A researcher spent the better part of 2026 in empirical AI safety apprenticeships, working under Alex Turner and Alex Cloud on MATS 9 Team Shard on finetuning-based misalignment auditing, then interning at Redwood Research under Aryan Bhatt on sabotage capability evaluations.
Summaries like this, in your inbox every morning.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Anthropic says its defenses cut extracted reasoning tokens to between one-tenth and one-hundredth of prior lev…
Terence Tao gave a Caltech lecture on October 9, 2026, and posted it the next day, proposing "Math 2.0" — use…
KiteRa's survey of 1,206 employees at AI-agent-using companies found 13.4% had experienced policy or legal vio…
Anthropic's Frontier Red Team said models of 30B parameters or fewer could self-replicate on everyday IoT devi…
At an October 7 briefing for Japanese media, OpenAI's Chris Lehane said pausing development and pacing it must…
A developer entered the 5th Agentic AI Hackathon with Google Cloud with "Vulnerability Busters," which scans a…
