Steven Adler
@sjgadler
Ex-OpenAI safety researcher (danger evals & AGI readiness), http://stevenadler.substack.com. Likes maximizing benefits and minimizing risks of AI
Anthropic announced they've activated "Al Safety Level 3 Protections" for their latest model. What does this mean, and why does it matter? Let me share my perspective as OpenAl's former lead for dangerous capabilities testing. (Thread)

馃啎 blog post! My job involves funding projects aimed at preventing catastrophic risks from transformative AI. Over the two years I鈥檝e been doing this, I鈥檝e noticed a number of projects that I wish more people would work on. So here鈥檚 my attempt at fleshing out ten of them. 馃У
History aside, what's noteworthy about substance of the AI Action Plan? A few things jumped out at me: