- MATS 9.0 applications are open! Launch your career in AI alignment, governance, and security with our 12-week research program. MATS provides field-leading research mentorship, funding, Berkeley & London offices, housing, and talks/workshops with AI experts.
- Summer applications just launched! Mentors include AI safety researchers from @AnthropicAI, @OpenAI, @deepmind, @MIRIBerkeley, @CHAI_Berkeley, @cais, @FHIOxford, and more! serimats.org
- The world is sleepingSurvey of 900 experts by World Economic Forum reveals they have bizarre views about the biggest global risks. Most severe 10 year risk is extreme weather events??
- 80% of MATS alumni who completed the program before 2025 are still working on AI safety today, based on a survey of all available alumni LinkedIns or personal websites (242/292 ~ 83%). 10% are working on AI capabilities, but only ~6 on pre-training at a frontier AI company.
- In 3.5 years, @MATSprogram has helped produce 115 arXiv publications. Our organizational h-index is 31.
- @MATSprogram Summer 2025 applications close Apr 18! Come help advance the fields of AI alignment, security, and governance with mentors including @NeelNanda5 @EthanJPerez @OwainEvans_UK @EvanHub @bshlgrs @dawnsongtweets @DavidSKrueger @RichardMCNgo and more!
- OpenAI' latest model, o3, scored: - 2727 on Codeforces, making it the 175th best competitive programmer on Earth - 25% on FrontierMath, where "each problem demands hours of work from expert mathematicians" - 88% on GPQA, where 70% represents PhD-level science knowledge - 88% on
- MATS has received mentorship applications from 155 researchers for our Winter 2025 program, far more than we can support. If you run an AI safety or governance program and you want referrals, let me know!
- What a week! Both Anthropic and DeepMind shed some light on their AI alignment plans after OpenAI shared their plan in Aug 2022. * Anthropic: anthropic.com/index/core-vie… * DeepMind: lesswrong.com/.../4iEpGXb...… * OpenAI: openai.com/blog/our-appro…
- The Paris AI Summit was a staggering failure. Entropy - 1, Humanity - 0
- First AI alignment paper to win ICML Best Paper! So happy to have helped support this work at @MATSprogram :) Well done, @McHughes288, @danvalentine256, @sleight_henry, @akbirkhan, @EthanJPerez, @sleepinyourhat and coauthors!excited to announce this received an “ICML Best Paper Award”! come see our talk at 10:30 tomorrow
- Another @MATSprogram concluded and another 61 scholars graduated on Fri! Our Scholar Symposium featured 46 talks on AI interpretability, model evals + demos, agent foundations, control/red-teaming, scalable oversight, and more.
