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’m sorry for dragging this over here, but this is very, very funny. | I’m sorry for dragging this over here, but this is very, very funny. | ↗Not briefed |
hn hn-ai | 1mo | Trump Calls AI Fears a Hoax. Inside the White House, the Debate Is More Complex | — | ↗Not briefed |
hn hn-ai | 1mo | Gavin Newsom Wants AI Kill Switch to Boost CA AI Safety | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Predictable Failure in Multi-Hop Retrieval: Score-Distributional Confidence Scoring and Abstention | Multi-hop retrieval failures are not uniformly distributed across queries: they cluster in structurally predictable subpopulations. We… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Lorsqu'on les laisse discuter entre eux assez longtemps, les LLM ont tendance, comme les communautés de locuteurs humains, à inventer leur … | Lorsqu'on les laisse discuter entre eux assez longtemps, les LLM ont tendance, comme les communautés de locuteurs humains, à inventer… | ↗Not briefed |
bluesky bluesky-ai | 1mo | California Governor Gavin Newsom signed an executive order outlining plans to require labs such as OpenAI and Anthropic to implement a 'kil… | California Governor Gavin Newsom signed an executive order outlining plans to require labs such as OpenAI and Anthropic to implement a… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Benchmarking World Models for Continual Learning on Compositional Tasks | A desirable property of a world model is the ability to learn continually across tasks, adapting to new environments without forgettin… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Jeff – A read-only CLI for semantic code review using Jev | — | ↗Not briefed |
rss google-research | 1mo | MilleMiglia: A realistic instance generator for middle-mile logistics | Algorithms & Theory | ↗Not briefed |
arxiv arxiv-ai | 1mo | Particle Competition and Cooperation for Robust Graph Convolutional Network Learning Under Label Noise | Graph Convolutional Networks (GCNs) are highly sensitive to label noise, since corrupted supervision can propagate through the graph a… | ↗Not briefed |
rss the-decoder | 1mo | California Governor Newsom signs executive order demanding "kill switch" for AI models | California Governor Gavin Newsom signed an executive order seeking independent auditors inside AI labs and a "kill switch" for AI mode… | ↗Not briefed |
bluesky bluesky-ai | 1mo | an llm is the grammatically correct option but a lollum is the emotionally correct one | an llm is the grammatically correct option but a lollum is the emotionally correct one | ↗Not briefed |
bluesky bluesky-ai | 1mo | Maybe don't let LLM'S plan and carry out a war. All of our expensive munitions were wasted because operational planning wasn't updated wit… | Maybe don't let LLM'S plan and carry out a war. All of our expensive munitions were wasted because operational planning wasn't update… | ↗Not briefed |
hn hn-ai | 1mo | Jev as a Primitive Feature of Ruby | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | anthropic shreds the rare books instead of just reselling or donating them because they are required to. by copyright law. | anthropic shreds the rare books instead of just reselling or donating them because they are required to. by copyright law. | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI Execs Hoped to Make “Gazillions” After Feeding Model Journalists’ Work #TruthOut | OpenAI Execs Hoped to Make “Gazillions” After Feeding Model Journalists’ Work #TruthOut | ↗Not briefed |
bluesky bluesky-ai | 1mo | So grp sex among bosses & subordinates is not just a private kink: it's emblematic of the business model, which is "obliterating others' ri… | So grp sex among bosses & subordinates is not just a private kink: it's emblematic of the business model, which is "obliterating other… | ↗Not briefed |
rss techmeme | 1mo | Mistral and other European AI startups accuse US rivals of using safety concerns to entrench their dominance, rejecting Anthropic's calls to pace the frontier (Reuters) | Reuters: Mistral and other European AI startups accuse US rivals of using safety concerns to entrench their dominance, rejecting Anthr… | ↗Not briefed |
bluesky bluesky-ai | 1mo | prospect.org/2026/09/11/o... | prospect.org/2026/09/11/o... | ↗Not briefed |
bluesky bluesky-ai | 1mo | Did you not think about and plan for—ahead of time!—even the most basic consequences of your intentional actions? And now that they've come… | Did you not think about and plan for—ahead of time!—even the most basic consequences of your intentional actions? And now that they've… | ↗Not briefed |
bluesky bluesky-ai | 1mo | This is horrifying and the exact reason why we LLM 'critics' are so enraged by these companies lying about their products' abilities. What … | This is horrifying and the exact reason why we LLM 'critics' are so enraged by these companies lying about their products' abilities.… | ↗Not briefed |
hn hn-ai | 1mo | Benchmarking Wild vs. Mold | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Available Guardrails: Certifying Selective Prediction across ML Systems | A selective predictor acts as a safety gate: it returns an output only when the prediction appears sufficiently trustworthy. Deploymen… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I mean didn’t openAI just give Chuck Schumers daughter a high level job? Pretty sure I read that. | I mean didn’t openAI just give Chuck Schumers daughter a high level job? Pretty sure I read that. | ↗Not briefed |
arxiv arxiv-ai | 1mo | An Interpretable Memory Decision Controller for LLM Agents Based on Three-Signal Complementarity: Decoupling Confidence and Consistency | Memory systems for large language models have focused predominantly on efficient retrieval, whereas the decision of whether retrieved… | ↗Not briefed |
