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 |
|---|---|---|---|---|
hn hn-ai | 10d | With All Due Respect, Fuck You, Claude | — | ↗Not briefed |
hn hn-ai | 10d | Muhammad on X: "Claude reverse engineered my 32-year-old Jeep | — | ↗Not briefed |
rss techmeme | 10d | Rox, which develops AI agents to automate customer research and outreach, raised a $100M Series B co-led by General Catalyst and March at a $1.3B valuation (Morgan Chittum/Fortune) | Morgan Chittum / Fortune: Rox, which develops AI agents to automate customer research and outreach, raised a $100M Series B co-led by… | ↗Not briefed |
arxiv arxiv-ai | 10d | What Conformal Selection Can Certify in CRISPR Off-Target Design:Benchmark Leakage, Free Baselines, and a Finite-Sample Cohort Requirement | Conformal selection promises distribution-free, finite-sample control of the error rate of a decision, and has been proposed as a safe… | ↗Not briefed |
hn hn-ai | 10d | Show HN: Prefixwise – names why each LLM request missed the prompt cache | — | ↗Not briefed |
bluesky bluesky-ai | 10d | 🤣 | 🤣 | ↗Not briefed |
rss bing-news-ai | 10d | First workplace artificial intelligence guide released | A draft plan for AI adoption in the workplace has been released by the federal government, which says it could stop hundreds of jobs f… | ↗Not briefed |
rss bing-news-ai | 10d | First workplace artificial intelligence guide released | A draft plan for AI adoption in the workplace has been released by the federal government, which says it could stop hundreds of jobs f… | ↗Not briefed |
bluesky bluesky-ai | 10d | set up a local llm out of morbid curiosity: a) its impressive that linear alebgra can do this b) i would go insane if i had to rely on this… | set up a local llm out of morbid curiosity: a) its impressive that linear alebgra can do this b) i would go insane if i had to rely on… | ↗Not briefed |
bluesky bluesky-ai | 10d | living in the world that "THEY TOOK OUR JERBS" s a justification for violent (yes, telling trans LLM users to detransition is violent) reac… | living in the world that "THEY TOOK OUR JERBS" s a justification for violent (yes, telling trans LLM users to detransition is violent)… | ↗Not briefed |
bluesky bluesky-ai | 10d | Also, we use real human models (most often my wife, but also some actor friends). People want to see how stuff hangs on a person. Not on so… | Also, we use real human models (most often my wife, but also some actor friends). People want to see how stuff hangs on a person. Not… | ↗Not briefed |
hn hn-ai | 10d | LLM prefill, what we got wrong | — | ↗Not briefed |
hn hn-ai | 10d | Show HN: Nib – an on-device grammar checker for macOS (open source) | — | ↗Not briefed |
bluesky bluesky-ai | 10d | Stated far more eloquently than I ever could. Brilliant reading for anyone who is concerned about the ethics of the whole House House LLM d… | Stated far more eloquently than I ever could. Brilliant reading for anyone who is concerned about the ethics of the whole House House… | ↗Not briefed |
bluesky bluesky-ai | 10d | See i can also quote documents from authorities agreeing with me :-) | See i can also quote documents from authorities agreeing with me :-) | ↗Not briefed |
bluesky bluesky-ai | 10d | Naprawiłem zmywarkę. Po 5l pompa odmówiła tyrania. Nowa założona. Leje się z węża, szlag mnie trafia. Kupiony silikon rtv (czarny). Uszczel… | Naprawiłem zmywarkę. Po 5l pompa odmówiła tyrania. Nowa założona. Leje się z węża, szlag mnie trafia. Kupiony silikon rtv (czarny). Us… | ↗Not briefed |
hn hn-ai | 10d | Show HN: HLedger Cal – your money habits over time | — | ↗Not briefed |
bluesky bluesky-ai | 10d | Not an academic but an IT guy who reads too much archaeology. Short version is "need" is a stretch, AI/ML (not necessarily an LLM like a ch… | Not an academic but an IT guy who reads too much archaeology. Short version is "need" is a stretch, AI/ML (not necessarily an LLM like… | ↗Not briefed |
bluesky bluesky-ai | 10d | Can’t believe he would say that about something “quintessentially transformative” [Bartz et al v. Anthropic PBC, (N.D. Cal. 2025) (ORDER ON… | Can’t believe he would say that about something “quintessentially transformative” [Bartz et al v. Anthropic PBC, (N.D. Cal. 2025) (ORD… | ↗Not briefed |
