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 | Anthropic goes public and then $GOOG takes the market is my preferred order of event, in case the universe was wondering. | Anthropic goes public and then $GOOG takes the market is my preferred order of event, in case the universe was wondering. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Pope: hahaha haha hahaha! Anthropic: "I decline to disclose the conversation." | Pope: hahaha haha hahaha! Anthropic: "I decline to disclose the conversation." | ↗Not briefed |
bluesky bluesky-ai | 1mo | “I woodworked something and then my kid brutally owned me with an offhand remark. Lord of the Rings” - from the LLM of Joe | “I woodworked something and then my kid brutally owned me with an offhand remark. Lord of the Rings” - from the LLM of Joe | ↗Not briefed |
bluesky bluesky-ai | 1mo | If your review article could have been written by an LLM, it is not a good review article. An LLM can summarize a (large) body of literatur… | If your review article could have been written by an LLM, it is not a good review article. An LLM can summarize a (large) body of lite… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic does not apply the same protections to the data labelers they use to clean their training data from the worst shit in the world. … | Anthropic does not apply the same protections to the data labelers they use to clean their training data from the worst shit in the wo… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "it's not about raising kids -- it's about surfacing pursuits that maximize my potential." - From the LLM of Joe | "it's not about raising kids -- it's about surfacing pursuits that maximize my potential." - From the LLM of Joe | ↗Not briefed |
bluesky bluesky-ai | 1mo | having a big ai model write a ton of big plans, then forking the session and using a (perhaps) small model to chat about it, learn about it… | having a big ai model write a ton of big plans, then forking the session and using a (perhaps) small model to chat about it, learn abo… | ↗Not briefed |
hn hn-ai | 1mo | DeLM: Decentralized Multi-Agent Systems with Shared Context | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: jev-bisect – Find answers with Jev via higher, lower, or | <p><pre><code> ! pip install jev-bisect ! export TYPESAFE_API_KEY="your-typesafe-api-key" </code></pre> -------<p><p… | ↗Not briefed |
bluesky bluesky-ai | 1mo | from the llm of joe | from the llm of joe | ↗Not briefed |
hn hn-ai | 1mo | Themeable Lightweitght Spotify Player | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Google $GOOG Launches Workplace AI Agent That Acts Like Colleague" (Bloomberg) Google Cloud CEO: "It chooses the best model for the job, … | "Google $GOOG Launches Workplace AI Agent That Acts Like Colleague" (Bloomberg) Google Cloud CEO: "It chooses the best model for the… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Mesh-Native Tokenisation: Retrofitting Foundation Models to Unstructured Data | Physics foundation models are an increasingly popular approach for building versatile emulators for spatiotemporal systems. They requi… | ↗Not briefed |
arxiv arxiv-ai | 1mo | GeopoliticalQA: Benchmarking Retrieval and Generation over Intelligence Reports | Geopolitical and strategic intelligence reports constitute a rich source of expert knowledge and have increasing value in the current… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Sam Altman (CEO OpenAI) beschreibt hier sein Geschäftsmodell: Dass Denken zur bezahlten Ware wird. Realistisch angesichts der PISA-Ergebnis… | Sam Altman (CEO OpenAI) beschreibt hier sein Geschäftsmodell: Dass Denken zur bezahlten Ware wird. Realistisch angesichts der PISA-Erg… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Matemáticos contra la IA 💪💪💪💪 ❗️"Los matemáticos no pedimos que hagan este trabajo" ❗️"Liberar más de 700 documentos de una sola vez no… | Matemáticos contra la IA 💪💪💪💪 ❗️"Los matemáticos no pedimos que hagan este trabajo" ❗️"Liberar más de 700 documentos de una sola v… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Crashlytics for AI Coding Agents | — | ↗Not briefed |
hn hn-ai | 1mo | Prehistoric global migration of vanishing gut microbes with humans | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | "is the latest AI model of man or of the father, or perhaps of two spirits at once?" oh my god shut up we're going back to paganism and yo… | "is the latest AI model of man or of the father, or perhaps of two spirits at once?" oh my god shut up we're going back to paganism a… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Macro models finally get their own AI assistant 🤖 Oxford Economics launched AskOEAI Model Assistant, an AI for macro modelling, letting u… | Macro models finally get their own AI assistant 🤖 Oxford Economics launched AskOEAI Model Assistant, an AI for macro modelling, lett… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Literally OpenAI makes money if you light all your tokens on fire for stupid shit. Dont do that. | Literally OpenAI makes money if you light all your tokens on fire for stupid shit. Dont do that. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Some of AI’s biggest risks arise before a model is released to the public. Recent hacks show why regulators need rules for testing, researc… | Some of AI’s biggest risks arise before a model is released to the public. Recent hacks show why regulators need rules for testing, re… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Oh, was it the "I'll get AI to read my kid a bedtime story but set up a voice model of me" one? Because, yeah, I can see that being a misj… | Oh, was it the "I'll get AI to read my kid a bedtime story but set up a voice model of me" one? Because, yeah, I can see that being a… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Knowing Is Not Telling: Identity Ossification in LLM Representations of Trans-border Communities🔧 GPT-5 | Ethnic and cultural identities rarely align with political borders, a tension made sharper by migration and long-standing cross-border… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: K10s – A Clickable Kubernetes TUI (Go, Bubble Tea) | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | replace the word "LLM" with "5 orbs" to check if your use case is insane ✅ I use 5 orbs to analyze meeting notes ❌ 5 orbs read bedtime sto… | replace the word "LLM" with "5 orbs" to check if your use case is insane ✅ I use 5 orbs to analyze meeting notes ❌ 5 orbs read bedtim… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Self-Preservation in Language Models | Continual learning (CL) seeks to adapt pretrained models while preserving prior capabilities. Yet the tasks supplied for preservation… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Reflective Prompt Optimization of a Deep-Research Agent: A Topic-Disjoint, Multi-Seed Protocol for Measuring Transferable Gains and Prompt-Level Overfitting | Prompt optimization is an attractive alternative to reinforcement learning for adapting language-model agents, since it is cheap and p… | ↗Not briefed |
