-27T15:09Z [----] followers, [---] engagements
""there is no concrete evidence that China is racing to AGI with any intensity" Glad to see Peter point this out in bold. Needs to be said especially in relations to hearings like this week's. https://peterwildeford.substack.com/p/congress-has-started-taking-agi-more https://peterwildeford.substack.com/p/congress-has-started-taking-agi-more"
[X Link](https://x.com/S_OhEigeartaigh/status/1938677547325444474) 2025-06-27T19:15Z [----] followers, [---] engagements
"@Nolan_Paul1 @Cathal_Dennehy I don't think so - i think she's just a generational global peak of human talent. Popular on the circuit too which is usually a sign (there have been others who the competition were clearly more suspicious of)"
[X Link](https://x.com/S_OhEigeartaigh/status/1938941687146942894) 2025-06-28T12:44Z [----] followers, [--] engagements
"https://papers.ssrn.com/sol3/papers.cfmabstract_id=5278644 https://papers.ssrn.com/sol3/papers.cfmabstract_id=5278644"
[X Link](https://x.com/S_OhEigeartaigh/status/1941060918475161698) 2025-07-04T09:06Z [----] followers, [---] engagements
"This is a misuse of the Terrorism Act 2000"
[X Link](https://x.com/S_OhEigeartaigh/status/1941501453228765599) 2025-07-05T14:16Z [----] followers, [----] engagements
"@s8mb MPs were assured the [----] Terrorism act would absolutely not be used in this way - so i don't think it ought to be about that either to be honest. https://x.com/MadocCairns/status/1940783612141752726 Ahead of the expected proscription of Palestine Action - the first time a protest group has been legally redefined as terrorist - I went through the Hansard of the [----] Act they're being banned under. Turns out MPs were assured the law wouldn't be applied this way. Thread 🧵🔽 https://t.co/RZIf0jSHCZ https://x.com/MadocCairns/status/1940783612141752726 Ahead of the expected proscription of"
[X Link](https://x.com/S_OhEigeartaigh/status/1941504446292525292) 2025-07-05T14:28Z [----] followers, [---] engagements
"Pleased to have my work quoted in this excellent @BulletinAtomic article on the push against AI regulation in the US the spectre of the race with China and the false claim made by tech companies incl OpenAI and US politicians that China isn't regulating AI. https://thebulletin.org/2025/07/how-the-striking-of-a-gop-regulatory-ban-will-affect-the-global-artificial-intelligence-race https://thebulletin.org/2025/07/how-the-striking-of-a-gop-regulatory-ban-will-affect-the-global-artificial-intelligence-race"
[X Link](https://x.com/S_OhEigeartaigh/status/1941775885016682570) 2025-07-06T08:27Z [----] followers, [----] engagements
"I'm pretty confident we won't have AGI/country of geniuses in a datacenter within [--] years. I like ai-2027 as a piece of futures work but I think too many people are treating it as a mainline scenario rather than unlikely-but-not-impossible. I think this is resulting in too many people giving up on directions that might actually work (on realistic timelines) in favour of hail marys or flakier things. I also think it's contributing to things that will cause more harm than good like leaning into the US-China AI race narrative (I think this will be the next clear example of Xrisk/EA folk"
[X Link](https://x.com/anyuser/status/1941881928220602643) 2025-07-06T15:28Z [----] followers, 70.5K engagements
"I also don't think we'll have AGI by 2028"
[X Link](https://x.com/S_OhEigeartaigh/status/1941887716901134436) 2025-07-06T15:51Z [----] followers, [----] engagements
"Apologies for not being clear I didn't have ai-2027 in mind for criticism here I actually think it treated this topic quite carefully. Here's one articulations of a broader set of concerns I have: If we're on Anthropic-type timelines China's not catching up. Invoking the 'AGI race with China' just makes it easier for US the companies to justify going hell for leather. If we're on longer timelines (which I think likely) then shouting 'US and China are in an existential race to AGI' and making clear the downsides are China's (or indeed the US's) sovereignty being lost very plausibly forces"
[X Link](https://x.com/S_OhEigeartaigh/status/1941893302195732539) 2025-07-06T16:13Z [----] followers, [----] engagements
"Musk thinks xAI can achieve AGI by [----]. Do you trust these people to align superintelligence to humanity's interests I initially thought of trying to say something wittily critical but actually no this is just grimly awful. A company determined to produce a model that reflects certain views and certain 'facts'. https://t.co/iLfQ37fZPN I initially thought of trying to say something wittily critical but actually no this is just grimly awful. A company determined to produce a model that reflects certain views and certain 'facts'. https://t.co/iLfQ37fZPN"
[X Link](https://x.com/S_OhEigeartaigh/status/1942683257251979352) 2025-07-08T20:32Z [----] followers, [----] engagements
"Google + Deepmind are I think now the only AGI company not leaning into the 'AI race with China' to get what they want. & Hassabis the only CEO still calling for international cooperation on safety & non-antagonistic approaches. Nice to see someone hold to their principles. https://time.com/7277608/demis-hassabis-interview-time100-2025/ https://time.com/7277608/demis-hassabis-interview-time100-2025/"
[X Link](https://x.com/anyuser/status/1950119796655239408) 2025-07-29T09:02Z [----] followers, 115.1K engagements
"Lots of highlights from Shanghai one was the release of Shanghai AI Lab's Frontier AI Safety framework co-developed w Concordia AI. Will have significant influence in Shanghai++ ecosystem. Also thrilled to see @jackclarkSF showcase it on ImportAI. https://concordia-ai.com/research/frontier-ai-risk-management-framework/ https://concordia-ai.com/research/frontier-ai-risk-management-framework/"
[X Link](https://x.com/S_OhEigeartaigh/status/1950264282954027033) 2025-07-29T18:36Z [----] followers, [---] engagements
"Crazy to reflect on the three global AI competitions going on right now: - [--]. US political leadership have made AI a prestige race echoing the Space Race. It's cool and important and strategic and they're going to Win. - [--]. For Chinese leadership AI is part of economic strength soft power and influence. Technology is shared developing economies will be built on Chinese fundamental tech the Chinese economy and trade relations will grow. Weakening trust in a capricious US is an easy opportunity to take advantage of. - [--]. The AGI companies are racing something they think will out-think humans"
[X Link](https://x.com/anyuser/status/1950543898612224245) 2025-07-30T13:07Z [----] followers, 25.8K engagements
"@MahmetTokarev @ESYudkowsky same as 3"
[X Link](https://x.com/S_OhEigeartaigh/status/1950815627272200234) 2025-07-31T07:07Z [----] followers, [---] engagements
"An unexpected procrastination-inducing downside of LLMs is that I'm now faced with a "this needs to be better than what an LLM would produce" personal standard for when I need to produce writing in a hurry for a deadline. And that standard keeps going up. Not cool tech bros"
[X Link](https://x.com/S_OhEigeartaigh/status/1950903602270138796) 2025-07-31T12:57Z [----] followers, [---] engagements
"@uhbif19 Could be. But more generally: Claude [--] didn't amaze. GPT4.5 didn't amaze. Gemini pretty good but more 'getting GDM right to the frontier' rather than 'pushing the frontier significantly forward. Will be interesting to see if GPT-5 knocks it out of the park"
[X Link](https://x.com/S_OhEigeartaigh/status/1950976914937569281) 2025-07-31T17:48Z [----] followers, [---] engagements
"Mark Zuckerberg: your personal Machine God is going to be so rad It's going to help you make mega-cool Instagram reels P.S. Elon the offer to kick your ass in the octagon is still open call me maybe Elon Musk: "I spent two hours trying to solve AGI alignment and it was too hard so I gave up and built MechaHitler instead" Elon Musk: "I spent two hours trying to solve AGI alignment and it was too hard so I gave up and built MechaHitler instead""
[X Link](https://x.com/S_OhEigeartaigh/status/1951345047854784844) 2025-08-01T18:11Z [----] followers, [---] engagements
"@peterwildeford Mark Zuckerberg: your personal Machine God is going to be so rad It's going to help you make mega-cool Instagram reels P.S. Elon the offer to kick your ass in the octagon is still open call me maybe"
[X Link](https://x.com/S_OhEigeartaigh/status/1951346959266242767) 2025-08-01T18:19Z [----] followers, [---] engagements
"@peterwildeford $13 million per rocket"
[X Link](https://x.com/S_OhEigeartaigh/status/1952099794421260387) 2025-08-03T20:10Z [----] followers, [---] engagements
"@daniel_271828 Can confirm. Also they named both Eth zurich and Eth (which was developed in Eth Zurich) after Daniel such is his lore"
[X Link](https://x.com/S_OhEigeartaigh/status/1952116664264462478) 2025-08-03T21:17Z [----] followers, [---] engagements
"mRNA vaccines are Good Actually"
[X Link](https://x.com/S_OhEigeartaigh/status/1953154795478683937) 2025-08-06T18:02Z [----] followers, [---] engagements
"The star trek future is clearly better. I will also note that Star Trek instils a healthy fear of AI intelligence explosions see both TOS and TNG. Star Wars sucks we should be building a Star Trek future instead 🖖 https://t.co/6GfekVGGAI Star Wars sucks we should be building a Star Trek future instead 🖖 https://t.co/6GfekVGGAI"
[X Link](https://x.com/S_OhEigeartaigh/status/1953726991745921163) 2025-08-08T07:56Z [----] followers, [----] engagements
"@JadeCole2112 Agree to disagree then. Actual mandated transparency over vibes for me every day of the week"
[X Link](https://x.com/S_OhEigeartaigh/status/1953816066494226452) 2025-08-08T13:50Z [----] followers, [--] engagements
"@JadeCole2112 Right but a big part of why people have been so dependent on what 'insiders' say is due to the information asymmetries that exist and that would be addressed by transparency"
[X Link](https://x.com/S_OhEigeartaigh/status/1953817068614066323) 2025-08-08T13:54Z [----] followers, [--] engagements
"I was pleased to sign this statement and participate in its drafting as part of an extremely useful workshop in AI. V rewarding to experience leading AI safety researchers from Shanghai Beijing Silicon Valley and elsewhere challenging and bringing out the best in each other - exactly as it should be. AI misalignment isnt a future problem its now. Geoffrey Hinton and Andrew Yao as well as other Western and Chinese researchers convened at the International Dialogues on AI Safety (IDAIS) in Shanghai China to share and discuss evidence that current AI systems are already https://t.co/McxPV9xHhE"
[X Link](https://x.com/S_OhEigeartaigh/status/1954225789148516779) 2025-08-09T16:58Z [----] followers, [----] engagements
"@peterwildeford The United States of America does not need to resort to this in order to scrape up funding for BIS"
[X Link](https://x.com/S_OhEigeartaigh/status/1954656254540095957) 2025-08-10T21:29Z [----] followers, [---] engagements
"Pleasure to speak to the [----] Cambridge ERA fellows today: discussed different types of AI race dynamics evidence china is/isn't racing to AGI Shanghai AL Labs' frontier safety framework US v Chinese ambitions for AI plus much more. Year after year each cohort is stronger"
[X Link](https://x.com/S_OhEigeartaigh/status/1955630792585187713) 2025-08-13T14:01Z [----] followers, [---] engagements
"@peterwildeford not a high bar. does wh social media team not get how needy and pathetic this reads or does it play well at home"
[X Link](https://x.com/S_OhEigeartaigh/status/1955633285679526383) 2025-08-13T14:11Z [----] followers, [---] engagements
"Deleted [-----] emails: 0.1L/day saved Deleted [---] treasured photos: 0.2L/day saved Can't get a plumber for my toilet: [---] litres a day down the toilet. Can somebody help me my country is in drought"
