I didn't want to post on Grok safety since I work at a competitor, but it's not about competition.
I appreciate the scientists and engineers at @xai but the way safety was handled is completely irresponsible. Thread below.
It was great to be part of this statement. I wholeheartedly agree.
It is a wild lucky coincidence that models often express dangerous intentions aloud, and it would be foolish to waste this opportunity. It is crucial to keep chain of thought monitorable as long as possible
It was great to be part of this statement. I wholeheartedly agree.
It is a wild lucky coincidence that models often express dangerous intentions aloud, and it would be foolish to waste this opportunity. It is crucial to keep chain of thought monitorable as long as possible
The questions is not whether LLMs *can* generalise out of distribution (they can - see image), but by *how much*. Here's Claude giving a reasonable response to an out of distribution prompt (a list of 20 randomly generated words).
The questions is not whether LLMs *can* generalise out of distribution (they can - see image), but by *how much*. Here's Claude giving a reasonable response to an out of distribution prompt (a list of 20 randomly generated words). https://t.co/6yW8ItjtN7
This is either phenomenal progress in robotics, significant progress in AI video generation, or mild progress in one skinny man's ability to hide under his shirt and run down a hill with a robot hat on.
This is either phenomenal progress in robotics, significant progress in AI video generation, or mild progress in one skinny man's ability to hide under his shirt and run down a hill with a robot hat on.
Pre-registering my prediction that by 2035 most people will be more concerned by questions like "is this celebrity a bad person because of how they treated their AI?" than the question "are AIs conscious".
I wonder how many people know that by default OpenAI not only *saves* what you type in but *uses it for training*, meaning OpenAI may only be able to fully delete your data by *deleting their entire model*. NYT vs OpenAI is not a zero sum game.
I wonder how many people know that by default OpenAI not only *saves* what you type in but *uses it for training*, meaning OpenAI may only be able to fully delete your data by *deleting their entire model*. NYT vs OpenAI is not a zero sum game.
The LinkedIn algorithm is bonkers. Go away for a week and now 40 people I've never heard of (bots?) are arguing over my post, one with the job title "Keynote Speaker"... 🙃
55 Followers 213 FollowingPhD student in Foundational AI @ucl @ai_ucl @uclcs
Enrichment Fellow @turinginst
2x ML Research Intern at Apple working on Differential Privacy
862 Followers 7K FollowingBITCOIN AND DIGITAL ASSET LEGISLATION, THE WORLD IS WAKING UP, THE COLLECTIVE CONSIOUSNESS OF HUMANITY EVOLVING. I’M HERE TO HELP YOU MOVE INTO QUANTUM
3K Followers 3K Following#AcceleratingSafeAI. Background in market and AI product development for Market Primes, Private Equity & Government Investors. All comments are mine alone.
10K Followers 1K FollowingWe work with @UCL staff, students and businesses to turn knowledge and ideas into solutions that benefit us all.
On LinkedIn: UCL Innovation & Enterprise
16K Followers 357 FollowingRuns an AI Safety research group in Berkeley (Truthful AI) + Affiliate at UC Berkeley. Past: Oxford Uni, TruthfulQA, Reversal Curse. Prefer email to DM.
5K Followers 325 FollowingCEO@Redwood Research (@redwood_ai), working on technical research to reduce catastrophic risk from AI misalignment. [email protected]
1K Followers 701 FollowingEuropean Commission (AI Office). PhD student @CambridgeMLG. Here to discuss ideas and have fun. Posts are my personal opinions; I don't speak for my employer.
25K Followers 206 FollowingWorking towards the safe development of AI for the benefit of all @UMontreal, @LawZero_ & @Mila_Quebec
A.M. Turing Award Recipient and most-cited AI researcher.
109K Followers 6K FollowingSearching for the numinous
🇦🇺 🇨🇦, currently live in 🇺🇸
Research @AsteraInstitute
https://t.co/maezekzRUb
https://t.co/2dWwZKrvrn
254K Followers 567 Followingnew book *Talent: How to Identify Energizers, Winners, and Creatives Around the World*, https://t.co/7bU5cUdOBc, Conversations with Tyler, Bloomberg Opinion.
40K Followers 16 FollowingThe Machine Intelligence Research Institute exists to maximize the probability that the creation of smarter-than-human intelligence has a positive impact.
137K Followers 536 FollowingProfessor of Computer Science. AI Safety & Security Researcher. AI Influencer. My opinions are now yours! For talks/interviews: [email protected]
9K Followers 651 FollowingPhilosopher, writer, diver.
Author of 'Other Minds,' 'Metazoa,' and 'Living on Earth.'
HPS, University of Sydney. Views my own.
5K Followers 291 FollowingLet's make AI doctors!
Views my own;
CEO @ https://t.co/wvoKT50fKX;
AI Researcher @ Berkeley;
If I block you it's like I'm moving to another convo at a party; nbd.
11K Followers 462 FollowingHelping the world prepare for extremely powerful AI @open_phil (views my own), writer and editor of Planned Obsolescence newsletter.
30K Followers 123 FollowingMechanistic Interpretability lead DeepMind. Formerly @AnthropicAI, independent. In this to reduce AI X-risk. Neural networks can be understood, let's go do it!
32K Followers 279 FollowingBlogger primarily on AI and AI x-risk but also other things at Don't Worry About the Vase (SS/WP/LW), founding Balsa Research to fix policy.
50K Followers 3K FollowingAI alignment + LLMs at Anthropic. On leave from NYU. Views not employers'. No relation to @s8mb. I think you should join @givingwhatwecan.
32K Followers 3K FollowingFocused on machine learning & society. Previously @Salesforce Research via @MetaMindIO. @Harvard '14, @Sydney_Uni '11. 🇦🇺 in SF.
Also @ https://t.co/IamBQbjRiN
109K Followers 166 FollowingUPMC Professor of Computer Science @ CMU, President Elect ICML Board, VP of Research @ Meta (Multimodal LLMs, AI Agents), ex-Director of AI research at @Apple
101K Followers 175 FollowingProfessor of computer science at UW and author of '2040' and 'The Master Algorithm'. Into machine learning, AI, and anything that makes me curious.
45K Followers 64 FollowingStudent of mind and nature, libertarian, chess player, cancer survivor. @ Keen, UAlberta, Amii, https://t.co/u8za2Kod54, The Royal Society, Turing Award
196K Followers 787 FollowingGerman Physicist. Author of "Lost in Math" & "Existential Physics".
There is no strength in numbers, have no such misconception.