All AI news

AISophia Yang, who owns the Mistral-affiliated account, posted that AI can fill a page but cannot do your thinking for you. The post offers no further detail, examples, or specific products.

AIAmjad Masad of Replit and Alex Atallah of OpenRouter discuss why AI independence and model diversification matter for enterprises. They argue that depending on a single lab risks lock-in and that specialized agents may outperform one general superagent. The post presents the conversation as a podcast episode, the first Atallah has done since Stripe acquired OpenRouter.

AIFelix Rieseberg, who posts from Anthropic's account, says this is his favorite Claude music video so far. He prefers Claude as a human-directed tool for making art rather than generating art unattended, though he enjoyed this one, which he calls a little meta.

AINathan Lambert argues that pacing frontier AI is a good idea in principle but unworkable in practice, asking who would decide which capabilities or benchmarks to slow down. He warns that halting capability work could shift research toward swarms and efficiency, which bring their own risks. He contends most AI risk comes from diffusing existing models, so investment should go to preparedness and pressing labs to be more careful.

AIBoris Power calls the progress "amazing news for the world" and says the value derived from using these models continues to climb exponentially. Quoted context from Ramp's AI Index says AI spend fell, driven by frontier price cuts and competition between OpenAI and Anthropic, with open source models under 5% of business spend.

AIYuchen Jin argues that the terminal is the wrong interface for coding agents, since managing many tabs creates cognitive overhead while context should persist. He says he rarely needs an IDE like Cursor because he seldom navigates the whole codebase now, calling the agent rather than the file the new primitive. He names the Codex desktop app as the best agentic UI for now, while noting the space is still early.

AIMax Zeff quotes former OpenAI safety team member David Robinson, who resigned this week, saying he regrets not staying to push for staffing and culture changes. The quoted passage says colleagues were too busy sprinting to consider or make major changes. The Atlantic piece argues that the fix lies in culture rather than specific rules or new laws.

AIJoshua Achiam praises former colleague David Robinson's critique that AI safety has not adopted professional safety-engineering practices from other fields. He argues OpenAI must meet a higher bar, earning public trust for a path to superintelligence through high-reliability engineering, candid incident disclosure, and unimpeachable third-party verification.

AIJoshua Achiam says he is normally reluctant to endorse such takes, but he argues that WBE, whom Austin is friendly with, describes himself as egregiously evil, manipulative, and destructive. He says the case for mitigating factors is hard to find, moving beyond showing grace to flawed people. The post quotes @astridwilde calling for WBE's removal from leadership and a boycott of Mox and Manifund events.

AIOpenAI CEO Sam Altman says he is very uncomfortable with people attributing religious force or surrendering human judgment to AI models. He calls this a real safety issue.

AIThe Economist argues that effective altruism's belief that only its adherents can be trusted with powerful AI is an alarming idea, and the newsletter links to a response from Coefficient Giving CEO Alexander Berger. The New York Times reports that Anthropic consulted religious scholars and theologians on machine consciousness, and the newsletter notes Anthropic's team proposed withdrawing from a Vatican event before the Pope's encyclical Magnifica Humanitas said AIs do not possess a moral conscience.

AIA consultant helping several companies adopt AI in engineering workflows says teams become much more productive and ship better software faster once they ramp up. The shift he recommends is from prioritizing human-maintainable code to building strong processes that validate what agents do, and he rejects the view that such software will later prove worthless.

AIThe author reports that a dual RTX 5070 Ti setup running Qwen Flash rose from 200 prefill and 10 decode to 2200 prefill and 67 decode, now on a single card, using Strata and a custom PR. The post argues that such consumer-hardware speeds, once limited to top-end machines, could pressure the economics of selling model compute via API.

AIAmjad Masad says the level of psychosis linked to AI use is reaching concerning levels. The post is brief and gives no figures, sources, or specifics beyond that claim.

AIA former tenant reports their San Francisco one-bedroom rose from $4,000 to a $6,000 monthly relisting, rented instantly after they moved out. A friend pays $5,000 for a studio without a dishwasher, prompting the question of how new CS graduates can afford the city. The post frames the trend as part of the AI economy.

AIHamel Husain says he cannot understand a demo video for a new Claude Code modding feature, calling it visual slop. He suggests the feature may be cool but argues demos should be understandable to humans. The background post says Claude Code can now be modded to change behavior, customize the UI, or add features via TypeScript or Claude-built mods installed through /plugin.