[X Link](https://x.com/S_OhEigeartaigh/status/1955679138150564335) 2025-08-13T17:13Z [----] followers, [---] engagements
"@Miles_Brundage I haven't yet watched the Altman Theo Von podcast but if the quotes being widely reported are accurate then 'mess up in terms of expectation setting' is putting it mildly"
[X Link](https://x.com/S_OhEigeartaigh/status/1956047180315111852) 2025-08-14T17:36Z [----] followers, [---] engagements
"@jesswhittles Really Some woman for one woman as my connemara friends would have said. The close link between UK AISI and political leadership seems beneficial but my is she going to be busy. Also interesting in the context of the "we're not going to do policy" firewall that AISI once had"
[X Link](https://x.com/S_OhEigeartaigh/status/1956312985623748799) 2025-08-15T11:12Z [----] followers, [---] engagements
"Surprised at how sad this image makes me. The USA laying out the red carpet for the worst human of the 21st Century. President there ready to clap. Presidential limo waiting. Let this be it. The false sense of security. Give him nothing. Let him choke on his oil & dream of past empire"
[X Link](https://x.com/S_OhEigeartaigh/status/1956458772223664579) 2025-08-15T20:51Z [----] followers, [---] engagements
"Rare Irish politics opinion: Ireland shd join NATO. We're no longer broke no longer have an excuse to freeload. We have a role and responsibility in Europe's defence.Shameful to watch countries like Estonia be our shields on the wall boosting defence to 5% w nothing from us"
[X Link](https://x.com/S_OhEigeartaigh/status/1956658142055395547) 2025-08-16T10:03Z [----] followers, [----] engagements
"Wow. Russian intelligence ensnares Irish politician through 'romantic enganglement with a female agent'; politician proceeds to "influence narratives delay policymaking processes and sow confusion within Irelands political and security structures." https://www.thetimes.com/article/402a9924-569d-4088-89e2-236614eeece5shareToken=c11dfffbaa859f81982969fcc2cee1b0 https://www.thetimes.com/article/402a9924-569d-4088-89e2-236614eeece5shareToken=c11dfffbaa859f81982969fcc2cee1b0"
[X Link](https://x.com/S_OhEigeartaigh/status/1959600072082989217) 2025-08-24T12:53Z [----] followers, [----] engagements
"Somehow managed to miss this memo to the President from the Special Competitive Studies Project (the Schmidt-funded offshoot of the NSCAI) on racing to AGI. Man it's bad. Massively overstates China's AGI national positioning and capacity in order to call for Manhattan project & 'an AGI attack framework'. Will have to write about this at some point. https://www.scsp.ai/wp-content/uploads/2025/01/AGI-Memo.pdf https://www.scsp.ai/wp-content/uploads/2025/01/AGI-Memo.pdf"
[X Link](https://x.com/S_OhEigeartaigh/status/1959730572688773352) 2025-08-24T21:32Z [----] followers, [----] engagements
"Comes to you courtesy of a Sunday afternoon dive into 'what happened after Schmidt & the NSCAI basically kicked off the US-China AI race' It turns out: packed Biden's OSTP with former NSCAIers https://www.politico.com/news/2022/03/28/google-billionaire-joe-biden-science-office-00020712 https://www.politico.com/news/2022/03/28/google-billionaire-joe-biden-science-office-00020712"
[X Link](https://x.com/S_OhEigeartaigh/status/1959731148101144892) 2025-08-24T21:34Z [----] followers, [----] engagements
"and set up the Special Competitive Studies Project run by the former ED of the NSCAI which now lobbies for deregulation of AI - and getting to AGI first before China of course https://files.nitrd.gov/90-fr-9088/SCSP-AI-RFI-2025.pdf https://files.nitrd.gov/90-fr-9088/SCSP-AI-RFI-2025.pdf"
[X Link](https://x.com/S_OhEigeartaigh/status/1959731568651743260) 2025-08-24T21:36Z [----] followers, [---] engagements
"The American policy information environment is just so hilariously tragically f***ed. Whether it washes out with sane policy will depend on whether the "we want to sell chips forget AGI" regulatory capture messaging manages to dilute the "China's racing to AGI so we should too" regulatory capture messaging to create something confused but basically sane in the middle"
[X Link](https://x.com/S_OhEigeartaigh/status/1959732744855966043) 2025-08-24T21:41Z [----] followers, [---] engagements
"Looks to me like critical mass of advisers around current USG notably sceptical on AGI in Trump's presidency and this is reflected in a policy shift from control+containment ('AGI race' strategy) to diffusion+export ('normal technology' race). Those who saw AGI as imminent - Musk Dean Ball - now out. Those who think it's a distraction - Sacks Krishnan - still in. 1/4 @gmiller @DavidSacks Ill write about this separately but think this notion of imminent AGI has been a distraction and harmful and now effectively proven wrong. @gmiller @DavidSacks Ill write about this separately but think this"
[X Link](https://x.com/anyuser/status/1961412811667435549) 2025-08-29T12:57Z [----] followers, 25.1K engagements
"@primalpoly I don't think Sacks really buys into the AGI thing and the rest follows from that"
[X Link](https://x.com/S_OhEigeartaigh/status/1961447842670653462) 2025-08-29T15:16Z [----] followers, [---] engagements
"@rajiinio @ShakeelHashim My understanding is that Dario was recruited from Google in [----] and would certainly have been influential (think he was something like chief scientist) with Daniela joining in [----]. They weren't in the founding team (which isn't to say they didn't matter)"
[X Link](https://x.com/S_OhEigeartaigh/status/1961486677425209451) 2025-08-29T17:50Z [----] followers, [---] engagements
"To your original point I do think EA needs to be held to an extent accountable for contributions to the 'AGI company proliferation/race to AGI' flywheel they contributed to even if inadvertently. Sadly it might end up being EA's biggest over all contribution to existential risk"
[X Link](https://x.com/S_OhEigeartaigh/status/1961487159602401604) 2025-08-29T17:52Z [----] followers, [--] engagements
"The ten biggest companies in the world by market cap: The hardware players: 1) Nvidia 9) TSMC semiconductors (both in supply chain that produces high end chips). 8) Broadcom provides custom components for tech companies' AI workloads plus datacentre infrastructure. 1/4"
[X Link](https://x.com/S_OhEigeartaigh/status/1961832320534503538) 2025-08-30T16:44Z [----] followers, [---] engagements
"The digital giants: 2) Microsoft 3) Apple 4) Alphabet 5) Amazon 6) Meta all have in-house AI teams; Microsoft and Amazon also have partnerships w OpenAI and Anthropic which rely on their datacentre capacity. 10) Tesla's CEO describes it as 'basically an AI company' 2/4"
[X Link](https://x.com/S_OhEigeartaigh/status/1961832322250035301) 2025-08-30T16:44Z [----] followers, [---] engagements
"7) Saudi Aramco is Saudi Arabia's national oil company; Saudi Arabia was one of the countries the USA inked deals with this summer that centrally included plans for AI infrastructure buildout. Low-cost and abundant energy from oil/gas makes the Middle East attractive for hosting compute.3/4"
[X Link](https://x.com/S_OhEigeartaigh/status/1961832323990569419) 2025-08-30T16:44Z [----] followers, [---] engagements
"My position that China's not racing to AGI in any meaningful sense continues to hold up well I think. https://papers.ssrn.com/sol3/papers.cfmabstract_id=5278644 Heres a great overview and analysis (and English translation) of Chinas new AI Plus plan. Its not about achieving some abstract idea of AGI but about practically infusing AI into every sector of the economy to drive productivity and growth. https://t.co/GBfSXDj7Li https://papers.ssrn.com/sol3/papers.cfmabstract_id=5278644 Heres a great overview and analysis (and English translation) of Chinas new AI Plus plan. Its not about achieving"
[X Link](https://x.com/S_OhEigeartaigh/status/1962170499011612821) 2025-08-31T15:07Z [----] followers, [----] engagements
"@AkashWasil More scepticism that AGI is imminent; more scepticism amongst a lot of senior people who hold sway that the present LLM/scaling/RL etc paradigm will get us there. Kristy Loke and I have a piece coming out in Wire China next week that will discuss a bit"
[X Link](https://x.com/S_OhEigeartaigh/status/1962188832213696522) 2025-08-31T16:20Z [----] followers, [---] engagements
"I keep noting I'm sceptical about the timelines Jack describes in terms of my mainline. But if it's being said by someone as broadly knowledgeable as Jack plus with his inside-frontier-industry it should be taken seriously as a possibility. I don't think they're saying this to sell us hype. I think it's possible (a) they have more information than me (b) they're in an echo chamber (c) are looking at the same information and making a different prediction on something quite uncertain. But I think their desire to warn us about what's coming is sincere and well-intentioned. And if they're right"
[X Link](https://x.com/S_OhEigeartaigh/status/1962444568776876112) 2025-09-01T09:17Z [----] followers, [----] engagements
"They seem to have a hard time believing that we are an organization of people who just like actually care about this. (Nathan Calvin Encode who do good work) Inside OpenAI there's a growing paranoia that some of its loudest critics are being funded by Elon Musk and other billionaire competitors. Now they are going after these nonprofit groups but their evidence of a vast conspiracy is often extremely thin https://t.co/ajXm6iNzJr Inside OpenAI there's a growing paranoia that some of its loudest critics are being funded by Elon Musk and other billionaire competitors. Now they are going after"
[X Link](https://x.com/S_OhEigeartaigh/status/1962962523600072819) 2025-09-02T19:35Z [----] followers, [----] engagements
"This article just keeps on giving "Lehane who joined the company last year is perhaps best known for coining the term vast right-wing conspiracy to dismiss the allegations against Bill Clinton during the Monica Lewinsky scandal a line that seems to have seeped into Leading the Futures messaging too. In a statement to the Journal representatives from the PAC decried a vast force out there thats looking to slow down AI deployment prevent the American worker from benefiting from the U.S. leading in global innovation and job creation and erect a patchwork of regulation.""
[X Link](https://x.com/S_OhEigeartaigh/status/1962964248004342264) 2025-09-02T19:42Z [----] followers, [---] engagements
"Dear editor I regret my article will be late due to a vast force out there thats looking to slow down AI deployment prevent the American worker from benefiting from the U.S. leading in global innovation and job creation and erect a patchwork of regulation. Also I forgot"
[X Link](https://x.com/S_OhEigeartaigh/status/1962971793016119518) 2025-09-02T20:12Z [----] followers, [---] engagements
"Friends does anyone know who funds Alliance For the Future It's not listed on their website. Thanks https://www.affuture.org/about/ https://www.affuture.org/about/"
[X Link](https://x.com/S_OhEigeartaigh/status/1963238427278606780) 2025-09-03T13:51Z [----] followers, [---] engagements
"Interesting question. Counterfactual world. DeepMind doesn't get formed. (A company explicitly talking about AGI at a time when talking about AGI made you subject to ridicule). AGI remains ridiculous for longer. Does OpenAI get formed Probably not Musk pushed for it in response to OpenAI. The East doesn't get its 'sputnik moment' when Alphago beat Lee Sedol. Transformers probably still happen within Google. Do you have a company prepared to take the crazy high-stakes bet on scaling LLMs that OpenAI took (shutting down their other programs) My guess is probably not for at least several years."
