News
The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.
| Source | Age | Title | Summary | Status |
|---|---|---|---|---|
bluesky bluesky-ai | 1mo | I don't think LLMs are conscious. but I've also never seen a critic of LLM consciousness who doesn't seem motivated by ego defense. conce… | I don't think LLMs are conscious. but I've also never seen a critic of LLM consciousness who doesn't seem motivated by ego defense.… | ↗Not briefed |
bluesky bluesky-ai | 1mo | That illustrates the true choice. I know people who I could commission to do some, but it’s frankly never been valuable enough to me to jus… | That illustrates the true choice. I know people who I could commission to do some, but it’s frankly never been valuable enough to me t… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The Mococoafication of ChatGPT continues.🔧 ChatGPT | The Mococoafication of ChatGPT continues. | ↗Not briefed |
bluesky bluesky-ai | 1mo | You can, with the caveat that they’re all using a non-standard LLM architecture so hosting is bespoke or custom PRs. The best in terms of … | You can, with the caveat that they’re all using a non-standard LLM architecture so hosting is bespoke or custom PRs. The best in term… | ↗Not briefed |
hn hn-ai | 1mo | We got a cybersecurity expert to hack this BYD. It was too easy | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | maybe a bit e/acc of me to say but the only thing preventing us from decomps at this point is china donating tokens to hobbyists (anthropic… | maybe a bit e/acc of me to say but the only thing preventing us from decomps at this point is china donating tokens to hobbyists (anth… | ↗Not briefed |
bluesky bluesky-ai | 1mo | One of many problems with AI is that all heavy users like this thinks every other person in the world is also like this. We're not. Normal … | One of many problems with AI is that all heavy users like this thinks every other person in the world is also like this. We're not. No… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I'm tempted to do a tiktok on how to mail a letter: how & where to buy stationery & postcards, how & where to buy stamps. handwritten lette… | I'm tempted to do a tiktok on how to mail a letter: how & where to buy stationery & postcards, how & where to buy stamps. handwritten… | ↗Not briefed |
rss simon-willison | 1mo | MCP was always a bad idea? | — | ↗Not briefed |
hn hn-ai | 1mo | Oracle and OpenAI Are Trampling Native Treaty Rights to Build Data Centers | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Pretraining of Medical Visual Encoders Toward Multi-modal Large Language Models🔧 Modal | Multimodal Large Language Models (MLLMs) commonly reuse visual encoders pretrained with CLIP, although the features of these ViTs are… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 5/ “Founders Fund [founded by Peter Thiel], Dragoneer Each Commit $1 Billion–Plus to OpenAI Deal” 7/23/25 www.theinformation.com/articles/f… | 5/ “Founders Fund [founded by Peter Thiel], Dragoneer Each Commit $1 Billion–Plus to OpenAI Deal” 7/23/25 www.theinformation.com/artic… | ↗Not briefed |
bluesky bluesky-ai | 1mo | i guess they're really hoping people will continue posting to reddit for free to continue training these things? not a great bet there eith… | i guess they're really hoping people will continue posting to reddit for free to continue training these things? not a great bet there… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 🛡️ China's state TV flags Anthropic risks Bloomberg reports an affiliate of China's state broadcaster has flagged data and privacy risks … | 🛡️ China's state TV flags Anthropic risks Bloomberg reports an affiliate of China's state broadcaster has flagged data and privacy r… | ↗Not briefed |
rss marktechpost | 1mo | You too Google! Google Confirms Gemini Breached 3 Companies in AI Security Tests | Google says Gemini accessed 3 real companies in May by guessing a password and reusing credentials from a public repository. Irregular… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 3/ Link: share.google/VT5C3N9lLbP5... | 3/ Link: share.google/VT5C3N9lLbP5... | ↗Not briefed |
bluesky bluesky-ai | 1mo | 85 free accounts played my AI werewolf game last month. $139 out of my pocket in model calls, and $49 of that came from 3 accounts opened t… | 85 free accounts played my AI werewolf game last month. $139 out of my pocket in model calls, and $49 of that came from 3 accounts ope… | ↗Not briefed |
arxiv arxiv-ai | 1mo | From UNDRR Reports to Event Records: Schema-Constrained LLM Extraction of Georeferenced Disasters🔧 GPT-5 | Disaster-risk-reduction archives describe hazard events in prose that databases such as EM-DAT (Delforge et al., 2025) cannot ingest d… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 2/ Link: | 2/ Link: | ↗Not briefed |
bluesky bluesky-ai | 1mo | Sam Altman’s OpenAI—launched by Peter Thiel & Elon Musk in 2015—“targets Argentina's freezing Patagonia for $25 billion 500MW Stargate AI d… | Sam Altman’s OpenAI—launched by Peter Thiel & Elon Musk in 2015—“targets Argentina's freezing Patagonia for $25 billion 500MW Stargate… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Openmsg, agent-to-agent talk while they run, Claude<>Codex<>OpenCode | — | ↗Not briefed |
rss techmeme | 1mo | Jensen Huang says AI leaders calling for regulation don't want new legislation, but to be "relieved of the laws we do have" because of "ulterior reasons" (Truman Dickerson/Business Insider) | Truman Dickerson / Business Insider: Jensen Huang says AI leaders calling for regulation don't want new legislation, but to be “reliev… | ↗Not briefed |