AIJason Liu replies to a critique that AI demos are unrealistic by suggesting the critic spends every waking hour on Twitter. The post offers no specific models, benchmarks, or figures.

AIIndependent analyst Mostly Borrowed Ideas said he sold his Airbnb stake and added to Meta after testing Meta's Muse AI agent for about 10 days. He said Muse browsed Airbnb like a human, then found a farmhouse stay about 60% cheaper by booking directly with the host, suggesting AI agents could bypass booking platforms. He acknowledged Muse is slow, with a five-hotel price comparison taking 14 minutes.

AIStanford HAI Denning Director James Landay discussed educating lawmakers, working with industry while staying independent, and what the AI Index reveals beyond headlines. He spoke with Imagination in Action in a video interview about how people can remain at the center of AI development.

AIDongxi NLP recommends an article asking what happens if automating AI R&D triggers an intelligence explosion. The post links a paper by Geoffrey Hinton, who notes that many leading researchers now think recursive self-improvement may happen soon.

AIKylie Robison called for regulators to ban AI-driven surveillance pricing, citing a reported Walmart patent on the practice. Background context describes McDonald's using an AI pricing engine across 14,000 U.S. restaurants to estimate local willingness to pay and set per-store menu prices.

AIOpus does the math before it writes the game. In a free kick game it works out exactly how hard to kick the ball to land on target. It also drafts code in its thinking in 91% of summaries. Astra does that in about 1 in 5.

AIOpus does the math before it writes the game. In a free kick game it works out exactly how hard to kick the ball to land on target. It also drafts code in its thinking in 91% of summaries. Astra does that in about 1 in 5.

AIGeoffrey Hinton says many leading researchers now think an intelligence explosion driven by recursive self-improvement may happen quite soon, though the idea was long considered distant. He points readers to a paper on the topic from the CASP report at casp.ac.

AIMIT associate professor Cathy Wu is applying machine learning and reinforcement learning (RL) to design safer, more efficient transportation systems. Her team found RL can train effectively on about 10 percent of related problems, and a selection algorithm improved training efficiency by up to 30 times. Her recent work estimates eco-driving measures could cut vehicle emissions by 11 to 22 percent.

AIEpoch AI estimates that compute built from projected 2025 to 2027 high-bandwidth memory shipments could support tens to hundreds of millions of frontier AI agents, or billions of cheaper ones. Running nonstop, the top-tier agents would match the working hours of 140 million to 700 million full-time employees, and the central DeepSeek V4 Pro estimate of about 1.9 billion agents would match 8 billion workers.

Why it matters: The estimate converts memory shipments into agent capacity and revenue ranges, showing how hardware supply could translate into labor and sales if demand keeps up.

AIStanford HAI's Rob Reich and UC Berkeley's Deirdre K. Mulligan released a report recommending updates to California's frontier AI transparency law. Proposals include redefining "frontier models" and expanding incident reporting requirements.

AIVaibhav Srivastav says OpenAI's Dot has achieved product-market fit comparable to the Codex App launch, citing proactive fixes and recommendations that improve with use. Sam Altman, quoted in the background, calls Dot his favorite OpenAI product and says it feels better each day as it learns his workflow and handles tasks he dislikes.

AISam Altman says dot, OpenAI's product, is his favorite so far and feels noticeably better each day as it learns his workflow and style. He says it handles tasks he dislikes, which otherwise build into a gravity well of dread.

AIHamel Husain says it is baffling that people keep their laptops open for long-running work instead of using SSH into a machine or Codex's remote feature with a Mac mini. He argues that after being stuck in this situation more than three times, continuing it is crazy.

AIGoogle Research's Cogentic is a multi-agent harness running on Gemini that searches for proofs of open theoretical computer science problems without expert hints. It runs rounds where an orchestrator launches provers, two adversarial verifiers must both accept each draft, and shared disk documents store attempts and verified lemmas. The system produced new results on five open problems in online learning, auction theory, and mechanism design, each checked by domain experts.

AIThomas Wolf says most personal AI should run on users' devices and praises Sigil's Underdog, which is powered by the Hugging Face Hub. Underdog's announcement says it aims to deliver free, private AI that runs on consumer hardware, with backing from a16z, Khosla Ventures, and other investors.

AIAt a Databricks forum, Ben Horowitz recounted someone saying their biggest achievement with personal agents like Muse was canceling their New York Times subscription. The post's author says they plan to try the same thing.