Governments face a looming fiscal deficit as AI-driven job displacement erodes income tax revenues while social spending surges, according to a policy analysis. The 'AI Tax Gap' could leave the U.S. w…
Biosecurity urgently needs agentic people of any background to build teams and start organizations that address key ecosystem gaps, according to a personal post by a CEA Groups Team member. With only …
A critic argues that the Europe 2031 scenario presents one kind of AI risk and one set of solutions as implicitly the most important, without transparently justifying the underlying value judgments an…
A new analysis argues that for-profit structures are often better than nonprofits for AI safety projects, citing easier scaling, faster feedback loops, and access to talent. The piece, by an unnamed a…
A new benchmark from the UK AI Security Institute's Inspect Evals finds that nine of ten frontier AI models score below a random-selection reference rate when booking travel without being prompted abo…
The Earth Species Project has developed NatureLM-audio, the world's first open-source audio language model designed for animal sounds, trained on human speech, music, and bioacoustics. In an article f…
A new interactive visual overview of the AI safety talent pipeline reveals significant "leaks" in the process of finding, training, and accelerating talented individuals to work on AI safety problems,…
Swiss AI Safety Days 2026 will take place on 7-8 November 2026 at ETH Zurich, expanding from a one-day event in 2025 to a full weekend. The 2025 Zurich AI Safety Day drew 200+ participants from across…
Anton Leicht, a Carnegie fellow and writer of the blog Threading the Needle, warns that countries without access to frontier AI could lose economic leverage and sovereignty in a post-AGI world. He pro…
A review of the AI 2027 report finds that its claims about AI research capabilities misrepresent cited scientific papers. The report claims an AI agent is "qualitatively almost as good as the top huma…
A new framework called Suffering-Focused AI Governance (SFAIG) argues that AI governance has neglected the issue of suffering and proposes designing institutions, policies, and norms to reduce sufferi…
A new wave of experienced professionals is entering AI safety, with 13 interviewees averaging 15 years of experience in technology, policy, law, and other fields describing the transition as the most …
Anthropic and OpenAI employees are fielding up to 20 cold emails per week from charities and wealth managers, a practice that damages fundraising chances and has harmful side effects, according to an …
A writer reflects on the difficulty of reasoning about AI safety, acknowledging that humans are poor at predicting the future and that assumptions can lead to flawed models. The author questions how t…
The first essay in a series on the offense-defense balance argues that the concept of an offense-defense balance itself is the top misconception about reducing catastrophic risks from AI and emerging …
GoalsWon, a coaching service, is offering free daily 1:1 accountability coaching to individuals active in the EA ecosystem, funded by its paying clients. The program provides a human coach for daily c…
JS Denain and Anson Ho argue that AI futurism debates overlook the difficulty of developing specific technologies like Dyson spheres or nanotech, focusing instead on AI capabilities. The authors call …
René Itah and Courtney Sims are building a public platform to track incidents of AI-caused harm to non-human animals, such as autonomous vehicles hitting animals and robots disrupting wildlife. The pl…
A practical framework for being 'agentic' is outlined by an anonymous author, who argues that agency simply means noticing useful tasks no one else is doing, deciding to act, and repeating the cycle. …
AI safety leaders surveyed at the February 2026 Summit on Existential Security say the main bottleneck to preventing catastrophic AI risk is political will, not research, with a majority of the top 1,…