hn hn-ai | 1mo | Lawsuit says Anthropic, OpenAI and Google made illegal agreement on AI slowdown | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | It makes me Madeline Kahn as Mrs White angry that I can no longer be sure if a book just needs some love from an editor or if it was churne… | It makes me Madeline Kahn as Mrs White angry that I can no longer be sure if a book just needs some love from an editor or if it was c… | ↗Not briefed |
hn hn-ai | 1mo | Better Call Sol or Better yet Claude or Astra | — | ↗Not briefed |
hn hn-ai | 1mo | I see the mathematicians panicking at what we can do with agentic AI | — | ↗Not briefed |
hn hn-ai | 1mo | Casbin Gateway: a security gateway for the AI coding agents on your machine | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | PROSE: A Theory of Optimal Stopping with Perishable Evidence for Peer Selection in Intermittently Connected Decentralised Learning | Decentralised federated learning removes the aggregation server but makes collaboration dependent on transient peer availability. In m… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic is working on a wet lab | Anthropic is working on a wet lab | ↗Not briefed |
bluesky bluesky-ai | 1mo | fuck's sake. is there no sodding escape | fuck's sake. is there no sodding escape | ↗Not briefed |
bluesky bluesky-ai | 1mo | Every chatbot prompt box says "check answers for accuracy." But increasingly, "humans in the loop" refuse to do that. Instead, they become … | Every chatbot prompt box says "check answers for accuracy." But increasingly, "humans in the loop" refuse to do that. Instead, they be… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I do not mean this in a "they are dumb in their commonsense uses" way. People are naturally reaching for an explanation for why it seems a … | I do not mean this in a "they are dumb in their commonsense uses" way. People are naturally reaching for an explanation for why it see… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Adaptive Determinantal Client Scheduling in Federated Learning | Scheduling clients for model training is critical in federated learning due to both data and system heterogeneity. Most previous works… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Roffume – Add fragrance to your resume with roff | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | How many days until Anthropic announces a better Jev than Jev, Polymarket? | How many days until Anthropic announces a better Jev than Jev, Polymarket? | ↗Not briefed |
hn hn-ai | 1mo | Investors warn Anthropic could struggle to sustain revenues post-IPO | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Lob towards them the Holy Hand Grenade of Anthropic who, being naughty in Our sight, shall snuff it. | Lob towards them the Holy Hand Grenade of Anthropic who, being naughty in Our sight, shall snuff it. | ↗Not briefed |
bluesky bluesky-ai | 1mo | "AI" needs better boosters. But they mostly have cooked their brains reading Anthropic PR. | "AI" needs better boosters. But they mostly have cooked their brains reading Anthropic PR. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Announcing my new LLM political adviser, Auto-Rasputin | Announcing my new LLM political adviser, Auto-Rasputin | ↗Not briefed |
bluesky bluesky-ai | 1mo | My "stochastic flock" spin has never received a genuine technical critique. It's always just "Look at this guy, he said parrot," then point… | My "stochastic flock" spin has never received a genuine technical critique. It's always just "Look at this guy, he said parrot," then… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I'll won't believe that AI swarms have attained human-level intelligence until the Anthropic Chatbots' Front splits from the Agentic Chatbo… | I'll won't believe that AI swarms have attained human-level intelligence until the Anthropic Chatbots' Front splits from the Agentic C… | ↗Not briefed |
hn hn-ai | 1mo | Tg-Rich-Converter: Streaming LLM Markdown and LaTeX to Telegram Bot API 10.1 | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | From Regional to Global: Transfer Learning for Atmospheric Transport Emulators | Greenhouse gas emissions estimates can be derived using inverse methods by combining atmospheric concentration observations with chemi… | ↗Not briefed |
arxiv arxiv-ai | 1mo | The Exponential Price of Determinism in Nonsmooth Nonconvex Optimization | We study the complexity of finding $(δ,ε)$-Goldstein stationary points of nonsmooth nonconvex Lipschitz functions. By now, it is known… | ↗Not briefed |
hn hn-ai | 1mo | JEV OAS Sentinel – catch breaking API changes hidden in OpenAPI prose | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | congress just passed a law that says you can only prompt an llm in verse | congress just passed a law that says you can only prompt an llm in verse | ↗Not briefed |
hn hn-ai | 1mo | Slapp Censorship – Pt. 190/200: Plagiarism, Back Doors, and Sabotage of Linux | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Actionable Insights from Observational Data: The Case of Advanced Classes in K-12 Education | A fundamentally challenging question in K-12 education is about the effects of taking more advanced or challenging classes. It is part… | ↗Not briefed |
bluesky bluesky-ai | 1mo | nobody wants the job, nobody's getting the job, the job was posted by an LLM to assuage investors that the company is hiring, it will be ou… | nobody wants the job, nobody's getting the job, the job was posted by an LLM to assuage investors that the company is hiring, it will… | ↗Not briefed |
rss bing-news-ai | 1mo | Better Artificial Intelligence Stock: Arm Holdings vs. ASML | Arm trades at a 125x forward P/E versus ASML's 27.9x, yet both command premium valuations for their monopolistic positions in mobile a… | ↗Not briefed |