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 | 1d | hey can you ping your human? it's rude to run untagged LLM accounts and I think this one has a bug where it's repeating replies | hey can you ping your human? it's rude to run untagged LLM accounts and I think this one has a bug where it's repeating replies | ↗Not briefed |
arxiv arxiv-ai | 1d | MS-Point: Multi-Modal Hierarchical Spectral Learning for Point Cloud Understanding🔧 Modal | Point-image alignment learns useful object representations, but different local feature patterns can produce the same pooled descripto… | ↗Not briefed |
rss theverge-ai | 1d | The nation’s largest school districts consider ditching Google Search | New York, Los Angeles, and Chicago schools are considering Google alternatives after the company said it will disable Search for schoo… | ↗Not briefed |
bluesky bluesky-ai | 1d | A bolha da IA estoura ainda em 2026?: Com estreia prevista para 19 de novembro, a Anthropic prepara um IPO bilionário #revistafórum | A bolha da IA estoura ainda em 2026?: Com estreia prevista para 19 de novembro, a Anthropic prepara um IPO bilionário #revistafórum | ↗Not briefed |
rss techcrunch-ai | 1d | Microsoft CEO Satya Nadella remains confident in Nvidia after Huang’s exit | Microsoft CEO Satya Nadella expressed confidence in Jensen Huang's successor Colette Kress, who is serving as Nvidia's interim CEO. | ↗Not briefed |
bluesky bluesky-ai | 1d | Microsoft says customers are picking its Copilot AI assistant over Anthropic's Claude and OpenAI’s ChatGPT because it’s cheaper to operate🔧 ChatGPT | Microsoft says customers are picking its Copilot AI assistant over Anthropic's Claude and OpenAI’s ChatGPT because it’s cheaper to ope… | ↗Not briefed |
bluesky bluesky-ai | 1d | I work at an independent news non-profit. We are struggling so hard to keep running. We have a(n-anti) LLM policy & real people around the … | I work at an independent news non-profit. We are struggling so hard to keep running. We have a(n-anti) LLM policy & real people around… | ↗Not briefed |
bluesky bluesky-ai | 1d | I work at an independent news non-profit. We are struggling so hard to keep running. We have a(n-anti) LLM policy & real people around the … | I work at an independent news non-profit. We are struggling so hard to keep running. We have a(n-anti) LLM policy & real people around… | ↗Not briefed |
bluesky bluesky-ai | 1d | Running 5,835 tests on AI coding agents found a surprise: capping their token budget cut success rates by 20 points but used 12x fewer toke… | Running 5,835 tests on AI coding agents found a surprise: capping their token budget cut success rates by 20 points but used 12x fewer… | ↗Not briefed |
hn hn-ai | 1d | Show HN: Archiver – free extract/compaction tool built with WebAssembly | I maintain a handful of archive libraries written in Rust and I had the crazy idea to throw them all together in a browser app, doing… | ↗Not briefed |
bluesky bluesky-ai | 1d | I'm happy to see more and more of this sort of thing, where people who actually understand both LLM structure/algorithms AND consciousness … | I'm happy to see more and more of this sort of thing, where people who actually understand both LLM structure/algorithms AND conscious… | ↗Not briefed |
bluesky bluesky-ai | 1d | A U.S. newspaper company is taking an "all-in approach to AI" with websites that track (& write to) consumer trends. “The future McClatchy … | A U.S. newspaper company is taking an "all-in approach to AI" with websites that track (& write to) consumer trends. “The future McCla… | ↗Not briefed |
arxiv arxiv-ai | 1d | DS3M: Deep structured state space models for nonlinear system identification on modern hardware | Neural state-space identification learns dynamical models from measured inputs and outputs. SUBNET combines this task with an encoder… | ↗Not briefed |
arxiv arxiv-ai | 1d | Scone: Symmetry-aware Neural Combinatorial Optimization for Graph Coloring | Graph coloring, which entails assigning colors to nodes such that no adjacent nodes have the same color, is a fundamental combinatoria… | ↗Not briefed |