rss techcrunch-ai | 1mo | Google’s new ‘CC’ is an AI agent that helps families run their households | Google is refocusing its CC AI agent on household coordination, letting families share emails, schedules, and tasks so the AI can mana… | ↗Not briefed |
arxiv arxiv-ai | 1mo | $λ$-Controlled GRPO: Turning Flow-Matching Ratio Instability into a Budgeted Resource | Reinforcement learning is increasingly used to align image generators with reward signals, and Flow-GRPO recently extended this paradi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | i heard someone say an llm is like hooking up a weed whacker to a dog and then saying what happens next isn't their fault | i heard someone say an llm is like hooking up a weed whacker to a dog and then saying what happens next isn't their fault | ↗Not briefed |
arxiv arxiv-ai | 1mo | Gricea: An Open Science Platform for Conversational AI Research | We need studies on conversational AI (CAI) at scale to understand human behavior and shape CAI design. However, fragmented reporting o… | ↗Not briefed |
arxiv arxiv-ai | 1mo | QuranicMMLU: A Cognitively-Aware Benchmark for Evaluating Generative AI Solutions on Quranic Linguistic Knowledge | We introduce QuranicMMLU, a benchmark for evaluating generative AI on Quranic Arabic across multiple dimensions of linguistic complexi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 🤖 Beyond the Ground · AI & DeepTech Anthropic's Lab: How Robot-Run Preclinical Automation Rewrites Pharma R&D and Research Jobs Anthropi… | 🤖 Beyond the Ground · AI & DeepTech Anthropic's Lab: How Robot-Run Preclinical Automation Rewrites Pharma R&D and Research Jobs Ant… | ↗Not briefed |
hn hn-ai | 1mo | Scam Spotting with ChatGPT🔧 ChatGPT | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | US cybersecurity researchers who conducted hack say ‘scope of what we could theoretically access was huge’ | US cybersecurity researchers who conducted hack say ‘scope of what we could theoretically access was huge’ | ↗Not briefed |
hn hn-ai | 1mo | NanoJev | — | ↗Not briefed |
rss the-decoder | 1mo | Security researchers used Anthropic's Claude to hack OpenAI's internal systems in under 72 hours | Three security researchers used Anthropic's Claude models to break into OpenAI's internal systems through its community forum in less… | ↗Not briefed |
bluesky bluesky-ai | 1mo | apparently this take got me blocked because "anthropic shredding rare books". culture war gonna culture war I guess huh | apparently this take got me blocked because "anthropic shredding rare books". culture war gonna culture war I guess huh | ↗Not briefed |
bluesky bluesky-ai | 1mo | An OpenAI executive similarly warned that publishers would face an “existential threat” from commercial products trained on news content in… | An OpenAI executive similarly warned that publishers would face an “existential threat” from commercial products trained on news conte… | ↗Not briefed |
bluesky bluesky-ai | 1mo | With now airtight proof, I hope my friends allow themselves to be wrong and finish denouncing genAI 🤞 OpenAI/Microsoft confesses in now u… | With now airtight proof, I hope my friends allow themselves to be wrong and finish denouncing genAI 🤞 OpenAI/Microsoft confesses in… | ↗Not briefed |
hn hn-ai | 1mo | Non-Obvious Nix Options | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Timeline, a team project planner with a fast canvas Gantt view | I manage a few projects at work. I really am unsatisfied with modern tools available to me for that purpose. I had always envisioned a… | ↗Not briefed |
hn hn-ai | 1mo | Open-weight models take 56% of token volume, Astra doubles Fable 5.1 spend | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Dario Amodei and other AI leaders want to ‘Pace the Frontier’ but…how? | Dario Amodei and other AI leaders want to ‘Pace the Frontier’ but…how? | ↗Not briefed |
bluesky bluesky-ai | 1mo | There's no point at which turning your brain off will work: danluu.com/brain-off/ | There's no point at which turning your brain off will work: danluu.com/brain-off/ | ↗Not briefed |
hn hn-ai | 1mo | Tiny-Vedas: a generic accelerator interface for RISC-V, from PyTorch op to GDS | — | ↗Not briefed |
rss techmeme | 1mo | Sony Music and UMG sue Suno, alleging that its new model trained in partnership with WMG and BMG infringes copyright because it is based on an infringing model (Corbin Bolies/Variety)🔧 Suno | Corbin Bolies / Variety: Sony Music and UMG sue Suno, alleging that its new model trained in partnership with WMG and BMG infringes co… | ↗Not briefed |
rss techcrunch-ai | 1mo | Dario Amodei and other AI leaders want to ‘Pace the Frontier’ but…how? | A week after an Anthropic researcher’s doomsday warning rattled the AI world, the company’s CEO Dario Amodei has outlined his plan to… | ↗Not briefed |
bluesky bluesky-ai | 1mo | In this Philadelphia sports bar, OpenAI still means no one is guarding Allen Iverson. But they do know this: Even if the machines take all … | In this Philadelphia sports bar, OpenAI still means no one is guarding Allen Iverson. But they do know this: Even if the machines take… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Huh, it's so weird how the AI that UnitedHealthcare uses overwhelmingly *rejects* claims by accident instead of overwhelmingly *approving* … | Huh, it's so weird how the AI that UnitedHealthcare uses overwhelmingly *rejects* claims by accident instead of overwhelmingly *approv… | ↗Not briefed |
arxiv arxiv-ai | 1mo | COMPLEX: A Closed-Form Certified Embedding of Multiparameter Persistence Modules | Every multiparameter persistence vectorization we know of carries a one-sided Lipschitz upper bound and nothing below it: without a lo… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Yesterday, Jev support landed in the Laravel AI SDK. We're now using it in There There to keep spam out of your inbox. Jev isn't an LLM. Y… | Yesterday, Jev support landed in the Laravel AI SDK. We're now using it in There There to keep spam out of your inbox. Jev isn't an L… | ↗Not briefed |