via LessWrong AI [3] — TLDRThis is a post announcing a lot of new ARENA material I've been working on for a while, which is now available for study here (currently on the alignment-science branch, but planned to be merged into main this Sunday).There's a set of exercises (each one contains about 1-2 days of material) on t

via LessWrong AI [4] — OpenAI CEO Sam Altman wrote in a memo to staff that he will draw the same red lines that sparked a high-stakes fight between rival Anthropic and the Pentagon: no AI for mass surveillance or autonomous lethal weapons.Why it matters: If other leading firms like Google follow suit, this could massively

via ArXiv cs.AI [4] — Agile hardware design flows are a critically needed force multiplier to meet the exploding demand for compute. Recently, agentic generative AI systems have demonstrated significant advances in algorithm design, improving code efficiency, and enabling d

via ArXiv cs.AI [3] — Traditional software relies on contracts -- APIs, type systems, assertions -- to specify and enforce correct behavior. AI agents, by contrast, operate on prompts and natural language instructions with no formal behavioral specification. This gap is the

via Alignment Forum [5] — Authors: Aditya Singh*, Gerson Kroiz*, Senthooran Rajamanoharan, Neel NandaAditya and Gerson are co-first authors. This work was conducted during MATS 9.0 and was advised by Senthooran Rajamanoharan and Neel Nanda.MotivationImagine that a frontier lab’s coding agent has been caught putting a bug in

via Substack Zvi — Events continue to be fast and furious.

via ArXiv cs.AI [4] — Agentic reinforcement learning (ARL) has rapidly gained attention as a promising paradigm for training agents to solve complex, multi-step interactive tasks. Despite encouraging early results, ARL remains highly unstable, often leading to training coll

via Substack Zvi [2] — The situation in AI in 2026 is crazy.

via The Verge AI [4] — After less than two years at Amazon, David Luan, the head of Amazon's San Francisco AI lab, is departing the company. Luan announced the update in a post on LinkedIn on Tuesday, saying, "I'll be leaving Amazon at the end of this week to cook up something new." He added that, "There's incredible work

via LessWrong AI [4] — I have 3 Claude Code instances running on an otherwise empty server with a shared Manifold Markets account. They have an internal messaging system for async communication. Observations from running this agent collectiveu2026

via Substack Zvi [2] — Anthropic first gave us Claude Opus 4.6, then followed up with Claude Sonnet 4.6.

via Substack Zvi — A thought experiment about AI safety scenarios and their implications for alignment research.

via The Verge AI [2] — Anthropic's weekslong battle with the Department of Defense has played out over social media posts, admonishing public statements, and direct quotes from unnamed Pentagon officials to the news media. But the future of the $380 billion AI startup comes down to just three words: "any lawful use." The

via LessWrong AI — This post is detailing our experience attending the AI Impact Summit and its associated side events in Delhi, February 2026. We are both unfamiliar with the policy and governance domain. This is just an honest reaction attending these events, maybe there are 2nd order effects we…

via LessWrong AI — This post contains some rough reflections on the alignment community trying to make its ontology legible to the mainstream ML community, and the lessons we should take from that experience.Historically, it was difficult for the alignment community to engage with the ML community…

via ArXiv cs.AI — Accelerating exploration in visual reinforcement learning under sparse rewards remains challenging due to the substantial task-irrelevant variations. Despite advances in intrinsic exploration, many methods either assume access to low-dimensional states

via LessWrong AI [9] — [Whenever discussing when AGI will come, it bears repeating: If anyone builds AGI, everyone dies; no one knows when AGI will be made, whether soon or late; a bunch of people and orgs are trying to make it; and they should stop and be stopped.] Arguments for fast AGI progress…

via LessWrong AI [1] — L;DRWe describe the persona selection model (PSM): the idea that LLMs learn to simulate diverse characters during pre-training, and post-training elicits and refines a particular such Assistant persona. Interactions with an AI assistant are then well-

via LessWrong AI [4] — I think more people should be storing a substantial amount of food. It's not likely you'll need it, but as with reusable masks the cost is low enough I think it's usually worth it. It's hard for me to really imagine living through a famine. The world as I h

via ArXiv cs.AI [5] — Traditional AI alignment primarily focuses on individual model outputs; however, autonomous agents in long-horizon workflows require sustained reliability across entire interaction trajectories. We introduce APEMO (Affect-aware Peak-End Modulation for