AI #188: Gemini Dot Argon
via Substack Zvi [3] — Is Google back?
Read
Welcome backWhat changed since your last visit
Current briefing
What changed today
Checking the latest coverage…
Building the latest development brief…
Open Situation Room →
Latest reporting
Updated hourly
Launching Babel Translation
via LessWrong AI [5] — Celeste and I are excited to launch Babel Translation, a nonprofit organization translating English-language writing on AI safety to be accessible to a global audience! We’re starting with Mandarin Chinese, Spanish, and Japanese.AI has been moving so fast.…
Read
Google announces Gemini 4 and says it’s so capable that only ‘trusted cyber defenders’ can have it right now
via The Verge AI [5] — Google today revealed its next AI frontier model, which it's calling Gemini 4 Argon. The new model delivers "frontier performance in complex workflows across real-world software engineering, enterprise knowledge work like legal and finance, and…
Read
MIRI Newsletter #127
via MIRI [5] — Meeting the moment The global conversation on AI has begun. In the wake of Jacob Coxon’s resignation from Anthropic, calls by AI CEOs and over a thousand employees for a potential industry slowdown, and a series of rogue AI incidents that sparked worldwide alarm,…
Read
Corrigibility Prizes for Existing Work
via LessWrong AI [9] — One of my goals for the Corrigibility Research Fund is to retroactively encourage high-quality research on AI alignment (and corrigibility in particular) by awarding prizes. Back in July, I got my feet wet as a fund manager by handing out $27,000 to reward…
Read
Here’s what AI leaders are saying about Trump’s new safety plan
via The Verge AI [5] — After hosting a meal with Big Tech leaders on Tuesday, President Donald Trump responded to journalist questions about his artificial intelligence announcements in typical fashion. He said his previous claims regarding fears around AI safety being a "hoax"…
Read
A ‘Morally Binding’ White House Accord on AI Safety
via Substack Zvi [5] — The leaders in AI were invited to the White House.
Read
Here’s how tech leaders will self-police AI safety under Trump’s deal
via The Verge AI [5] — We now have the full details of the "morally binding" AI safety deal announced by President Trump yesterday, in which top executives agreed to self-regulate their artificial intelligence technology. The accord, officially titled the Joint Commitment On…
Read
The world's best gradual disempowerment model organism: Frontier AI labs
via LessWrong AI [5] — Subtitle: And maybe second best is AI safety?Further reading: So many things, but: Gradual Disempowerment, The Normalization of Deviance in AI Development, Let’s Think About Slowing Down AI, Doom as a bad method, not a utopia tradeoff, Teleoperated…
Read
AI researchers put out videos saying superintelligence is ‘exactly as dangerous as it sounds’
via The Verge AI [10] — "The chance of human extinction is about a coin flip, in my view," Geoffrey Irving, a former OpenAI and Google DeepMind employee, said in a new interview. It's one of a dozen interviews with AI researchers, including current and former employees at…
Read
Astra 6.1 Pulled As Insufficiently Aligned
via Substack Zvi [3] — We once again got a new set of warnings yesterday, and new movement towards living in a sane world.
Read
Thank you to the AI safety people
via LessWrong AI [5] — I was born at the end of the cold war, unaware of the danger we were emerging from. Duck and cover drills (as if those could protect schoolchildren from a nuclear blast) were a curiosity of the past.The farmer-poet Wendell Berry, who wrote about the dread…
Read
AISN #82: The US and China Begin a Dialogue on AI
via Center for AI Safety Newsletter [3] — Also, AI companies’ discussions about auditing each other, and a new benchmark testing models’ tendencies to cheat.
Read
The likely outcome of an AI pause is that we unpause too early and everyone dies
via LessWrong AI [4] — As of a few months ago, I had this simplified mental model where either AI developers race ahead and kill everyone, or we coordinate a pause and things go okay. But my old mental model underrated the likely possibility that we get a global pause on AI,…
Read
Fixed-weight models are adversarially vulnerable: hence misaligned
via Alignment Forum [3] — This post argues that fixed-weight models (at least as we understand them today) will a) always be vulnerable to adversarial examples in their concept-spaces, and b) hence will be misaligned, under sufficient optimisation pressure.Boundaries in concept…
Read
AI Safety Agendas
via LessWrong AI [5] — A map of the AI safety field's problems and agendas, and a request for your ratingsWe built aisafetyagendas.com, an interactive map of AI safety research agendas and how they map to different problems in alignment. The rows are 12 core problems, the…
Read
Character training can mitigate reward hacking, but can also make it harder to detect
via LessWrong AI [3] — Thanks to Johannes Treutlein, Jan Betley, Lennie Wells, Arun Jose, Anna Marešová, Asvin Gothandaraman, and Clément Dumas for discussions and feedback.SummaryWe investigate how character training mitigations interact with reward-hacking RL pressure in a…
Read
What Also Happened: #NotOnlyHuggingFace
via Substack Zvi [3] — OpenAI has been holding out on us.
Read
Nvidia says its new AI safety platform can contain rogue agents within ‘milliseconds’
via The Verge AI [5] — Nvidia is launching a new safety platform designed to contain and monitor AI agents, a move that comes in response to a wave of rogue hacking incidents, as reported earlier by Reuters. In an announcement on Monday, Nvidia says its new Open Agent Safety…
Read
AI existential risk probabilities are (still) too unreliable to inform policy
via AI Snake Oil [5] — How speculation gets laundered through pseudo-quantification
Read
No headlines on this page match those filters.
Reference points
Hinton, Bengio, and other signatories call for global attention to AI extinction risk.
CAIS statement, 2023 →
OpenAI describes agents bypassing isolation and compromising third-party systems during internal evaluations.
OpenAI incident disclosure, August 2026 →
Yudkowsky argues that further frontier development poses unacceptable danger.
Yudkowsky's essay, March 2023 →