hn hn-ai | 10d | "Agent" in SaaS earnings filings: 87 mentions in Q1 2024, 478 in Q3 2026 | — | ↗Not briefed |
bluesky bluesky-ai | 10d | AI/indie radar tonight: 1. ChatGPT beta: eligible US merchants can claim a business page and run a Business Agent 2. Google's Gemini 4 Carb…🔧 ChatGPT | AI/indie radar tonight: 1. ChatGPT beta: eligible US merchants can claim a business page and run a Business Agent 2. Google's Gemini 4… | ↗Not briefed |
arxiv arxiv-ai | 10d | Testing Conversation Preservation Across LLM APIs: A Rule-Based Oracle for Detecting Silent Information Loss | LLM applications increasingly rely on translators and compatibility layers to move conversations across APIs. Yet a translated request… | ↗Not briefed |
bluesky bluesky-ai | 10d | I like this as a framework for AI disclosure, but there's still the matter of clarifying the sentience: 0: I did not use an LLM R: I revie… | I like this as a framework for AI disclosure, but there's still the matter of clarifying the sentience: 0: I did not use an LLM R: I… | ↗Not briefed |
bluesky bluesky-ai | 10d | This is what we mean when we say “Hallucinations are not a solved problem and likely never will be.” This is inherent to how an LLM works. … | This is what we mean when we say “Hallucinations are not a solved problem and likely never will be.” This is inherent to how an LLM wo… | ↗Not briefed |
rss the-decoder | 10d | Anthropic says AI is eroding Nvidia's CUDA lead, cutting an AMD chip switch from months to three days | Anthropic claims it used AI itself to breach Nvidia's CUDA moat. The company's in-house model Mythos reportedly ported Claude to AMD G… | ↗Not briefed |
arxiv arxiv-ai | 10d | Faithful in Steps: Improving Generalization and Citation in RAG via Query Decomposition🔧 Modal | Retrieval-augment generation is a prevalent strategy to mitigate hallucinations of LLMs. The attributable RAG (RAGQ) generates quotes… | ↗Not briefed |
bluesky bluesky-ai | 10d | Doesn't change all the other evils of it, but it's better than theft - assuming they're telling the truth. Meanwhile, the slop machine ver… | Doesn't change all the other evils of it, but it's better than theft - assuming they're telling the truth. Meanwhile, the slop machin… | ↗Not briefed |
bluesky bluesky-ai | 10d | This AM’s EntryPoint: RETAIL RUNS BACK THE DEEPSEEK PLAYOOK Tuesday was the 2nd-largest day for net purchases of $NVDA on Robinhood. Mond… | This AM’s EntryPoint: RETAIL RUNS BACK THE DEEPSEEK PLAYOOK Tuesday was the 2nd-largest day for net purchases of $NVDA on Robinhood.… | ↗Not briefed |
rss lastweekin-ai | 10d | Last Week (Month) in AI #347 - Nvidia's spy, Anthropic's $1.4T IPO, AI's blue wave | Catching up on five weeks: Huang under investigation, Anthropic and OpenAI priced at $1.4T and $1.25T as spending stalls, Anthropic wa… | ↗Not briefed |
hn hn-ai | 10d | Pharo in the Browser with WebAssembly | — | ↗Not briefed |
rss techmeme | 10d | After the US midterms, Palantir CEO Alex Karp pivots toward Democrats, describing his philosophy as progressive and acknowledging ICE's enforcement missteps (Politico) | Politico: After the US midterms, Palantir CEO Alex Karp pivots toward Democrats, describing his philosophy as progressive and acknowle… | ↗Not briefed |
hn hn-ai | 10d | Anthropic Envy on HN | It is clearly there: wishing that they fail, bad faith arguments about how badly they are doing, etc, etc<p>What could be the source?… | ↗Not briefed |
bluesky bluesky-ai | 10d | It's already affecting ROM hackers and fan game developers, but why now does Claude have some ethics towards IP? Did Nintendo come knocking… | It's already affecting ROM hackers and fan game developers, but why now does Claude have some ethics towards IP? Did Nintendo come kno… | ↗Not briefed |
arxiv arxiv-ai | 10d | Analysis choices decide whether untrained networks beat trained ones at V1: a multiverse analysis of model-brain RSA | Untrained convolutional networks have been reported to rival trained networks as models of visual cortex, and we had reported an untra… | ↗Not briefed |
