All AI news

AIJoshua Keating argues that keeping a human in the loop on lethal AI decisions is not enough if the humans rely too heavily on AI outputs. He cites a CNN-reported case in which an analyst's AI-assisted report falsely identified a Chinese ship's cargo as nuclear components, nearly prompting a boarding during the Iran war. The piece links this to automation bias and to military AI cases in Gaza and Minab, and warns that AI integration early in a nuclear decision chain is harder to regulate than autonomous launch.

AIAI is changing developer work, and the article recommends three skills: directing AI agents, reviewing AI output instead of trusting the first answer, and using saved time for judgment-heavy problems such as customer needs and tradeoffs. It cites GitHub Copilot's built-in Rubber Duck agent, which uses a second model to critique plans, code, and tests. The author argues that developers remain responsible for outcomes while AI handles more implementation.

AIVenture investors have funded defense-tech companies such as SpaceX, Anduril, and Castelion, but the article argues that production capacity in the supplier base is now the bottleneck. Most of America's machine shops and manufacturers are small, with 83% of machine shops employing fewer than 20 people, and 61% of tier-two-and-below defense manufacturers cite tooling, automation, or production-line limits as top expansion barriers.

AIThomas Wolf joked that the first good AI movie will feature Ben Affleck explaining learning rates while Matt Damon plays the base model. The quoted post adds a mock Affleck line about freezing base weights and using a 2e-4 learning rate.

AIMerve Noyan notes that zero-shot classification models are now being rebranded as decision models. She says she did not expect this shift in 2026 but is glad people have finally discovered these models.

AIHugging Face's Merve Noyan said she thinks she and PewDiePie would get along well and that he would love being on their Slack. The post is a playful reply to Ben Burtenshaw's suggestion that she could teach PewDiePie about local agents and post-training.

AIThore Graepel, a core member of the AlphaGo team, argues that current large language models do not truly reason, despite chain-of-thought gains in math and coding. He says they lack an explicit, inspectable epistemic state, keep knowledge and reasoning intertwined in their weights, and often produce post-hoc explanations. He proposes systems that maintain an auditable epistemic state and evaluate each step by how much it resolves uncertainty.

AIDaniel Kokotajlo, who leads the AI Futures Project, testified before a Senate subcommittee on September 30, 2026. He argued that Anthropic and OpenAI are racing toward superintelligence by automating AI research and development, and that his team thinks this could happen as early as 2028. He warned that declining monitorability and models that appear aligned during evaluations make misalignment harder to detect, and he recommended greater industry transparency and redirecting compute away from AI R&D.

AILucas Beyer calls SWE-sweep a useful new benchmark, where agents must find and fix bugs in a repo checked out at an earlier commit, scored against unit tests from real later bugfixes. He notes two limitations: a model may find valid bugs that don't match the tested ones, and the construction makes training on the test set easy. He advises not overemphasizing small ranking differences once models score highly.

AIWill Depue, who is affiliated with OpenAI, says his tweet is not being misinterpreted and that he has heard the same claim from Anthropic employees about OpenAI. He argues that a later interpretation shifting the framing to US versus China is weak and only marginally better.

AIThe author recalls a senior engineer who dismissed a newcomer's question with "oops, forgot," and says LLMs now answer patiently through text, diagrams, videos, and more. The post frames this shift as making dismissive gatekeeping obsolete, building on Andrej Karpathy's tips for turning LLM outputs into easier-to-read formats such as ASD-STE100 writing, diagrams, HTML pages, and generated explainer videos.

AIJason Liu says people claimed the same about coding, suggesting AI could change travel booking similarly. He reacts to a post contrasting booking a flight and hotel in two minutes alone with a ten-minute call where an assistant reviews flight options and sends map screenshots.

AIA post from OpenAI-affiliated account @willdepue shares a quote attributed to an Anthropic policy lead, calling its wording dangerous. The quoted post reads "You can't do safety from second place," which the main post presents as the concerning statement.

AIVaibhav Srivastav, an OpenAI-affiliated account, posted that people overestimate what they can achieve in one year but underestimate what they can accomplish over a decade. A quoted post recalled how ChatGPT looked in 2022, four years earlier, to frame that point.