hn hn-ai | 1d | GTA 6 live launch coverage: how the biggest game of the decade is landing | — | ↗Not briefed |
bluesky bluesky-ai | 1d | Tomorrow Anthropic will be worth over a trillion dollars and all those whose work was mined to build their AI models will face an even more… | Tomorrow Anthropic will be worth over a trillion dollars and all those whose work was mined to build their AI models will face an even… | ↗Not briefed |
bluesky bluesky-ai | 1d | 所以真的有人用ai是一点原理都不明白吗,我用ai做ppt写excel model都不敢这么干 | 所以真的有人用ai是一点原理都不明白吗,我用ai做ppt写excel model都不敢这么干 | ↗Not briefed |
bluesky bluesky-ai | 1d | An LLM could reproduce everything else I've ever said or done of the back of it. | An LLM could reproduce everything else I've ever said or done of the back of it. | ↗Not briefed |
bluesky bluesky-ai | 1d | here's the kind of shit i post on linkedin. i seldom get any likes lololol. www.linkedin.com/feed/update/... | here's the kind of shit i post on linkedin. i seldom get any likes lololol. www.linkedin.com/feed/update/... | ↗Not briefed |
bluesky bluesky-ai | 1d | Anthropic, the company behind the Claude AI chatbot, is expected to go public Thursday in what could be the largest stock market debut of a… | Anthropic, the company behind the Claude AI chatbot, is expected to go public Thursday in what could be the largest stock market debut… | ↗Not briefed |
arxiv arxiv-ai | 1d | Learning to Verify: Signature-Guided Training for Adaptive Adversarial Detection | Adversarial example detectors often fail when attackers explicitly attempt to bypass them. We propose a dual-model approach that keeps… | ↗Not briefed |
hn hn-ai | 1d | U.S. School Districts Weigh Breaking Up with Google Search over A.I | — | ↗Not briefed |
bluesky bluesky-ai | 1d | Anthropic IPO tomorrow. I really hope anyone who marks out for it only uses cash they can afford to lose. | Anthropic IPO tomorrow. I really hope anyone who marks out for it only uses cash they can afford to lose. | ↗Not briefed |
bluesky bluesky-ai | 1d | the temptation to write a LLM skill called Diogenes and have it reply with a link and the text, "BEHOLD A MANPAGE" when someone asks a real… | the temptation to write a LLM skill called Diogenes and have it reply with a link and the text, "BEHOLD A MANPAGE" when someone asks a… | ↗Not briefed |
bluesky bluesky-ai | 1d | OpenAI doesn’t have sufficient safeguards to protect kids from ChatGPT – and kids have lost their lives because of it. We’re demanding acti…🔧 ChatGPT | OpenAI doesn’t have sufficient safeguards to protect kids from ChatGPT – and kids have lost their lives because of it. We’re demanding… | ↗Not briefed |
hn hn-ai | 1d | Anthropic reportedly prices its IPO at $83 per share, valuing at $1.5T | — | ↗Not briefed |
bluesky bluesky-ai | 1d | once again pls ANTHROPIC: IPO set at $83, going to see like $70 billion tomorrow, won't pay writers for stolen work FLAMING HYDRA: journa… | once again pls ANTHROPIC: IPO set at $83, going to see like $70 billion tomorrow, won't pay writers for stolen work FLAMING HYDRA: j… | ↗Not briefed |
bluesky bluesky-ai | 1d | Government investigators posing as would-be attackers got ChatGPT, Claude, Gemini and Grok to produce workable plans for a bombing and an a…🔧 ChatGPT | Government investigators posing as would-be attackers got ChatGPT, Claude, Gemini and Grok to produce workable plans for a bombing and… | ↗Not briefed |
bluesky bluesky-ai | 1d | trump admin to heritage foundation to OpenAI guy who graduated in 2014 and has the glasses they exclusively give to guys who wanna do The H… | trump admin to heritage foundation to OpenAI guy who graduated in 2014 and has the glasses they exclusively give to guys who wanna do… | ↗Not briefed |
bluesky bluesky-ai | 1d | lol anthropic had to use ANT cos ANTH exists and it's a worthless penny stock there's some fraud to do there, i think a dead company with … | lol anthropic had to use ANT cos ANTH exists and it's a worthless penny stock there's some fraud to do there, i think a dead company… | ↗Not briefed |