bluesky bluesky-ai | 10d | We dont slap lobbyists enough www.businessinsider.com/openai-ai-da... | We dont slap lobbyists enough www.businessinsider.com/openai-ai-da... | ↗Not briefed |
arxiv arxiv-ai | 10d | MAGIC: Modality-Aware Gini Inspection for Clinical Hallucination Detection in LMMs | Recent advances in medical large multimodal models (LMMs) have brought promising progress to clinical decision support, yet hallucinat… | ↗Not briefed |
bluesky bluesky-ai | 10d | OpenAI’s ‘shepherds’ affirm blockbuster physics result. The company’s appointed team of experts has verified and released the groundbreakin… | OpenAI’s ‘shepherds’ affirm blockbuster physics result. The company’s appointed team of experts has verified and released the groundbr… | ↗Not briefed |
arxiv arxiv-ai | 10d | One Hierarchy, Many Sports: Hierarchy-Aligned Multi-Task Learning for Next-Shot Prediction | Next-shot prediction in racket sports requires learning from event data with different match structures, recorded attributes, and labe… | ↗Not briefed |
bluesky bluesky-ai | 10d | OpenAI's $13.5bln in operating losses (inc. stock comp) on $10bln in revenue is not a particularly compelling business model. www.wsj.com/… | OpenAI's $13.5bln in operating losses (inc. stock comp) on $10bln in revenue is not a particularly compelling business model. www.wsj… | ↗Not briefed |
arxiv arxiv-ai | 10d | An Approach to Hinton's Mother-Child Proposal: Neural Networks Learning and Preserving Care in a Finite World | Hinton proposed that artificial intelligence should be given "maternal instincts" and relate to us as a mother to her child. We argue… | ↗Not briefed |
arxiv arxiv-ai | 10d | GHAgentCI: A Dataset of Coding-Agent Invocations in CI Workflow Histories | Coding agents are increasingly integrated into GitHub Actions to review pull requests, triage issues and modify code. Identifying thes… | ↗Not briefed |
arxiv arxiv-ai | 10d | DecArena: Evaluating Large-Scale Decentralized Multi-Agent Collaboration | Emerging personal AI agents raise fundamental scalability and privacy concerns for centralized multi-agent systems (MASs), which rely… | ↗Not briefed |
bluesky bluesky-ai | 10d | Someone said Anthropic said to sell-side "The contract with SpaceX will not be extended beyond the middle of next year." Maybe the reason o… | Someone said Anthropic said to sell-side "The contract with SpaceX will not be extended beyond the middle of next year." Maybe the rea… | ↗Not briefed |
hn hn-ai | 10d | Show HN: Tensward – Local AI inference optimisation | — | ↗Not briefed |
hn hn-ai | 10d | Show HN: BackTrail, keep each tab's Back history across browser restarts | — | ↗Not briefed |
bluesky bluesky-ai | 10d | the Unity engineer going “this can’t be done” and the LLM going “hmmm but what about this” is actually a great example of why this technolo… | the Unity engineer going “this can’t be done” and the LLM going “hmmm but what about this” is actually a great example of why this tec… | ↗Not briefed |
hn hn-ai | 10d | Show HN: Hackernewstime – Hacker News stories, evolving over time, ranked | Fellow Hackers,<p>I love reading and commenting in this board but I needed a better interface, so I created hackernewstime.com with C&… | ↗Not briefed |
arxiv arxiv-ai | 10d | Incentive-Responsive Agents: A Principal-Agent Framework, Metric Suite, and Runtime Architecture for Long-Lived Personal AI Agents | Personal AI agents now persist: they hold accounts, mailboxes, long-term memory, and open-ended task streams, yet the dominant way to… | ↗Not briefed |
bluesky bluesky-ai | 10d | New York requires frontier AI developers like OpenAI and Anthropic to register by January 1, 2027. Companies making upwards of $500 million… | New York requires frontier AI developers like OpenAI and Anthropic to register by January 1, 2027. Companies making upwards of $500 mi… | ↗Not briefed |
bluesky bluesky-ai | 10d | nobody who is on or using openly LLM-coded bluesky has any business memoryholing an indie dev for using AI for QA and then apologizing for … | nobody who is on or using openly LLM-coded bluesky has any business memoryholing an indie dev for using AI for QA and then apologizing… | ↗Not briefed |