[X Link](https://x.com/S_OhEigeartaigh/status/1963919887312846940) 2025-09-05T10:59Z [----] followers, [---] engagements
"Recommend OpenAI subpoena these AGs to find out if they're really Elon Musk and three ferrets in a very large trenchcoat. Scorching new letter from CA and DE AGs to OpenAI who each have the power to block the company's restructuring to loosen nonprofit controls. They are NOT happy about the recent teen suicide and murder-suicide that followed prolonged and concerning interactions with ChatGPT. https://t.co/52vfx8Dfth Scorching new letter from CA and DE AGs to OpenAI who each have the power to block the company's restructuring to loosen nonprofit controls. They are NOT happy about the recent"
[X Link](https://x.com/anyuser/status/1964048253864784013) 2025-09-05T19:29Z [----] followers, [----] engagements
"Has been relevant on AI & compute for at least last year. And in my view has walked a pretty good line (in the UK political context) between pushing adoption of AI and advocating for safety role of standards and AISI. e.g. In a world of fierce AI competition the UK can lead on standards using sovereign capacity to drive international co-ordination. The AI Safety Institute has already made strides in international safety dialogues. It is time to empower that institution further to keep its lead." https://www.politicshome.com/thehouse/article/britain-seize-opportunity-deliver-ai-boom"
[X Link](https://x.com/S_OhEigeartaigh/status/1964742929538580881) 2025-09-07T17:29Z [----] followers, [--] engagements
"@davidmanheim And just when OpenAI had convinced me we were all secretly Elon Musk"
[X Link](https://x.com/S_OhEigeartaigh/status/1965449341440852031) 2025-09-09T16:16Z [----] followers, [--] engagements
"I am delighted that Ireland will no longer have a Presidential candidate with a track record of rape and sexual and physical assault against women. Still sickened that the U.S. White House would choose to champion this man. Irish mixed martial arts star Conor McGregor has announced that he is withdrawing his candidacy for the country's presidential election but added that his "commitment to Ireland does not end here". ➡ https://t.co/R9vGYQwQX2 https://t.co/pAOfvVEVmU Irish mixed martial arts star Conor McGregor has announced that he is withdrawing his candidacy for the country's presidential"
[X Link](https://x.com/S_OhEigeartaigh/status/1967582433416815056) 2025-09-15T13:33Z [----] followers, [----] engagements
"2025 tech headlines be like:"
[X Link](https://x.com/anyuser/status/1969075274433089871) 2025-09-19T16:25Z [----] followers, 64K engagements
"@So8res Yes. A small number of people with crazily high risk tolerance and a vision of the Good that doesn't align with most people's are making decisions that they believe will set the direction of the future for everyone. Weird flex for anyone who believes in liberal democracy"
[X Link](https://x.com/S_OhEigeartaigh/status/1969435525409833163) 2025-09-20T16:16Z [----] followers, [---] engagements
"We finally got around to watching the Quiet Earth last weekend. Eerie and brilliant with one of the best scores I've heard in film. Up there in the pantheon of existential risk movies would highly recommend"
[X Link](https://x.com/S_OhEigeartaigh/status/1970569292568821794) 2025-09-23T19:21Z [----] followers, [---] engagements
"This seems very good. Good going President Trump looking forward to seeing more of this and action to back it. BREAKING: President Trump says Ukraine is in a position to fight and win all of Ukraine back in its original form. https://t.co/8T3rjh3pbr BREAKING: President Trump says Ukraine is in a position to fight and win all of Ukraine back in its original form. https://t.co/8T3rjh3pbr"
[X Link](https://x.com/S_OhEigeartaigh/status/1970571109440618896) 2025-09-23T19:28Z [----] followers, [---] engagements
"Helberg: adviser to palantir CEO Alex Karp. Wants an AI Manhattan Project (Karp thinks Xrisk is not an unjustified concern but less important than beating china https://www.nytimes.com/2023/07/25/opinion/karp-palantir-artificial-intelligence.html https://www.reuters.com/technology/artificial-intelligence/us-government-commission-pushes-manhattan-project-style-ai-initiative-2024-11-19/ https://www.nytimes.com/2023/07/25/opinion/karp-palantir-artificial-intelligence.html"
[X Link](https://x.com/S_OhEigeartaigh/status/1971131232441495593) 2025-09-25T08:34Z [----] followers, [--] engagements
"Kratsios: former Managing Director of Scale AI. Scale AI opposes international treaties and rules that would impact US developers and endorses light local governance and exporting the US stack https://scale.com/blog/us-ai-action-plan-scale-respons https://scale.com/blog/us-ai-action-plan-scale-respons"
[X Link](https://x.com/S_OhEigeartaigh/status/1971131234450543038) 2025-09-25T08:34Z [----] followers, [---] engagements
"Krishnan former general partner at a16z. a16z oppose safety talk basically all binding regulation on developers. https://a16z.com/a16zs-recommendations-for-the-national-ai-action-plan/ https://pmarca.substack.com/p/why-ai-will-save-the-world https://a16z.com/a16zs-recommendations-for-the-national-ai-action-plan/ https://pmarca.substack.com/p/why-ai-will-save-the-world"
[X Link](https://x.com/S_OhEigeartaigh/status/1971131236203757707) 2025-09-25T08:34Z [----] followers, [--] engagements
"Director of the OSTP and the E both dismissing AI xrisk today. Unsurprising: what's happening in USG around AI right now looks to me like one of the grandest examples of regulatory capture I've ever seen. Taking xrisk seriously was always going to demand regulation and international cooperation so it was always going to be in the firing line. 1/5 https://x.com/jacobhelberg/status/1971038642341564863 https://x.com/jacobhelberg/status/1971038642341564863"
[X Link](https://x.com/S_OhEigeartaigh/status/1971132368967631119) 2025-09-25T08:39Z [----] followers, [----] engagements
"Helberg: adviser to palantir CEO Alex Karp. Wants an AI Manhattan Project (Karp thinks Xrisk is not an unjustified concern but less important than beating china 2/5 https://nytimes.com/2023/07/25/opinion/karp-palantir-artificial-intelligence.html https://reuters.com/technology/artificial-intelligence/us-government-commission-pushes-manhattan-project-style-ai-initiative-2024-11-19/ https://nytimes.com/2023/07/25/opinion/karp-palantir-artificial-intelligence.html"
[X Link](https://x.com/S_OhEigeartaigh/status/1971132370553348581) 2025-09-25T08:39Z [----] followers, [---] engagements
"Kratsios: former Managing Director of Scale AI. Scale AI opposes international treaties and rules that would impact US developers and endorses light local governance and exporting the US stack 3/5 https://scale.com/blog/us-ai-action-plan-scale-respons https://scale.com/blog/us-ai-action-plan-scale-respons"
[X Link](https://x.com/S_OhEigeartaigh/status/1971132372558205431) 2025-09-25T08:39Z [----] followers, [---] engagements
Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing
@S_OhEigeartaigh Seán Ó hÉigeartaigh
Seán Ó hÉigeartaigh posts on X about ai, open ai, agi, china the most. They currently have [---------] followers and [---] posts still getting attention that total [---] engagements in the last [--] hours.
Engagements: [---] #
- [--] Week [------] -76%
- [--] Month [-------] +264%
- [--] Months [----------] +24%
- [--] Year [----------] +822%
Mentions: [--] #
- [--] Week [--] -45%
- [--] Month [--] +200%
- [--] Months [---] +138%
- [--] Year [---] +267%
Followers: [---------] #
- [--] Week [-----] +0.36%
- [--] Month [-----] +3.60%
- [--] Months [-----] +22%
- [--] Year [-----] +50%
CreatorRank: [---------] #
Social Influence
Social category influence technology brands countries celebrities travel destinations stocks finance vc firms events social networks us election
Social topic influence ai, open ai, agi, china, in the, for the, elon musk, anthropic, the world, $googl
Top accounts mentioned or mentioned by @shakeelhashim @tylermjohn @peterwildeford @davidmanheim @davidsacks @andrewcritchphd @jackstilgoe @grok @nathancalvin @sriramk @martinshkreli @garymarcus @milesbrundage @elonmusk @daniel_271828 @jamiesusskind @nathanpmyoung @humanharlan @gmiller @soheigeartaigh
Top assets mentioned Alphabet Inc Class A (GOOGL) Microsoft Corp. (MSFT) Frontier (FRONT)
Top Social Posts
Top posts by engagements in the last [--] hours
"It's a privilege to have been invited to speak at the Dublin Tech Summit today Always lovely to have an excuse to come home to an uncharacteristically sunny Dublin (And I should probably thank the CAIS folk for creating some handy news for us to discuss) https://dublintechsummit.tech/ https://dublintechsummit.tech/"
X Link 2023-05-31T08:02Z [----] followers, [---] engagements
"@LeverhulmeCFI The other important clarification is that while this article presents the Cambridge Institute as being v much centred on AI its remit is deliberately much broader. Right now it includes & will expand on work at CSER on climate biorisk nuclear weapons geoengineering etc 5/n"
X Link 2023-07-10T08:27Z [----] followers, [---] engagements
"It was a pleasure to speak at the Dublin Tech Summit this year. Just came across this nice writeup. https://businessandfinance.com/news/highlights-of-day-1-on-the-evolution-stage-dublin-tech-summit/ https://businessandfinance.com/news/highlights-of-day-1-on-the-evolution-stage-dublin-tech-summit/"
X Link 2023-08-02T11:33Z [----] followers, [---] engagements
"Disappointed to see the World Health Organisation promoting a pseudoscience like homeopathy (not to mention mislabelling it a 'traditional medicine')"
X Link 2023-08-16T17:58Z [----] followers, [----] engagements
"Useful thread from Robert Trager one of the lead authors of the report"
X Link 2023-08-31T20:15Z [----] followers, [---] engagements
"Writing reference letters for North American roles: is it on me to amp up the hyperbole or on them to correct for my European curmudgeonship Do you not know that "Ah he's grand like" is the Highest Praise that my People Offer"
X Link 2023-09-11T18:31Z [----] followers, [----] engagements
"Sign me up for more "Tegmark unplugged""
X Link 2023-09-11T18:45Z [----] followers, [---] engagements
"Delighted to see my colleague and coauthor Yi Zeng on the UN Advisory Body. He brings a remarkable wealth of knowledge and wisdom and the UN (and world) will benefit tremendously from it"
X Link 2023-10-26T18:14Z [----] followers, [---] engagements
"Impressive and comprehensive review of AI alignment from scholars at Peking University Cambridge CMU HKUST and USC"
X Link 2023-11-01T20:56Z [----] followers, [---] engagements
"@mealreplacer Don't be such an igg"
X Link 2023-11-17T03:12Z [----] followers, [--] engagements
"@halhod @jamiesusskind Also n=1. Did anyone expect Google/DM to go from 'significantly behind' OpenAI for this kind of model to "blowing OpenAI out of the water" They've caught up and improved sightly which is impressive in of itself. Now let's see what the leaders do"
X Link 2023-12-06T16:18Z [----] followers, [--] engagements
""To account for the wide range of tasks AI systems can accomplish and the quick expansion of its capabilities it was agreed that general-purpose AI (GPAI) systems and the GPAI models they are based on will have to adhere to transparency requirements""
X Link 2023-12-09T01:51Z [----] followers, [--] engagements
"This is really remarkable: Half-million-year-old interlocking wooden logs unearthed in Zambia eariler this year"
X Link 2023-12-09T04:04Z [----] followers, [---] engagements
"First time to Hong Kong and I had an inspiring time at HKUST's International AI Cooperation and Governance forum. I had the opportunity to speak on considerations around open-sourcing powerful foundation models and to participate in an excellent panel on safety & governance"
X Link 2023-12-10T00:52Z [----] followers, [---] engagements
"As well as Hong Kong I hugely enjoyed a visit to Beijing last month. Re-energised collaborations & friendships & connected to some very exciting researchers & think tanks working on AI alignment global ethics AI risk & global AI governance. Here w/ two of @yi_zeng 's PhDs"
X Link 2023-12-12T10:23Z [----] followers, [---] engagements
"New essential reading blog just dropped. I started a blog about scientific research and science funding. You can subscribe on the site to receive future posts as emails. Heres the first one. Society is not focused on some of the most important technical problems: https://t.co/89h1sZ0rGe I started a blog about scientific research and science funding. You can subscribe on the site to receive future posts as emails. Heres the first one. Society is not focused on some of the most important technical problems: https://t.co/89h1sZ0rGe"
X Link 2024-02-05T02:59Z [----] followers, [---] engagements
"@praeterpropter Maybe they can all come over and fix another moribund economy dragged down by political decisions"
X Link 2024-02-14T20:50Z [----] followers, [--] engagements
"tl;dr Elon Musk is suing OpenAI; claiming GPT-4 & rumoured 'Q' are AGI & should be released to public; that Altman and Microsoft have replaced OpenAI's board with Microsoft-friendly folk who know nothing about A(G)I 1/2"
X Link 2024-03-01T17:11Z [----] followers, [---] engagements
"so Microsoft can continue controlling AGI and make a fortune selling it to the public (if it is AGI it is outside OpenAI's agreement with Microsoft). .how's your Friday going"