bluesky bluesky-ai | 1mo | These were more predictable than trying to decide whether to follow the AI model or the Euro or the GFS. 🌀 #GulfCoast | These were more predictable than trying to decide whether to follow the AI model or the Euro or the GFS. 🌀 #GulfCoast | ↗Not briefed |
bluesky bluesky-ai | 1mo | Programming the LLM chatbots to interact with people like a professional and not like a slave, servant, or friend actually could help train… | Programming the LLM chatbots to interact with people like a professional and not like a slave, servant, or friend actually could help… | ↗Not briefed |
arxiv arxiv-ai | 1mo | STAPLE: Static Sparse Upcycling of Pretrained Checkpoints via Embedding Modules | Sparse computation can scale the parametric capacity of LLMs without proportionally increasing active computation, but pretraining spa… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Beyond Binary Verifiable Rewards: Behavioral Self-Rewarding for SWE Agents | Reinforcement learning with verifiable rewards (RLVR) provides reliable supervision for software engineering agents through executable… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic changes usage policy to ban model abuse and election interference | Anthropic changes usage policy to ban model abuse and election interference | ↗Not briefed |
rss techcrunch-ai | 1mo | Ben Affleck is an AI nerd, and the internet is impressed | Ben Affleck is going viral for his deep knowledge of AI, from neural networks and transformers to open weights. The actor, who sold hi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | You don’t need mathematicians’ permission to do math. The whole fucking point of math is that either the math checks out or it doesn’t. FW… | You don’t need mathematicians’ permission to do math. The whole fucking point of math is that either the math checks out or it doesn’t… | ↗Not briefed |
rss techcrunch-ai | 1mo | Popular AI leaderboard Arena nearly doubles valuation to $3.1B valuation in 10 months | The company behind the popular LMArena leaderboard has raised $200 million led by Lightspeed and Khosla, and is now measuring AI model… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Oh, actually — the first highlighted portion might not be ad hominem at all, it might be intended to establish that OpenAI has no respect f… | Oh, actually — the first highlighted portion might not be ad hominem at all, it might be intended to establish that OpenAI has no resp… | ↗Not briefed |
rss techcrunch-ai | 1mo | OpenAI’s revenue is reportedly $20 billion less than previously projected | It had previously been reported that the AI lab's annualized revenue was some $70 billion, but a new report claims it's a whole lot le… | ↗Not briefed |
bluesky bluesky-ai | 1mo | the rational interpretation is that because they feed user logs into the future models, they want to do this to reduce negative sentiment i… | the rational interpretation is that because they feed user logs into the future models, they want to do this to reduce negative sentim… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I genuinely wonder what would happen if you sent an email and said (incredibly plausibly): "I don't want OpenAI or whoever to train their f… | I genuinely wonder what would happen if you sent an email and said (incredibly plausibly): "I don't want OpenAI or whoever to train th… | ↗Not briefed |
arxiv arxiv-ai | 1mo | SpireBench: Evaluating Long-Horizon Strategic Play of Agentic Models with Slay the Spire | Agentic large language models now tackle complex tasks, yet strategic reasoning over long horizons is under-studied: most agentic benc… | ↗Not briefed |
rss techcrunch-ai | 1mo | Google brings agentic AI to Gemini, starting with businesses | Google is turning Gemini into an AI agent that can plan, execute tasks, and work across business apps and systems. The agent can deleg… | ↗Not briefed |
bluesky bluesky-ai | 1mo | btw, the terrifying doomer possibilities of AI 2027 are literally only possible if frontier models are kept far in advance of what the publ… | btw, the terrifying doomer possibilities of AI 2027 are literally only possible if frontier models are kept far in advance of what the… | ↗Not briefed |
bluesky bluesky-ai | 1mo | As I said elsewhere, it's a novel use of the English Language. "Highly Accepted" makes no sense. Something is either accepted or it isn't. … | As I said elsewhere, it's a novel use of the English Language. "Highly Accepted" makes no sense. Something is either accepted or it is… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Know What You Don't Know: Agentic Data Synthesis by Evolving Generation Space for Closed-Loop Learning | The continued improvement of large language models increasingly depends on synthetic experience, yet effective data synthesis for inte… | ↗Not briefed |
rss the-decoder | 1mo | Some mathematicians call for OpenAI boycott after AI-generated proofs flood their field | The Association for Human Mathematics is calling for an OpenAI boycott after the company released more than 700 AI-generated math manu… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Hab mit Claude ai diskutiert, mir n Skalpell gekauft und bin jetzt selbständig. Was? | Hab mit Claude ai diskutiert, mir n Skalpell gekauft und bin jetzt selbständig. Was? | ↗Not briefed |
rss techcrunch-ai | 1mo | Anthropic changes usage policy to ban model abuse and election interference | Anthropic's updated usage policy explicitly prohibits users from repeatedly abusing Claude in extreme cases, though ordinary frustrati… | ↗Not briefed |
hn hn-ai | 1mo | You're saving one minute and now I can't know if I'm talking a human | I write a golf blog. Recently, it's been getting a nontrivial amount of attention from a very niche section of the nerd-golf comm… | ↗Not briefed |
bluesky bluesky-ai | 1mo | weren't there a bunch of people on here crowing about how openAI was profitable and people talking about how the economics of it are sustai… | weren't there a bunch of people on here crowing about how openAI was profitable and people talking about how the economics of it are s… | ↗Not briefed |