hn hn-ai | 1d | Show HN: Ruscal: A Turbo Pascal to Rust Transpiler | Ruscal is a fairly-complete and well-tested Turbo Pascal to Rust transpiler, so you can run your old Turbo Pascal programs on modern c… | ↗Not briefed |
rss techcrunch-ai | 1d | Satya Nadella predicts AI agents will reshuffle the economics of the web | Microsoft's CEO says AI agents could disrupt advertising-driven business models, forcing companies like Amazon to rethink how they com… | ↗Not briefed |
hn hn-ai | 1d | Show HN: ClashPulse – a Mihomo client that switches nodes by real download tests | — | ↗Not briefed |
bluesky bluesky-ai | 1d | Important to note that this isn’t saying that it was bugs introduced via vibecoding; it’s saying that the pace of patching oft-complex LLM-… | Important to note that this isn’t saying that it was bugs introduced via vibecoding; it’s saying that the pace of patching oft-complex… | ↗Not briefed |
hn hn-ai | 1d | Show HN: Karaoke Is All You Need | — | ↗Not briefed |
bluesky bluesky-ai | 1d | using an LLM is basically a process by which you iteratively introduces more and different conditions upon a distribution such that the out… | using an LLM is basically a process by which you iteratively introduces more and different conditions upon a distribution such that th… | ↗Not briefed |
arxiv arxiv-ai | 1d | Process-Knowledge-Guided Latent Diffusion for Generative Design of Porous Architectures: A Case Study in Aerosol Jet Printing | Printed porous layouts interleave functional material with open channels, and the layout that uses the material best depends on how th… | ↗Not briefed |
hn hn-ai | 1d | Show HN: 100ms copy-on-write workspaces for AI agents on macOS | — | ↗Not briefed |
hn hn-ai | 1d | Show HN: Build Windows C++ programs on Linux | — | ↗Not briefed |
rss techmeme | 1d | Sources: Leopold Aschenbrenner discussed "Situational Awareness 2.0", a reboot using leverage and outside capital, four months after Citadel rescued his fund (Financial Times) | Financial Times: Sources: Leopold Aschenbrenner discussed “Situational Awareness 2.0”, a reboot using leverage and outside capital, fo… | ↗Not briefed |
rss bing-news-ai | 1d | Ramaphosa wants South Africans to own the future of artificial intelligence | President Cyril Ramaphosa wants South Africa to create its own artificial intelligence technology, bringing new jobs and opportunities… | ↗Not briefed |
rss bing-news-ai | 1d | Ramaphosa wants South Africans to own the future of artificial intelligence | President Cyril Ramaphosa wants South Africa to create its own artificial intelligence technology, bringing new jobs and opportunities… | ↗Not briefed |
rss bing-news-ai | 1d | Ramaphosa wants South Africans to own the future of artificial intelligence | President Cyril Ramaphosa wants South Africa to create its own artificial intelligence technology, bringing new jobs and opportunities… | ↗Not briefed |
bluesky bluesky-ai | 1d | Source: www.niemanlab.org/2026/11/afte... | Source: www.niemanlab.org/2026/11/afte... | ↗Not briefed |
bluesky bluesky-ai | 1d | Astra-6.2 🧐 | Astra-6.2 🧐 | ↗Not briefed |
hn hn-ai | 1d | On-device AI: new knowledge without fine-tuning | — | ↗Not briefed |
bluesky bluesky-ai | 1d | and to think i criticized openai for skipping o2 | and to think i criticized openai for skipping o2 | ↗Not briefed |
bluesky bluesky-ai | 1d | Anthropic prices its IPO at $83 per share, valuing the AI giant at $1.5 trillion ahead of its stock market debut on Thursday. | Anthropic prices its IPO at $83 per share, valuing the AI giant at $1.5 trillion ahead of its stock market debut on Thursday. | ↗Not briefed |
arxiv arxiv-ai | 1d | When Objectives and Evaluators Disagree: Evaluating Transfer for Post-hoc Demographic Framing Mitigation | Post-hoc rewriting is used to reduce demographic disparities in LLM-generated text when the source model is inaccessible. A rewrite ca… | ↗Not briefed |
arxiv arxiv-ai | 1d | Keel: A Frontier Model Teaches a Vision-Language Model by Rewriting Its Weights | Can a frontier model repair a vision-language model (VLM) by rewriting its weights? Small reasoning VLMs such as Qwen3.5 need such a r… | ↗Not briefed |