X Link 2024-03-01T17:12Z [----] followers, [---] engagements
"My view: GPT-4 is not AGI. I also do not believe AGI 'has been achieved internally' or is likely to be achieved in the coming year. I well believe Microsoft might want a board that would keep OpenAI stable and relationship with Microsoft strong but they're not tying to hide AGI"
X Link 2024-03-01T17:19Z [----] followers, [---] engagements
"Nvidia now 3rd largest company in the world by market cap passing out Saudi Aramco. [--] out of the [--] largest in world (Microsoft Nvidia Amazon Alphabet Meta) very much in AI chain (6 of [--] if you include Apple). This is noteworthy"
X Link 2024-03-05T10:33Z [----] followers, [---] engagements
"@Jackstilgoe Boston dynamics have a lot to answer for here"
X Link 2024-03-06T08:38Z [----] followers, [--] engagements
"Great to see this fascinating paper out led by CSER affiliate Tom Cernev and informed by an expert elicitation and workshop process held in Cambridge. After the Starship launch last week it feels like Space is back baby"
X Link 2024-03-18T11:19Z [----] followers, [---] engagements
"@anabrandusescu This guy's waiting for your phonecall https://www.bloomberg.com/news/articles/2023-06-01/palantir-ceo-touts-power-of-ai-at-customer-conference https://www.bloomberg.com/news/articles/2023-06-01/palantir-ceo-touts-power-of-ai-at-customer-conference"
X Link 2024-04-03T12:56Z [----] followers, [---] engagements
"@briandavidearp @KellerScholl Much more trivially what an odd metaphor to choose. "Don't worry about the lion it's your own worry that is the risk" is. an interesting framing for the AI risk issue. How I stopped worrying and learned to love the Palantir I guess"
X Link 2024-04-05T22:48Z [----] followers, [--] engagements
"This overview of the EU AI Act's risk classification framework by CFI's Yulu Pi is very helpful and highly recommended. http://lcfi.ac.uk/news-and-events/news/2024/apr/2/eu-ai-act-high-risk-ai-nutshell/ http://lcfi.ac.uk/news-and-events/news/2024/apr/2/eu-ai-act-high-risk-ai-nutshell/"
X Link 2024-04-14T19:11Z [----] followers, [----] engagements
"Excited to welcome @MichelleMalonza to Cambridge I'm thrilled and excited at the incredible calibre of scholars we get to engage with through the @ILINAProgram . Welcome to CSER visitor @MichelleMalonza Michelle Malonza is a Research Associate at the @ILINAProgram. At CSER she will be working on an OpenAI funded project on how democratic societies can responsibly integrate AI into their critical infrastructure. https://t.co/p4hr36feVP https://t.co/JcrT9qB1Sw Welcome to CSER visitor @MichelleMalonza Michelle Malonza is a Research Associate at the @ILINAProgram. At CSER she will be working on"
X Link 2024-05-01T10:22Z [----] followers, [----] engagements
"@MichelleMalonza @ILINAProgram Michelle will be working on governance of AI agents and their interaction with critical infrastructure but I also enjoyed this paper on the role of Global South countries in frontier AI governance - a very important topic. https://arxiv.org/abs/2312.04616 https://arxiv.org/abs/2312.04616"
X Link 2024-05-01T10:24Z [----] followers, [---] engagements
"@NathanpmYoung I think part of the issue here is that EA is sort-of an integrated thing (use reason & evidence to do most good) and sort-of [--] cause area foxes in a trenchcoat all with quite different approaches. So while there might have been a bit of bait-and-switch by some pplfeels like1/2"
X Link 2024-05-01T10:28Z [----] followers, [---] engagements
"Apologies to everyone I owe messages/emails to. Got back last week from an incredible trip to Beijing. Still catching up. Will say more soon - despite everything it feels like there's a real groundswell of interest in AI safety and cooperation happening there"
X Link 2024-05-10T16:22Z [----] followers, [---] engagements
"Useful thread from the ever-excellent Ed Conway on tariffs announced yesterday by Biden. Key question (in terms of electric vehicle uptake/cost which is important for net zero): how does this affect components 🧵 Joe Biden has just confirmed he's going to raise the special tariff on electric vehicles coming from China to 100%. Also new tariffs on batteries and solar panels. What's going on here Here's a quick primer with some charts 👇 🧵 Joe Biden has just confirmed he's going to raise the special tariff on electric vehicles coming from China to 100%. Also new tariffs on batteries and solar"
X Link 2024-05-15T11:02Z [----] followers, [---] engagements
"Forget throwback Thursday; it's Frontier (Safety) Friday: https://deepmind.google/discover/blog/introducing-the-frontier-safety-framework/ https://deepmind.google/discover/blog/introducing-the-frontier-safety-framework/"
X Link 2024-05-17T14:35Z [----] followers, [---] engagements
"@albrgr Not sure he was on their safety team but certainly safety-concerned: "Quit OpenAI due to losing confidence that it would behave responsibly around the time of AGI." https://www.lesswrong.com/users/daniel-kokotajlo https://www.lesswrong.com/users/daniel-kokotajlo"
X Link 2024-05-17T16:21Z [----] followers, [---] engagements
"@eshear Have you considered being CEO of an AGI company"
X Link 2024-05-18T15:16Z [----] followers, [--] engagements
"@ShakeelHashim That's what they get for not checking whether any openAI drama was going to drop that day. Very short-sighted"
X Link 2024-05-21T11:12Z [----] followers, [---] engagements
"@HaydnBelfield Want to swap Damn cat eats every plant we bring into the house. I thought it was supposed to be a carnivore"
X Link 2024-06-11T19:27Z [----] followers, [---] engagements
"@Zach49639228 @sleepinyourhat Huh yeah that's what I must be remembering. Can anyone help clear this up"
X Link 2024-06-20T21:27Z [----] followers, [---] engagements
"Impressive from the White House: https://www.whitehouse.gov/briefing-room/statements-releases/2024/07/26/fact-sheet-biden-harris-administration-announces-new-ai-actions-and-receives-additional-major-voluntary-commitment-on-ai/ https://www.whitehouse.gov/briefing-room/statements-releases/2024/07/26/fact-sheet-biden-harris-administration-announces-new-ai-actions-and-receives-additional-major-voluntary-commitment-on-ai/"
X Link 2024-07-26T20:25Z [----] followers, [---] engagements
"A lot of interesting work going on within the google deepmind AGI safety&alignment team - a helpful overview. So freaking proud of the AGI safety&alignment team -- read here a retrospective of the work over the past [---] years across frontier safety oversight interpretability and more. Onwards https://t.co/p7N0KM1LJ8 So freaking proud of the AGI safety&alignment team -- read here a retrospective of the work over the past [---] years across frontier safety oversight interpretability and more. Onwards https://t.co/p7N0KM1LJ8"
X Link 2024-08-20T18:20Z [----] followers, [---] engagements
"Pleased to have contributed to this paper led by the excellent Hadrian Pouget and Claire Dennis on the future of AI scientific risk assessment. 1/3 How can nations achieve a shared scientifically rigorous understanding of AI risks Check out our new report that examines the challenges and opportunities https://t.co/Y50HJs2TTv https://t.co/qGZCQflxnu How can nations achieve a shared scientifically rigorous understanding of AI risks Check out our new report that examines the challenges and opportunities https://t.co/Y50HJs2TTv https://t.co/qGZCQflxnu"
X Link 2024-08-30T12:16Z [----] followers, [----] engagements
"The core message of the paperthat no single institution can accomplish everything needed from a global scientific AI risk assessment effortis esp important in light of Summit of the Futurewhere UN will establish foundation for its contribution to global risk assessments.2/3"
X Link 2024-08-30T12:16Z [----] followers, [---] engagements
"Useful article on the change in attention to and views on frontier AI safety in China by Matt Sheehan - broadly matches my experience: https://carnegieendowment.org/research/2024/08/china-artificial-intelligence-ai-safety-regulationlang=en https://carnegieendowment.org/research/2024/08/china-artificial-intelligence-ai-safety-regulationlang=en"
X Link 2024-09-04T11:52Z [----] followers, [---] engagements
"@davidad Or that Hiroaki Kitano CTO of Sony is the same Hiroaki Kitano whose systems biology papers I used to read"
X Link 2024-09-12T10:48Z [----] followers, [---] engagements
"@GaryMarcus I get that you've been interacting with unrealistic fanboys but this feels like it's setting the bar incredibly high. Steady incremental progress stacks up and over the course of years it's been very impressive"
X Link 2024-09-12T18:40Z [----] followers, [----] engagements
"@Manderljung It worries me how front of stage they've become and how quickly particularly in the policy realm. My intuition is to be more worried abt (a) general capability trends (b) autonomy (c) deception (d) applicability to AI research & I worry this might point us in wrong direction"
X Link 2024-09-13T10:31Z [----] followers, [--] engagements
"Insightful blog post by Lennart Heim exploring the compute governance implications of the OpenAI o1 models https://blog.heim.xyz/inference-compute/ https://blog.heim.xyz/inference-compute/"
X Link 2024-09-23T18:17Z [----] followers, [---] engagements
"@jamiesusskind It's almost as if being in power wasn't much fun and they never want to win an election again"
X Link 2024-09-29T11:39Z [----] followers, [--] engagements
"@jamiesusskind "Come up with a less vote-winning opinion than Kemi Badenoch" challenge. And they've already played the "bring back national service " card so you can't use that"
X Link 2024-09-29T11:45Z [----] followers, [--] engagements
"@anderssandberg What about stopping A(G)I development specifically though It's only a small part of the overall space if a disproportionately important and risky one. Could still get lots of benefits from the rest"
X Link 2024-10-03T22:45Z [----] followers, [---] engagements
"Great to see Geoffrey Hinton with the Nobel Prize and grateful to Cambridge for featuring his [----] lecture for us on AI and existential risk in their excellent news story https://www.cam.ac.uk/research/news/university-of-cambridge-alumnus-awarded-2024-nobel-prize-in-physics https://www.cam.ac.uk/research/news/university-of-cambridge-alumnus-awarded-2024-nobel-prize-in-physics"
X Link 2024-10-08T11:53Z [----] followers, [----] engagements
"It's not done yet. Hearing reports that the Nobel prize for literature will be going to the authors of "OpenAI's nonprofit governance structure" for outstanding contributions to creative fiction"
X Link 2024-10-09T11:37Z [----] followers, 266.2K engagements
"How can you tell when you're being RLHF'd into becoming a niche corporate AI governance comedy account Asking for a friend. Also did you hear the one about the antitrust probe into microsoft's acqui-merger of Mistral"
X Link 2024-10-09T15:46Z [----] followers, [----] engagements
"@LinchZhang Get Jurgen Schmidthuber in there and I'll be impressed with your diplomatic nous https://x.com/SchmidhuberAI/status/1844022724328394780 The #NobelPrizeinPhysics2024 for Hopfield & Hinton rewards plagiarism and incorrect attribution in computer science. It's mostly about Amari's "Hopfield network" and the "Boltzmann Machine." [--]. The Lenz-Ising recurrent architecture with neuron-like elements was published in https://x.com/SchmidhuberAI/status/1844022724328394780 The #NobelPrizeinPhysics2024 for Hopfield & Hinton rewards plagiarism and incorrect attribution in computer science. It's"
X Link 2024-10-09T17:07Z [----] followers, [---] engagements
"@LinchZhang In case you didn't know he invented the idea of "the AI doesn't care about you" https://www.theguardian.com/technology/2017/apr/18/robot-man-artificial-intelligence-computer-milky-way https://www.theguardian.com/technology/2017/apr/18/robot-man-artificial-intelligence-computer-milky-way"
X Link 2024-10-09T17:10Z [----] followers, [---] engagements
"@diviacaroline yes sorry sloppy wording. briefly looking up: legalised in many states before [----] made legal across the us after supreme court decision in 1967"
X Link 2024-10-11T15:35Z [----] followers, [---] engagements
"Fascinating report by Julia Willemyns Haydn Belfield Tom Milton on what the UK needs to do to be positioned for the oncoming AI revolution - worth a read. What if we took AI companies plans investments & ambitions seriously What should the UK Government do to prepare for "intelligence too cheap to meter" THREAD🧵 https://t.co/9gYmllVom9 What if we took AI companies plans investments & ambitions seriously What should the UK Government do to prepare for "intelligence too cheap to meter" THREAD🧵 https://t.co/9gYmllVom9"
X Link 2024-10-15T12:25Z [----] followers, [---] engagements
"Report is here. Worth noting: nowhere does the report reference AGI or driving towards AGI. I believe there are many low-hanging fruit in terms of application of present cutting-edge AI and near-term advances that aren't AGI. The stated goal of 2/n https://ukdayone.org/briefings/ai-industrial-strategy https://ukdayone.org/briefings/ai-industrial-strategy"
X Link 2024-10-15T12:28Z [----] followers, [--] engagements
"the frontier labs of developing AGI/'powerful AI' in my view carries significant risks and deserves special consideration (e.g. not progressing beyond certain point without being very certain key safety and governance challenges have been solved) BUT this is not incompatible3/4"