AITinker says Fulcrum trains its Echo writing model by building on a base model that already writes well, tailoring both SFT and RL to separate the default LLM voice from authors' voices. The post calls this customization approach both cheap and effective. Fulcrum says Echo beats frontier models at writing tasks such as fiction and technical explanations, at a training cost under $5K.

AIGuillermo Rauch praised the release of SvelteKit 3 and said he is excited for Async Svelte and Remote Functions. He built a small app that built and deployed in 15 seconds end-to-end, using Fx and Opus to handle the work.

AIIdeogram praised photographers for their craft and said it hopes to help them save hours of work. The post contains no concrete product features, model names, or figures, and it does not specify what kind of editing help is offered.

AIYuchen Jin says no AI today can generate an Andrej Karpathy video from a prompt, and he hopes Karpathy will return, noting he has not uploaded a YouTube video in over a year. The post builds on Karpathy's suggestion that LLMs can produce bespoke explainer videos on arbitrary topics, though that idea is still emerging.

AIBen Tossell argues that personal agents should proactively create apps and websites for their users. The post makes this claim without offering specific products, features, or benchmarks.

AIKylie Robison tried to book an international flight last night using Instinct, an agentic booking tool, but ended up neither completing the booking nor avoiding the search herself. She had been inclined to defend agentic flight booking as a way to offload tasks, but her experience left her reconsidering that position.

AIJournalist Kylie Robison jokingly argues that it is her duty as a journalist to make this about her, linking to a post on X. The post is brief and offers no substantive news, and the quoted reply from @dwr only comments that "dot" is a useful name for a personal agent if it launches a dot-shaped hardware device.

AIHarrison Chase argues that every agent harness requires a durable runtime, citing pi-durable as an example alongside deepagents built on LangGraph. The post frames durable execution as a basic requirement for agent systems rather than an optional feature. Pi 1.0 shipped with Pi Durable, which the referenced @pidotdev post invites users to customize.

AIRollouts account for most of RL's compute cost, and splitting them from training across separate engines can introduce numerical mismatches. In MoE models, such mismatches can even route tokens to different experts. Fireworks says it co-builds both engines so training stays fast and consistent.

AIBoris Power, who is listed as associated with OpenAI, posted only that he is excited for the next transportation revolution, without naming a product, company, or figures. The post links to Bryan Caplan's piece on rideshare economics, which estimates an equilibrium price of about $2 an hour for autonomous vehicles even when they sit empty half the time.

AIThe author jokes about founding a consulting firm that would use agents for work, Griffin for meetings, and Griffin for interviews to fill remote roles. They then question whether remote engineers and consultancies would still be needed if that became reality. The quoted Tavus post says Griffin passed a video Turing test with 48% of live interlocutors believing it was human.

AICline says two of its largest tasks in the last 30 days each processed 9B tokens, costing about $8,500 on Claude Opus 5 versus about $300 on DeepSeek V4 Pro. The post says that is roughly 30x cheaper for the same token count, letting users run large-horizon work without spending thousands or waiting on limit resets.

AIarXiv has introduced stricter rate limiting for all submitters to fairly distribute moderator time. The post links this move to Vibe research, where turning ideas into papers is easier, while noting that standards for judging research output have not kept pace.

AIHarvard physicist Matthew Schwartz argues that LLMs are poorly matched to science when used as human-style collaborators, so he built a toolkit for exact quantitative calculations. Working with Claude, the approach surfaced connections to ecology, population genetics, and a dozen other fields, with domain experts steering it toward interesting questions.

AIFrançois Chollet argues the key difference between base LLMs and modern LRMs is a shift from transductive answer prediction to inductive prediction of the program or reasoning chain behind an answer. He says this enables test-time induction and substantial fluid intelligence in LRMs, which he claims base LLMs largely lack. He cites ARC 1 results: base LLMs remain around 10-15%, while LRMs of the same size or smaller saturated the benchmark in 2025.

AIEmber-1, built on the Kimi K3 base model, is reported by @kickingkeys to cost about 20% less and use about 36% fewer reasoning tokens across 124 test prompts. The same tester says it runs about 2x faster, while the source's painting examples show its own expressive style.