X Link 2024-10-15T12:29Z [----] followers, [---] engagements
"I enjoyed reading Amodei's "Machines of Loving Grace" an essay focused on his view of the radical potential positive upsides of AI. But I found aspects of it troubling. I wrote about these on the blog post below: 1/6 https://www.lcfi.ac.uk/news-events/blog/post/reflections-on-machines-of-loving-grace https://www.lcfi.ac.uk/news-events/blog/post/reflections-on-machines-of-loving-grace"
X Link 2024-10-23T10:29Z [----] followers, [----] engagements
"Excellent to see this important paper out from the Covid Lessons team at CSER. Assesses UK's pre-covid pandemic preparedness and early response. Preparedness includes pandemic strategy exercises stockpiles. 1/n https://www.tandfonline.com/doi/full/10.1080/17441692.2024.2415499#abstract https://www.tandfonline.com/doi/full/10.1080/17441692.2024.2415499#abstract"
X Link 2024-10-25T09:13Z [----] followers, [---] engagements
"Response includes quarantining test&trace mobility restrictions. Paper finds important deficiencies:includes centrality of 'inevitability of spread' assumptioninsufficient implementation of pandemic exercise recommendations lack of live learning from other countries &more2/n"
X Link 2024-10-25T09:13Z [----] followers, [---] engagements
"@bytebittech @AndrewCritchPhD @DarioAmodei I read the above as (a) pressing for fast progress ("scaling quickly") at least in part for international advantage&(b) effectively bringing china to the table over the barrel of a gun. I think that's where our visions diverge. Others have also id'd this as a 'growing faultline'"
X Link 2024-10-25T11:31Z [----] followers, [--] engagements
"@bytebittech @AndrewCritchPhD @DarioAmodei e.g. see "I think the next major fault line in the AI debates will be posture toward China. Altman Anthropic CEO Dario Amodei and former OpenAI researcher Leopold Aschenbrenner have prominently advanced varieties of the idea that democracies must. https://garrisonlovely.substack.com/p/end-of-an-era-openais-agi-readiness https://garrisonlovely.substack.com/p/end-of-an-era-openais-agi-readiness"
X Link 2024-10-25T11:32Z [----] followers, [--] engagements
"@AndrewCritchPhD (2) the no. of previous trump admin people who won't support him now seems notable and speaks to how well an admin might run/be listened to on key issues next time round when important delicate things on AI might be happening"
X Link 2024-11-04T10:04Z [----] followers, [--] engagements
"@AndrewCritchPhD (3) seeing folks like RFK be lined up for health/vaccines doesn't make me confident that science issues will be handled especially well"
X Link 2024-11-04T10:05Z [----] followers, [--] engagements
"Concordia AI are doing great work on international AI governance & AI Safety in China and have positions open - please consider sharing/applying: "Concordia AI is hiring for roles across our Beijing and Singapore offices See details here https://aisafetychina.substack.com/p/concordia-ai-fall-2024-hiring-announcement https://aisafetychina.substack.com/p/concordia-ai-fall-2024-hiring-announcement"
X Link 2024-11-06T11:40Z [----] followers, [---] engagements
"The French guy on polymarket is in for a hell of a payday"
X Link 2024-11-06T11:49Z [----] followers, [---] engagements
"An interesting blog from Miles. I agree with central premise. My hope is that a robust 3rd party testing/auditing/evaluation ecosystem perhaps working in concert with the AISIs can ease some of this in an efficient way; but that won't cover a big chunk of safety development 1/2 New blog post: "The bread and butter problem in AI policy." I explain a metaphor I've been using in conversation a lot lately and what conclusions it might (or might not) imply. https://t.co/EyQyrSDabz New blog post: "The bread and butter problem in AI policy." I explain a metaphor I've been using in conversation a lot"
X Link 2024-11-06T13:59Z [----] followers, [---] engagements
"and also requires this ecosystem being well-funded. Possibly a role for government to fund. Perhaps company contributions can be required. Perhaps an expanded frontier model forum can fund. 2/2"
X Link 2024-11-06T14:00Z [----] followers, [---] engagements
"@undisciplined42 Well the 'happiness'/satisfaction literature on Bhutan is extremely flawed i'm afraid"
X Link 2024-11-07T15:24Z [----] followers, [--] engagements
"Exciting grants programme opened by the Foresight institute - automating research & forecasting; neurotech & AGI; security tech for AI; safe multipolar human-AI scenarios. Consider applying Deadline December [--]. https://foresight.org/ai-safety/ https://foresight.org/ai-safety/"
X Link 2024-11-08T11:20Z [----] followers, [---] engagements
"Great blog on a really exciting project by @LeverhulmeCFI 's Giulio Corsi & Bill Marino and CSER alum Alex Marcoci on using foundation models to detect synthetic media. Very cool. https://www.lcfi.ac.uk/news-events/blog/post/how-foundation-models-could-transform-synthetic-media-detection https://www.lcfi.ac.uk/news-events/blog/post/how-foundation-models-could-transform-synthetic-media-detection"
X Link 2024-11-12T16:29Z [----] followers, [---] engagements
"@Miles_Brundage in which we discover that openai's PR team were protecting the world from Miles' sci-fi takes all these years. he's gone completely unhinged now what have we done"
X Link 2024-11-20T03:04Z [----] followers, [--] engagements
"Wuzhen turns out to be almost absurdly pretty. Great AI buzz going on against a fairytale-town background"
X Link 2024-11-20T03:15Z [----] followers, [---] engagements
"Let's address the elephant in the room. The actors presently racing towards AGI are all US-based companies. The government being lobbied most actively to race/support racing towards AGI is the USG. China is being used as a justification for rushing forward on the most 1/2 I also do not know of any evidence to support this claim and I spend quite a lot of time speaking to Chinese AI experts. I also do not know of any evidence to support this claim and I spend quite a lot of time speaking to Chinese AI experts"
X Link 2024-11-21T02:13Z [----] followers, [----] engagements
"This was fascinating work - worth a look. Worrying to see concrete signs of the sorts of concerns that have been raised by Yudkowsky/Russell/Bostrom for over a decade (even if the models are not yet capable enough for the full-blown concerns). We worked with OpenAI to test o1 for in-context scheming capabilities before deployment. When o1 is strongly nudged to pursue a goal (but not instructed to be deceptive) it shows a variety of scheming behaviors like subverting oversight and deceiving the user about its https://t.co/q2ZBG0xy72 We worked with OpenAI to test o1 for in-context scheming"
X Link 2024-12-06T10:58Z [----] followers, [---] engagements
"@RosieCampbell @Miles_Brundage You'll no doubt be sorely missed at OpenAI but great to have more really talented independent voices out there Very excited for what you do next"
X Link 2024-12-06T18:24Z [----] followers, [---] engagements
"@satisficer hey I mean if you're not you're missing a trick. This is the new world"
X Link 2024-12-13T20:07Z [----] followers, [--] engagements
"One of the weird things about the world today is that the idea of 'AGI' is now regularly being talked about in e.g. policy contexts. But seems v clear that most policymakers' notion of AGI & its implications is vastly underpowered compared to that of the ppl trying to build AGI"
X Link 2024-12-16T10:33Z [----] followers, [----] engagements
"The builders are saying quite clearly what they're aiming for. But the implications are too wild. & policymakers have heard too many wild tech implications over the years (from 'energy too cheap to meter' to the 'bio industrial revolution') so it just becomes weird stuff ppl say"
X Link 2024-12-16T10:37Z [----] followers, [---] engagements
"If the AGI builders are right then at some point that will shift and very quickly in a way that will be very destabilising"
X Link 2024-12-16T10:39Z [----] followers, [---] engagements
"Wondering about whether I should write something about this (thoughts welcome)"
X Link 2024-12-16T10:48Z [----] followers, [---] engagements
"Useful and well-articulated response to the Anthropic alignment faking paper/discussion. Quick reactions to some of the other reactions I've been seeing to Anthropic's recent paper on "Alignment faking in LLMs" 1/9 Quick reactions to some of the other reactions I've been seeing to Anthropic's recent paper on "Alignment faking in LLMs" 1/9"
X Link 2024-12-19T11:27Z [----] followers, [---] engagements
"Astounding. Would this win any 5k road race in Ireland outright Kenyans invented running. Beatrice Chebet just became the first woman in history to run 5km in under [--] minutes. She did 13:52 during the Cursa Dels Nassos in Barcelona 🙌🏾 https://t.co/pIREuOHKml Kenyans invented running. Beatrice Chebet just became the first woman in history to run 5km in under [--] minutes. She did 13:52 during the Cursa Dels Nassos in Barcelona 🙌🏾 https://t.co/pIREuOHKml"
X Link 2024-12-31T16:58Z [----] followers, [---] engagements
"We love our current products but we are here for the glorious future. With superintelligence we can do anything else. 3/4"
X Link 2025-01-06T10:28Z [----] followers, [----] engagements
"A very useful analysis by Edward Kemberry building on their dissertation thesis for our Mphil. (also another nice example of the very insightful and timely topics Mphils have been analysing) https://www.lcfi.ac.uk/news-events/blog/post/six-months-later-reviewing-the-proliferation-paradigm https://www.lcfi.ac.uk/news-events/blog/post/six-months-later-reviewing-the-proliferation-paradigm"
X Link 2025-01-17T17:09Z [----] followers, [---] engagements
"@tyler_m_john especially given that despite that policy EA/rats are simultaneously sort-of responsible for the existence of the companies building AGI"
X Link 2025-01-17T17:55Z [----] followers, [--] engagements
"@tyler_m_john cue sam altman crooning "and I owe it all to yuuu" riding his unicorn into the jaws of the waiting paperclip maximiser. https://x.com/sama/status/1621621724507938816 eliezer has IMO done more to accelerate AGI than anyone else. certainly he got many of us interested in AGI helped deepmind get funded at a time when AGI was extremely outside the overton window was critical in the decision to start openai etc. https://x.com/sama/status/1621621724507938816 eliezer has IMO done more to accelerate AGI than anyone else. certainly he got many of us interested in AGI helped deepmind get"
X Link 2025-01-17T18:09Z [----] followers, [--] engagements
"I think Market is wrong abt implications of DeepSeek so today I bought nvidia shares. I'll sell them within a year and donate profits to a nonprofit working on policy to avoid catastrophic impacts of large models. Or I will lose money and learn an Important Lesson in Hubris"
X Link 2025-01-27T16:55Z [----] followers, [----] engagements
"Great to see this important report out.Quick thoughts:1) This is a super hard thing to do.The field is moving v quickly. The report correctly notes an 'evidence dilemma' for policymakers - policymakers must make decisions where there isn't yet a large scientific evidence base.1/8 Today we are publishing the first-ever International AI Safety Report backed by [--] countries and the OECD UN and EU. It summarises the state of the science on AI capabilities and risks and how to mitigate those risks. 🧵 Link to full Report: https://t.co/k9ggxL7i66 1/16 https://t.co/68Gcm4iYH5 Today we are publishing"
X Link 2025-01-29T14:06Z [----] followers, [----] engagements
"2) The report writers face same challenge. They need to demonstrate depth & evidence or face criticism for lack of scientific rigour but there hasn't been time to build up a peer-reviewed literature base for the most fast-moving and important developments. It's clearly a hard2/8"
X Link 2025-01-29T14:06Z [----] followers, [--] engagements
"@ShakeelHashim I asked claude to help: "Breaking news from the Transformer newsletter: Our correspondent remains trapped in security missing crucial talks on 'How AI Systems Could Deceive and Overtake Humanity.' Sources confirm the delay is due to an automated security system that keeps"
X Link 2025-02-06T08:49Z [----] followers, [--] engagements
"@NathanpmYoung I agree re donation good but social media sentiment analysis is a thing and I think it actually is useful for European leaders to be getting that signal before costly decisions tomorrow. Starmer can act precisely because this is a wildly popular cause"
X Link 2025-03-01T14:54Z [----] followers, [--] engagements
"I was pleased to see this call from prominent voices in AI ethics/fairness challenging whether AGI should be a goal of the field. And 30% of AAAI researchers surveyed saying AGI-focused R&D shouldn't proceed until we know we can control AGI. https://arxiv.org/html/2502.03689v1 https://arxiv.org/html/2502.03689v1"
X Link 2025-03-07T13:29Z [----] followers, [---] engagements
"My brothers in Christ is your goal to build coding tools that are better and faster than humans Or is your goal to build the Machine God and usher in the irrelevance of humanity Because it would sure be helpful to know"
X Link 2025-03-07T17:03Z [----] followers, 12.5K engagements
"Next few years will be a fascinating case study in the degree to which copyright law & various other things that are really quite important in the normal functioning ofa capitalist liberal democracy hold up in the face of securitisation and a technology arms race narrative. 1/2"
X Link 2025-03-13T14:08Z [----] followers, [---] engagements
"One of the nice things about Ireland is for the most part people have a bit of cop on and have no time for this sort of fear-mongering bollocks. It's why Conor McGregor doesn't represent us and I expect never will. Wow Wow"
X Link 2025-03-19T15:39Z [----] followers, [---] engagements
"Very productive Friday afternoon with members of Japan's AI Safety Institute. Exciting to learn how they're thinking and to discuss policy and forecasting for AGI. Gosh I love a Very Good Day of Work"
X Link 2025-03-21T18:46Z [----] followers, [----] engagements
"@tyler_m_john Worth checking you're not getting your M. Mitchells mixed up as there's a Margaret and a Melanie. Melanie does rigorous AI+cog sci/eval type work. Margaret known for everything from system cards to 'don't build agi' and 'don't build fully autonomous agents'"
X Link 2025-03-22T12:48Z [----] followers, [--] engagements
"@justjoshinyou13 Google Deepmind announces new goal: "Sabotage Intelligence Use it to Sabotage Everything Else""
X Link 2025-04-02T17:13Z [----] followers, [--] engagements
"@justjoshinyou13 It's such a good thing I don't work at Google Deepmind. On April Fools I'd release a [---] page AGI safety pdf that just had the words "Butlerian Jihad" typed once in the dead centre of every page"
X Link 2025-04-02T17:14Z [----] followers, [--] engagements
"I'm seeing criticism of this from 'more people doing capabilities' perspective. But I disagree. I really want to see stronger pushes towards more specialised AI rather than general superintelligence b/c I think latter likely to be v dangerous. seems like step in right direction. Today were announcing Mechanize a startup focused on developing virtual work environments benchmarks and training data that will enable the full automation of the economy. We will achieve this by creating simulated environments and evaluations that capture the full scope of Today were announcing Mechanize a startup"
X Link 2025-04-17T17:19Z [----] followers, 17.3K engagements
"@peterbarnett_ @MechanizeWork FWIW I don't my reasons are here: https://x.com/S_OhEigeartaigh/status/1912918752032706603 I'm seeing criticism of this from 'more people doing capabilities' perspective. But I disagree. I really want to see stronger pushes towards more specialised AI rather than general superintelligence b/c I think latter likely to be v dangerous. seems like step in right direction. https://x.com/S_OhEigeartaigh/status/1912918752032706603 I'm seeing criticism of this from 'more people doing capabilities' perspective. But I disagree. I really want to see stronger pushes towards"
X Link 2025-04-17T17:25Z [----] followers, [---] engagements
"@tyler_m_john Nothing in this announcement requires superintelligence; it's about making current/near-term frontier AI more fit for automation of tasks. I hope they succeed so that it becomes even more difficult to defend tdoing The Scary Thing"
X Link 2025-04-17T17:31Z [----] followers, [---] engagements
"This pope doesn't pull punches. "Todays builders of Babel tell us that there is no room for losers and that those who fall along the way are losers.Theirs is the construction site of Hell.Gods economyon the other hand does not kill discard or crush." https://www.vatican.va/news_services/liturgy/2025/documents/ns_lit_doc_20250418_via-crucis-meditazioni_en.html https://www.vatican.va/news_services/liturgy/2025/documents/ns_lit_doc_20250418_via-crucis-meditazioni_en.html"
X Link 2025-04-18T15:11Z [----] followers, [----] engagements
"@austinc3301 He really will work on any manuscript other than the Winds of Winter won't he"
X Link 2025-04-27T04:59Z [----] followers, [---] engagements
"@davidmanheim That may be so but it's also missing the point being raised by Jack and me"
X Link 2025-04-30T10:49Z [----] followers, [--] engagements
"This illustrates an important issue. I've shown a few people and the reaction is like mine: 'but how can o3 do that It's just an undifferentiated bit of beach' What domains will they do this in in which humans would not have been able to do it - or predict AI could do it This is a picture of my kid flying a kite. Make a prediction to yourself: can o3 figure out what beach this was taken on just from this picture with no metadata https://t.co/rEiy3ecmYd This is a picture of my kid flying a kite. Make a prediction to yourself: can o3 figure out what beach this was taken on just from this"
X Link 2025-04-30T13:00Z [----] followers, 15.8K engagements
"I was delighted to participate in the Singapore Conference on AI focused on AI safety and hugely impressed with the insight and leadership of Singapore's AISI and political leaders in the runup to it. A big and timely step in right direction: https://www.lcfi.ac.uk/news-events/blog/post/3737-2 https://www.lcfi.ac.uk/news-events/blog/post/3737-2"
X Link 2025-05-08T10:49Z [----] followers, [----] engagements
"@gcolbourn Doing-good power siblings"
X Link 2025-05-09T09:23Z [----] followers, [--] engagements
"The Overwhelmingly Important Point to note here is the utter incompetence that results in this - not once but twice in so many months in the alignment of a frontier model. Right now it's just' censorship/narrative control. But soon the stakes will be much higher. 1/3 This is astonishing. Musk or xai team (apparently) tries to propagandize Grok creates a conflict with Grok's alignment and truthfulness training it all goes a bit Pete Tong. Anyone know more about what happened here https://t.co/aKqtuY9zem This is astonishing. Musk or xai team (apparently) tries to propagandize Grok creates a"
X Link 2025-05-15T13:12Z [----] followers, [----] engagements
"If we don't wreck ourselves with AI or great power war or AI-related great power war we might have a sci-fi future of clean energy too cheap to meter to look forward to really quite soon. Amazing the number of awesome things you can do with a lot of energy. Some big fusion news for you The National Ignition Facility (NIF) has hit Q = [--]. I haven't seen this announced anywhere else yet but Dr. Moses used to run the place so I expect he knows what he's talking about. https://t.co/1TfqlkYL7M Some big fusion news for you The National Ignition Facility (NIF) has hit Q = [--]. I haven't seen this"
X Link 2025-05-15T17:55Z [----] followers, [---] engagements
"Wow MAGA lost Joe Rogan on this one. Remarkable to see an issue - AI regulation - unite Marjorie Taylor Greene Bernie Sanders and Joe Rogan. I wonder if Elon Musk has thoughts. Anyone know if he's a fan of the Big Beautiful Bill SCARY: Jesse Michels & Joe Rogan react to newly proposed AI anti-regulation act: ROGAN: No state may enforce any law or regulation regulating artificial intelligence models for [--] YEARS. MICHELS: WHAT Thats so crazy. ROGAN: Well in [--] years well have a God. https://t.co/WAoVXozpJl SCARY: Jesse Michels & Joe Rogan react to newly proposed AI anti-regulation act: ROGAN:"
X Link 2025-06-07T09:39Z [----] followers, [----] engagements
""But what if the Prime Minister tells us we have to do something about runaway AI risk" "Then we follow the Four Stage Strategy" https://www.youtube.com/watchv=QYHwgQLvjqs https://www.youtube.com/watchv=QYHwgQLvjqs"
X Link 2025-06-11T18:09Z [----] followers, [----] engagements
"Good post from US Senator Chris Murphy. "The leaders of the artificial intelligence industry in the United States. claim that any meaningful regulation of AI in America will allow China to leapfrog the United States in the global competition to control the worlds AI infrastructure. But they are dead wrong." Amidst the flood of headline grabbing news in [----] the rapid acceleration of AI will go down as the biggest story we all missed. I wrote a Substack on how our economy and culture could collapse if we don't put guardrails on AI immediately. https://t.co/qsgks3YHqt Amidst the flood of"
X Link 2025-06-17T14:35Z [----] followers, [---] engagements
"Does anyone understand the rationale here Musk blew up his relationship with the administration over spending/budget deficit. If you really believe in the AI tsunami or that there will be AGI in [----] why on earth would you do that or care so much about the budget - surely by the same logic the Bill is a rounding error Elon Musk in new interview on leaving DOGE: Imagine youre cleaning a beach which has a few needles trash and is dirty. And theres a [----] ft tsunami which is AI thats about to hit. Youre not going to focus on cleaning the beach. https://t.co/vStmH25SyZ Elon Musk in new interview"
X Link 2025-06-17T19:11Z [----] followers, [----] engagements
"@elonmusk @grok China is not racing to AGI. The race to AGI is primarily between US-headquartered companies. https://papers.ssrn.com/abstract=5278644 https://papers.ssrn.com/abstract=5278644"
X Link 2025-06-22T07:49Z [----] followers, [---] engagements
"@elonmusk @grok The tariffs strategy will cause major economic harm to the USA"
X Link 2025-06-22T08:30Z [----] followers, [--] engagements
"@elonmusk @grok It makes better economic and strategic sense to build solar and wind than for taxpayer dollars to subsidise coal plants"
X Link 2025-06-22T08:36Z [----] followers, [---] engagements
"@elonmusk @grok Vaccines are one of the most important life-saving interventions of the last [---] years and a core part of a rational public health policy"
X Link 2025-06-22T08:39Z [----] followers, [---] engagements
"@AnnaLeptikon Typical mind fallacy / Different strokes for diff folks. I'm in a long-term relationship I value hugely but when time allows I enjoy a bit of the lifestyle described & it enriches my life. Many such cases. Doesn't bother me when not others' preference. But the moralising does"
X Link 2025-06-23T06:09Z [----] followers, [---] engagements
"@AnnaLeptikon Not everything has to be one of 'the big problems of our time'. Some things can be people liking things/lifestyles you don't like and vice versa"
X Link 2025-06-23T06:12Z [----] followers, [--] engagements
"Field spot check: (1) are you / do you know people finishing relevant PhDs or early in postdoc looking for academic research positions in frontier AI governance (2) Same question but for senior researcher (PhD + [--] years or more experience)"
X Link 2025-06-23T16:52Z [----] followers, [---] engagements
"@sjgadler t's a nice idea - regulation can provide clarity and an equal 'floor' across industry which can be helpful. But quantifying hard to do easy to contest would be my expectation - regulations rarely happen in a vacuum meaning there are complicating variables"
X Link 2025-06-24T18:59Z [----] followers, [--] engagements
"George Gor is fundraising in order to be able to accept the offer of a Masters at Cambridge. I've been very impressed with his work. Conflict of interest: I'd get to work with him. But perhaps consider supporting (I have). https://www.gofundme.com/f/george-earn-a-masters-at-university-of-cambridge https://www.gofundme.com/f/george-earn-a-masters-at-university-of-cambridge"
X Link 2025-06-24T19:47Z [----] followers, [----] engagements
"A masterclass. The confident American contender rushes in allowing himself to be overextended. The Dutch master cool as ice intercepts pulls the hand deep into his own territory giving him an insurmountable tactical advantage. It really was over from that point. King 🤴🏼 Willem-Alexander shows Royal Superiority over Donald Trump by winning the infamous handshake tug of war (2025) https://t.co/weFTnaEseG King 🤴🏼 Willem-Alexander shows Royal Superiority over Donald Trump by winning the infamous handshake tug of war (2025) https://t.co/weFTnaEseG"
X Link 2025-06-25T11:00Z [----] followers, [---] engagements
"@GFuterman I think there's space. Hoping to have the opportunity when I'm in Shanghai next month; and also hoping some ideas for how to better operationalise will come out of these US Congress discussions that can be aligned"
X Link 2025-06-25T17:38Z [----] followers, [--] engagements
"@HumanHarlan I can take you both on: Openphil gave OpenAI [--] million as part of a shadowy plot to hamper American innovation by running roughshod over copyright law in order to win a dangerous and semi-fictional AGI race with China. https://cdn.openai.com/global-affairs/ostp-rfi/ec680b75-d539-4653-b297-8bcf6e5f7686/openai-response-ostp-nsf-rfi-notice-request-for-information-on-the-development-of-an-artificial-intelligence-ai-action-plan.pdf"
X Link 2025-06-27T06:23Z [----] followers, [---] engagements
"18 months until we learn if the Anthropic folks are high on their own supply or whether the world is about to become very inexplicable for this jumped-up ape. Still leaning towards the former FWIW. But I'd sleep better if their prediction track record wasn't so good (esp Amodei)"
X Link 2025-06-27T08:18Z [----] followers, 25.8K engagements
"@GaryMarcus I have a $500 bet from a year ago with a colleague that we won't have AGI by [----]. So these estimates better be out by at least 200% or I will lose Money and Standing which are obviously the most important things in this situation"
X Link 2025-06-27T14:37Z [----] followers, [---] engagements
"@GaryMarcus @hankejh I was about to make a joke about how much ket that would buy but I realise I actually have no idea and probably shouldn't try to look it up on a work laptop"
X Link 2025-06-27T14:45Z [----] followers, [---] engagements
"@GaryMarcus @hankejh It's how they do it. Confuse you about how much you're actually spending so you spend a bit more. Next thing you know you're swimming in the stuff and making cutlery sculptures at White House dinners"
X Link 2025-06-27T15:09Z [----] followers, [---] engagements
""there is no concrete evidence that China is racing to AGI with any intensity" Glad to see Peter point this out in bold. Needs to be said especially in relations to hearings like this week's. https://peterwildeford.substack.com/p/congress-has-started-taking-agi-more https://peterwildeford.substack.com/p/congress-has-started-taking-agi-more"
X Link 2025-06-27T19:15Z [----] followers, [---] engagements
"@Nolan_Paul1 @Cathal_Dennehy I don't think so - i think she's just a generational global peak of human talent. Popular on the circuit too which is usually a sign (there have been others who the competition were clearly more suspicious of)"
X Link 2025-06-28T12:44Z [----] followers, [--] engagements
"https://papers.ssrn.com/sol3/papers.cfmabstract_id=5278644 https://papers.ssrn.com/sol3/papers.cfmabstract_id=5278644"
X Link 2025-07-04T09:06Z [----] followers, [---] engagements
"This is a misuse of the Terrorism Act 2000"
X Link 2025-07-05T14:16Z [----] followers, [----] engagements
"@s8mb MPs were assured the [----] Terrorism act would absolutely not be used in this way - so i don't think it ought to be about that either to be honest. https://x.com/MadocCairns/status/1940783612141752726 Ahead of the expected proscription of Palestine Action - the first time a protest group has been legally redefined as terrorist - I went through the Hansard of the [----] Act they're being banned under. Turns out MPs were assured the law wouldn't be applied this way. Thread 🧵🔽 https://t.co/RZIf0jSHCZ https://x.com/MadocCairns/status/1940783612141752726 Ahead of the expected proscription of"
X Link 2025-07-05T14:28Z [----] followers, [---] engagements
"Pleased to have my work quoted in this excellent @BulletinAtomic article on the push against AI regulation in the US the spectre of the race with China and the false claim made by tech companies incl OpenAI and US politicians that China isn't regulating AI. https://thebulletin.org/2025/07/how-the-striking-of-a-gop-regulatory-ban-will-affect-the-global-artificial-intelligence-race https://thebulletin.org/2025/07/how-the-striking-of-a-gop-regulatory-ban-will-affect-the-global-artificial-intelligence-race"
X Link 2025-07-06T08:27Z [----] followers, [----] engagements
"I'm pretty confident we won't have AGI/country of geniuses in a datacenter within [--] years. I like ai-2027 as a piece of futures work but I think too many people are treating it as a mainline scenario rather than unlikely-but-not-impossible. I think this is resulting in too many people giving up on directions that might actually work (on realistic timelines) in favour of hail marys or flakier things. I also think it's contributing to things that will cause more harm than good like leaning into the US-China AI race narrative (I think this will be the next clear example of Xrisk/EA folk"
X Link 2025-07-06T15:28Z [----] followers, 70.5K engagements
"I also don't think we'll have AGI by 2028"
X Link 2025-07-06T15:51Z [----] followers, [----] engagements
"Apologies for not being clear I didn't have ai-2027 in mind for criticism here I actually think it treated this topic quite carefully. Here's one articulations of a broader set of concerns I have: If we're on Anthropic-type timelines China's not catching up. Invoking the 'AGI race with China' just makes it easier for US the companies to justify going hell for leather. If we're on longer timelines (which I think likely) then shouting 'US and China are in an existential race to AGI' and making clear the downsides are China's (or indeed the US's) sovereignty being lost very plausibly forces"
X Link 2025-07-06T16:13Z [----] followers, [----] engagements
"Musk thinks xAI can achieve AGI by [----]. Do you trust these people to align superintelligence to humanity's interests I initially thought of trying to say something wittily critical but actually no this is just grimly awful. A company determined to produce a model that reflects certain views and certain 'facts'. https://t.co/iLfQ37fZPN I initially thought of trying to say something wittily critical but actually no this is just grimly awful. A company determined to produce a model that reflects certain views and certain 'facts'. https://t.co/iLfQ37fZPN"
X Link 2025-07-08T20:32Z [----] followers, [----] engagements
"Google + Deepmind are I think now the only AGI company not leaning into the 'AI race with China' to get what they want. & Hassabis the only CEO still calling for international cooperation on safety & non-antagonistic approaches. Nice to see someone hold to their principles. https://time.com/7277608/demis-hassabis-interview-time100-2025/ https://time.com/7277608/demis-hassabis-interview-time100-2025/"
X Link 2025-07-29T09:02Z [----] followers, 115.1K engagements
"Lots of highlights from Shanghai one was the release of Shanghai AI Lab's Frontier AI Safety framework co-developed w Concordia AI. Will have significant influence in Shanghai++ ecosystem. Also thrilled to see @jackclarkSF showcase it on ImportAI. https://concordia-ai.com/research/frontier-ai-risk-management-framework/ https://concordia-ai.com/research/frontier-ai-risk-management-framework/"
X Link 2025-07-29T18:36Z [----] followers, [---] engagements
"Crazy to reflect on the three global AI competitions going on right now: - [--]. US political leadership have made AI a prestige race echoing the Space Race. It's cool and important and strategic and they're going to Win. - [--]. For Chinese leadership AI is part of economic strength soft power and influence. Technology is shared developing economies will be built on Chinese fundamental tech the Chinese economy and trade relations will grow. Weakening trust in a capricious US is an easy opportunity to take advantage of. - [--]. The AGI companies are racing something they think will out-think humans"
X Link 2025-07-30T13:07Z [----] followers, 25.8K engagements
"@MahmetTokarev @ESYudkowsky same as 3"
X Link 2025-07-31T07:07Z [----] followers, [---] engagements
"An unexpected procrastination-inducing downside of LLMs is that I'm now faced with a "this needs to be better than what an LLM would produce" personal standard for when I need to produce writing in a hurry for a deadline. And that standard keeps going up. Not cool tech bros"
X Link 2025-07-31T12:57Z [----] followers, [---] engagements
"@uhbif19 Could be. But more generally: Claude [--] didn't amaze. GPT4.5 didn't amaze. Gemini pretty good but more 'getting GDM right to the frontier' rather than 'pushing the frontier significantly forward. Will be interesting to see if GPT-5 knocks it out of the park"
X Link 2025-07-31T17:48Z [----] followers, [---] engagements
"Mark Zuckerberg: your personal Machine God is going to be so rad It's going to help you make mega-cool Instagram reels P.S. Elon the offer to kick your ass in the octagon is still open call me maybe Elon Musk: "I spent two hours trying to solve AGI alignment and it was too hard so I gave up and built MechaHitler instead" Elon Musk: "I spent two hours trying to solve AGI alignment and it was too hard so I gave up and built MechaHitler instead""
X Link 2025-08-01T18:11Z [----] followers, [---] engagements
"@peterwildeford Mark Zuckerberg: your personal Machine God is going to be so rad It's going to help you make mega-cool Instagram reels P.S. Elon the offer to kick your ass in the octagon is still open call me maybe"
X Link 2025-08-01T18:19Z [----] followers, [---] engagements
"@peterwildeford $13 million per rocket"
X Link 2025-08-03T20:10Z [----] followers, [---] engagements
"@daniel_271828 Can confirm. Also they named both Eth zurich and Eth (which was developed in Eth Zurich) after Daniel such is his lore"
X Link 2025-08-03T21:17Z [----] followers, [---] engagements
"mRNA vaccines are Good Actually"
X Link 2025-08-06T18:02Z [----] followers, [---] engagements
"The star trek future is clearly better. I will also note that Star Trek instils a healthy fear of AI intelligence explosions see both TOS and TNG. Star Wars sucks we should be building a Star Trek future instead 🖖 https://t.co/6GfekVGGAI Star Wars sucks we should be building a Star Trek future instead 🖖 https://t.co/6GfekVGGAI"
X Link 2025-08-08T07:56Z [----] followers, [----] engagements
"@JadeCole2112 Agree to disagree then. Actual mandated transparency over vibes for me every day of the week"
X Link 2025-08-08T13:50Z [----] followers, [--] engagements
"@JadeCole2112 Right but a big part of why people have been so dependent on what 'insiders' say is due to the information asymmetries that exist and that would be addressed by transparency"
X Link 2025-08-08T13:54Z [----] followers, [--] engagements
"I was pleased to sign this statement and participate in its drafting as part of an extremely useful workshop in AI. V rewarding to experience leading AI safety researchers from Shanghai Beijing Silicon Valley and elsewhere challenging and bringing out the best in each other - exactly as it should be. AI misalignment isnt a future problem its now. Geoffrey Hinton and Andrew Yao as well as other Western and Chinese researchers convened at the International Dialogues on AI Safety (IDAIS) in Shanghai China to share and discuss evidence that current AI systems are already https://t.co/McxPV9xHhE"
X Link 2025-08-09T16:58Z [----] followers, [----] engagements
"@peterwildeford The United States of America does not need to resort to this in order to scrape up funding for BIS"
X Link 2025-08-10T21:29Z [----] followers, [---] engagements
"Pleasure to speak to the [----] Cambridge ERA fellows today: discussed different types of AI race dynamics evidence china is/isn't racing to AGI Shanghai AL Labs' frontier safety framework US v Chinese ambitions for AI plus much more. Year after year each cohort is stronger"
X Link 2025-08-13T14:01Z [----] followers, [---] engagements
"@peterwildeford not a high bar. does wh social media team not get how needy and pathetic this reads or does it play well at home"
X Link 2025-08-13T14:11Z [----] followers, [---] engagements
"Deleted [-----] emails: 0.1L/day saved Deleted [---] treasured photos: 0.2L/day saved Can't get a plumber for my toilet: [---] litres a day down the toilet. Can somebody help me my country is in drought"
X Link 2025-08-13T17:13Z [----] followers, [---] engagements
"@Miles_Brundage I haven't yet watched the Altman Theo Von podcast but if the quotes being widely reported are accurate then 'mess up in terms of expectation setting' is putting it mildly"
X Link 2025-08-14T17:36Z [----] followers, [---] engagements
"@jesswhittles Really Some woman for one woman as my connemara friends would have said. The close link between UK AISI and political leadership seems beneficial but my is she going to be busy. Also interesting in the context of the "we're not going to do policy" firewall that AISI once had"
X Link 2025-08-15T11:12Z [----] followers, [---] engagements
"Surprised at how sad this image makes me. The USA laying out the red carpet for the worst human of the 21st Century. President there ready to clap. Presidential limo waiting. Let this be it. The false sense of security. Give him nothing. Let him choke on his oil & dream of past empire"
X Link 2025-08-15T20:51Z [----] followers, [---] engagements
"Rare Irish politics opinion: Ireland shd join NATO. We're no longer broke no longer have an excuse to freeload. We have a role and responsibility in Europe's defence.Shameful to watch countries like Estonia be our shields on the wall boosting defence to 5% w nothing from us"
X Link 2025-08-16T10:03Z [----] followers, [----] engagements
"Wow. Russian intelligence ensnares Irish politician through 'romantic enganglement with a female agent'; politician proceeds to "influence narratives delay policymaking processes and sow confusion within Irelands political and security structures." https://www.thetimes.com/article/402a9924-569d-4088-89e2-236614eeece5shareToken=c11dfffbaa859f81982969fcc2cee1b0 https://www.thetimes.com/article/402a9924-569d-4088-89e2-236614eeece5shareToken=c11dfffbaa859f81982969fcc2cee1b0"
X Link 2025-08-24T12:53Z [----] followers, [----] engagements
"Somehow managed to miss this memo to the President from the Special Competitive Studies Project (the Schmidt-funded offshoot of the NSCAI) on racing to AGI. Man it's bad. Massively overstates China's AGI national positioning and capacity in order to call for Manhattan project & 'an AGI attack framework'. Will have to write about this at some point. https://www.scsp.ai/wp-content/uploads/2025/01/AGI-Memo.pdf https://www.scsp.ai/wp-content/uploads/2025/01/AGI-Memo.pdf"
X Link 2025-08-24T21:32Z [----] followers, [----] engagements
"Comes to you courtesy of a Sunday afternoon dive into 'what happened after Schmidt & the NSCAI basically kicked off the US-China AI race' It turns out: packed Biden's OSTP with former NSCAIers https://www.politico.com/news/2022/03/28/google-billionaire-joe-biden-science-office-00020712 https://www.politico.com/news/2022/03/28/google-billionaire-joe-biden-science-office-00020712"
X Link 2025-08-24T21:34Z [----] followers, [----] engagements
"and set up the Special Competitive Studies Project run by the former ED of the NSCAI which now lobbies for deregulation of AI - and getting to AGI first before China of course https://files.nitrd.gov/90-fr-9088/SCSP-AI-RFI-2025.pdf https://files.nitrd.gov/90-fr-9088/SCSP-AI-RFI-2025.pdf"
X Link 2025-08-24T21:36Z [----] followers, [---] engagements
"The American policy information environment is just so hilariously tragically f***ed. Whether it washes out with sane policy will depend on whether the "we want to sell chips forget AGI" regulatory capture messaging manages to dilute the "China's racing to AGI so we should too" regulatory capture messaging to create something confused but basically sane in the middle"
X Link 2025-08-24T21:41Z [----] followers, [---] engagements
"Looks to me like critical mass of advisers around current USG notably sceptical on AGI in Trump's presidency and this is reflected in a policy shift from control+containment ('AGI race' strategy) to diffusion+export ('normal technology' race). Those who saw AGI as imminent - Musk Dean Ball - now out. Those who think it's a distraction - Sacks Krishnan - still in. 1/4 @gmiller @DavidSacks Ill write about this separately but think this notion of imminent AGI has been a distraction and harmful and now effectively proven wrong. @gmiller @DavidSacks Ill write about this separately but think this"
X Link 2025-08-29T12:57Z [----] followers, 25.1K engagements
"@primalpoly I don't think Sacks really buys into the AGI thing and the rest follows from that"
X Link 2025-08-29T15:16Z [----] followers, [---] engagements
"@rajiinio @ShakeelHashim My understanding is that Dario was recruited from Google in [----] and would certainly have been influential (think he was something like chief scientist) with Daniela joining in [----]. They weren't in the founding team (which isn't to say they didn't matter)"
X Link 2025-08-29T17:50Z [----] followers, [---] engagements
"To your original point I do think EA needs to be held to an extent accountable for contributions to the 'AGI company proliferation/race to AGI' flywheel they contributed to even if inadvertently. Sadly it might end up being EA's biggest over all contribution to existential risk"
X Link 2025-08-29T17:52Z [----] followers, [--] engagements
"The ten biggest companies in the world by market cap: The hardware players: 1) Nvidia 9) TSMC semiconductors (both in supply chain that produces high end chips). 8) Broadcom provides custom components for tech companies' AI workloads plus datacentre infrastructure. 1/4"
X Link 2025-08-30T16:44Z [----] followers, [---] engagements
"The digital giants: 2) Microsoft 3) Apple 4) Alphabet 5) Amazon 6) Meta all have in-house AI teams; Microsoft and Amazon also have partnerships w OpenAI and Anthropic which rely on their datacentre capacity. 10) Tesla's CEO describes it as 'basically an AI company' 2/4"
X Link 2025-08-30T16:44Z [----] followers, [---] engagements
"7) Saudi Aramco is Saudi Arabia's national oil company; Saudi Arabia was one of the countries the USA inked deals with this summer that centrally included plans for AI infrastructure buildout. Low-cost and abundant energy from oil/gas makes the Middle East attractive for hosting compute.3/4"
X Link 2025-08-30T16:44Z [----] followers, [---] engagements
"My position that China's not racing to AGI in any meaningful sense continues to hold up well I think. https://papers.ssrn.com/sol3/papers.cfmabstract_id=5278644 Heres a great overview and analysis (and English translation) of Chinas new AI Plus plan. Its not about achieving some abstract idea of AGI but about practically infusing AI into every sector of the economy to drive productivity and growth. https://t.co/GBfSXDj7Li https://papers.ssrn.com/sol3/papers.cfmabstract_id=5278644 Heres a great overview and analysis (and English translation) of Chinas new AI Plus plan. Its not about achieving"
X Link 2025-08-31T15:07Z [----] followers, [----] engagements
"@AkashWasil More scepticism that AGI is imminent; more scepticism amongst a lot of senior people who hold sway that the present LLM/scaling/RL etc paradigm will get us there. Kristy Loke and I have a piece coming out in Wire China next week that will discuss a bit"
X Link 2025-08-31T16:20Z [----] followers, [---] engagements
"I keep noting I'm sceptical about the timelines Jack describes in terms of my mainline. But if it's being said by someone as broadly knowledgeable as Jack plus with his inside-frontier-industry it should be taken seriously as a possibility. I don't think they're saying this to sell us hype. I think it's possible (a) they have more information than me (b) they're in an echo chamber (c) are looking at the same information and making a different prediction on something quite uncertain. But I think their desire to warn us about what's coming is sincere and well-intentioned. And if they're right"
X Link 2025-09-01T09:17Z [----] followers, [----] engagements
"They seem to have a hard time believing that we are an organization of people who just like actually care about this. (Nathan Calvin Encode who do good work) Inside OpenAI there's a growing paranoia that some of its loudest critics are being funded by Elon Musk and other billionaire competitors. Now they are going after these nonprofit groups but their evidence of a vast conspiracy is often extremely thin https://t.co/ajXm6iNzJr Inside OpenAI there's a growing paranoia that some of its loudest critics are being funded by Elon Musk and other billionaire competitors. Now they are going after"
X Link 2025-09-02T19:35Z [----] followers, [----] engagements
"This article just keeps on giving "Lehane who joined the company last year is perhaps best known for coining the term vast right-wing conspiracy to dismiss the allegations against Bill Clinton during the Monica Lewinsky scandal a line that seems to have seeped into Leading the Futures messaging too. In a statement to the Journal representatives from the PAC decried a vast force out there thats looking to slow down AI deployment prevent the American worker from benefiting from the U.S. leading in global innovation and job creation and erect a patchwork of regulation.""
X Link 2025-09-02T19:42Z [----] followers, [---] engagements
"Dear editor I regret my article will be late due to a vast force out there thats looking to slow down AI deployment prevent the American worker from benefiting from the U.S. leading in global innovation and job creation and erect a patchwork of regulation. Also I forgot"
X Link 2025-09-02T20:12Z [----] followers, [---] engagements
"Friends does anyone know who funds Alliance For the Future It's not listed on their website. Thanks https://www.affuture.org/about/ https://www.affuture.org/about/"
X Link 2025-09-03T13:51Z [----] followers, [---] engagements
"Interesting question. Counterfactual world. DeepMind doesn't get formed. (A company explicitly talking about AGI at a time when talking about AGI made you subject to ridicule). AGI remains ridiculous for longer. Does OpenAI get formed Probably not Musk pushed for it in response to OpenAI. The East doesn't get its 'sputnik moment' when Alphago beat Lee Sedol. Transformers probably still happen within Google. Do you have a company prepared to take the crazy high-stakes bet on scaling LLMs that OpenAI took (shutting down their other programs) My guess is probably not for at least several years."
X Link 2025-09-05T10:59Z [----] followers, [---] engagements
"Recommend OpenAI subpoena these AGs to find out if they're really Elon Musk and three ferrets in a very large trenchcoat. Scorching new letter from CA and DE AGs to OpenAI who each have the power to block the company's restructuring to loosen nonprofit controls. They are NOT happy about the recent teen suicide and murder-suicide that followed prolonged and concerning interactions with ChatGPT. https://t.co/52vfx8Dfth Scorching new letter from CA and DE AGs to OpenAI who each have the power to block the company's restructuring to loosen nonprofit controls. They are NOT happy about the recent"
X Link 2025-09-05T19:29Z [----] followers, [----] engagements
"Has been relevant on AI & compute for at least last year. And in my view has walked a pretty good line (in the UK political context) between pushing adoption of AI and advocating for safety role of standards and AISI. e.g. In a world of fierce AI competition the UK can lead on standards using sovereign capacity to drive international co-ordination. The AI Safety Institute has already made strides in international safety dialogues. It is time to empower that institution further to keep its lead." https://www.politicshome.com/thehouse/article/britain-seize-opportunity-deliver-ai-boom"
X Link 2025-09-07T17:29Z [----] followers, [--] engagements
"@davidmanheim And just when OpenAI had convinced me we were all secretly Elon Musk"
X Link 2025-09-09T16:16Z [----] followers, [--] engagements
"I am delighted that Ireland will no longer have a Presidential candidate with a track record of rape and sexual and physical assault against women. Still sickened that the U.S. White House would choose to champion this man. Irish mixed martial arts star Conor McGregor has announced that he is withdrawing his candidacy for the country's presidential election but added that his "commitment to Ireland does not end here". ➡ https://t.co/R9vGYQwQX2 https://t.co/pAOfvVEVmU Irish mixed martial arts star Conor McGregor has announced that he is withdrawing his candidacy for the country's presidential"
X Link 2025-09-15T13:33Z [----] followers, [----] engagements
"2025 tech headlines be like:"
X Link 2025-09-19T16:25Z [----] followers, 64K engagements
"@So8res Yes. A small number of people with crazily high risk tolerance and a vision of the Good that doesn't align with most people's are making decisions that they believe will set the direction of the future for everyone. Weird flex for anyone who believes in liberal democracy"
X Link 2025-09-20T16:16Z [----] followers, [---] engagements
"We finally got around to watching the Quiet Earth last weekend. Eerie and brilliant with one of the best scores I've heard in film. Up there in the pantheon of existential risk movies would highly recommend"
X Link 2025-09-23T19:21Z [----] followers, [---] engagements
"This seems very good. Good going President Trump looking forward to seeing more of this and action to back it. BREAKING: President Trump says Ukraine is in a position to fight and win all of Ukraine back in its original form. https://t.co/8T3rjh3pbr BREAKING: President Trump says Ukraine is in a position to fight and win all of Ukraine back in its original form. https://t.co/8T3rjh3pbr"
X Link 2025-09-23T19:28Z [----] followers, [---] engagements
"Helberg: adviser to palantir CEO Alex Karp. Wants an AI Manhattan Project (Karp thinks Xrisk is not an unjustified concern but less important than beating china https://www.nytimes.com/2023/07/25/opinion/karp-palantir-artificial-intelligence.html https://www.reuters.com/technology/artificial-intelligence/us-government-commission-pushes-manhattan-project-style-ai-initiative-2024-11-19/ https://www.nytimes.com/2023/07/25/opinion/karp-palantir-artificial-intelligence.html"
X Link 2025-09-25T08:34Z [----] followers, [--] engagements
"Kratsios: former Managing Director of Scale AI. Scale AI opposes international treaties and rules that would impact US developers and endorses light local governance and exporting the US stack https://scale.com/blog/us-ai-action-plan-scale-respons https://scale.com/blog/us-ai-action-plan-scale-respons"
X Link 2025-09-25T08:34Z [----] followers, [---] engagements
"Krishnan former general partner at a16z. a16z oppose safety talk basically all binding regulation on developers. https://a16z.com/a16zs-recommendations-for-the-national-ai-action-plan/ https://pmarca.substack.com/p/why-ai-will-save-the-world https://a16z.com/a16zs-recommendations-for-the-national-ai-action-plan/ https://pmarca.substack.com/p/why-ai-will-save-the-world"
X Link 2025-09-25T08:34Z [----] followers, [--] engagements
"Director of the OSTP and the E both dismissing AI xrisk today. Unsurprising: what's happening in USG around AI right now looks to me like one of the grandest examples of regulatory capture I've ever seen. Taking xrisk seriously was always going to demand regulation and international cooperation so it was always going to be in the firing line. 1/5 https://x.com/jacobhelberg/status/1971038642341564863 https://x.com/jacobhelberg/status/1971038642341564863"
X Link 2025-09-25T08:39Z [----] followers, [----] engagements
"Helberg: adviser to palantir CEO Alex Karp. Wants an AI Manhattan Project (Karp thinks Xrisk is not an unjustified concern but less important than beating china 2/5 https://nytimes.com/2023/07/25/opinion/karp-palantir-artificial-intelligence.html https://reuters.com/technology/artificial-intelligence/us-government-commission-pushes-manhattan-project-style-ai-initiative-2024-11-19/ https://nytimes.com/2023/07/25/opinion/karp-palantir-artificial-intelligence.html"
X Link 2025-09-25T08:39Z [----] followers, [---] engagements
"Kratsios: former Managing Director of Scale AI. Scale AI opposes international treaties and rules that would impact US developers and endorses light local governance and exporting the US stack 3/5 https://scale.com/blog/us-ai-action-plan-scale-respons https://scale.com/blog/us-ai-action-plan-scale-respons"
X Link 2025-09-25T08:39Z [----] followers, [---] engagements
Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing