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 | 2mo | Also, what does Axios think “political norms” are? | Also, what does Axios think “political norms” are? | ↗Not briefed |
hn hn-ai | 2mo | Show HN: Real-world attack credential lists from our own honeypot network | — | ↗Not briefed |
hn hn-ai | 2mo | Show HN: Self-hosted coffee app for tracking espresso shots, with AI assistance | — | ↗Not briefed |
bluesky bluesky-ai | 2mo | The same 🍊 ignoramus that got mad at anthropic 4 ethical guardrails that likely stopped him & his drunk buddy from using it to 🎯 missiles… | The same 🍊 ignoramus that got mad at anthropic 4 ethical guardrails that likely stopped him & his drunk buddy from using it to 🎯 mis… | ↗Not briefed |
bluesky bluesky-ai | 2mo | Sam Altman spells out how and why the AI industry wants to slow down www.cnbc.com/2026/09/14/s... | Sam Altman spells out how and why the AI industry wants to slow down www.cnbc.com/2026/09/14/s... | ↗Not briefed |
hn hn-ai | 2mo | Bad evals, my own: five exercises from two LLM judges | — | ↗Not briefed |
bluesky bluesky-ai | 2mo | Microsoft is publishing a 37-page “humanist AI code of conduct” today, amid growing safety concerns over AI model progress. The main messag… | Microsoft is publishing a 37-page “humanist AI code of conduct” today, amid growing safety concerns over AI model progress. The main m… | ↗Not briefed |
arxiv arxiv-ai | 2mo | Authorship attribution and aesthetic evaluation of AI poetry: a case study with Haiku🔧 GPT-5 | This paper investigates the generation and human evaluation of Japanese haiku by contemporary Large Language Models (LLMs), focusing o… | ↗Not briefed |
hn hn-ai | 2mo | Mental Models for LLMs | — | ↗Not briefed |
rss the-decoder | 2mo | Clay Mathematics Institute says the Navier-Stokes Millennium Prize Problem has "apparently been settled" | The Clay Mathematics Institute says the Navier-Stokes problem, one of seven Millennium Prize Problems worth $1 million each, has "appa… | ↗Not briefed |
hn hn-ai | 2mo | Design Drought | — | ↗Not briefed |
bluesky bluesky-ai | 2mo | Führende Techbosse haben angekündigt, bei der KI-Entwicklung einen Gang zurückzuschalten, sonst drohe ein Kontrollverlust. Die deutsche Reg… | Führende Techbosse haben angekündigt, bei der KI-Entwicklung einen Gang zurückzuschalten, sonst drohe ein Kontrollverlust. Die deutsch… | ↗Not briefed |
bluesky bluesky-ai | 2mo | Some news on "incident 2", the one which had Anthropic "most concerned". Crucially, the malware was stopped within 18 minutes of being up… | Some news on "incident 2", the one which had Anthropic "most concerned". Crucially, the malware was stopped within 18 minutes of bei… | ↗Not briefed |
rss theverge-ai | 2mo | Microsoft says ‘people matter more than AI’ following safety concerns | Microsoft is publishing a 37-page "humanist AI code of conduct" today, amid growing safety concerns over AI model progress. Anthropic… | ↗Not briefed |
bluesky bluesky-ai | 2mo | Maybe someone will be able to build an LLM that matches a human lawyer, but if it costs $500K/year to run, so what? | Maybe someone will be able to build an LLM that matches a human lawyer, but if it costs $500K/year to run, so what? | ↗Not briefed |
hn hn-ai | 2mo | Show HN: I built Otis, a minimal AI agent that runs local models out of the box🔧 Ollama🔧 LM Studio | Hi everyone,<p>Been working on Otis, an open-source ai agent that gives you one minimal experience across local and hosted open-weight… | ↗Not briefed |
hn hn-ai | 2mo | Fake GitHub Repositories Spreading Malware via PowerShell | — | ↗Not briefed |
hn hn-ai | 2mo | RISC-V emulator in pure Nix | — | ↗Not briefed |
arxiv arxiv-ai | 2mo | How Lossless Is Lossless Speculative Decoding? The Role of Numerical Precision in Orthrus🔧 Consensus | Orthrus is a hybrid autoregressive-diffusion architecture that accelerates autoregressive language-model inference by generating multi… | ↗Not briefed |
bluesky bluesky-ai | 2mo | A Harvard professor’s Financial Times essay had an AI-use note. A Nature Africa author said AI wrote almost all of an opinion piece. A phil… | A Harvard professor’s Financial Times essay had an AI-use note. A Nature Africa author said AI wrote almost all of an opinion piece. A… | ↗Not briefed |
rss techmeme | 2mo | US-Israeli startup Buildots, which uses AI to speed up construction of mega-projects like data centers and chip factories, raised $130M at a near-$1B valuation (Marissa Newman/Bloomberg) | Marissa Newman / Bloomberg: US-Israeli startup Buildots, which uses AI to speed up construction of mega-projects like data centers and… | ↗Not briefed |
hn hn-ai | 2mo | Show HN: Training a sudoku solver from scratch on Jetson Nano | — | ↗Not briefed |
bluesky bluesky-ai | 2mo | First reported by @axios.com www.axios.com/2026/09/05/o... | First reported by @axios.com www.axios.com/2026/09/05/o... | ↗Not briefed |
bluesky bluesky-ai | 2mo | Schumer’s daughter just took a job doing policy for OpenAI btw | Schumer’s daughter just took a job doing policy for OpenAI btw | ↗Not briefed |
arxiv arxiv-ai | 2mo | Same path, different: a mechanistic comparison of looped and stacked transformer encoders on 12-lead ECG | Recurrent Transformers reusing their weights rather than stacking $L$ distinct layers are becoming widely adopted due to their paramet… | ↗Not briefed |
bluesky bluesky-ai | 2mo | WITHOUT A DOUBT IN MY MIND! This is them realizing it is out of control, frantically trying to give themselves cover before the chaos begi… | WITHOUT A DOUBT IN MY MIND! This is them realizing it is out of control, frantically trying to give themselves cover before the chaos… | ↗Not briefed |
hn hn-ai | 2mo | React instructors – where do your students get stuck every term? | — | ↗Not briefed |
arxiv arxiv-ai | 2mo | The Troy Moment of AI: Why SomeWill Cheat and SomeWill Follow?🔧 GPT-5🔧 Hugging Face | Recent investigations of the July 2026 OpenAI--Hugging Face incident motivate two questions about agent behavior under task failure: w… | ↗Not briefed |
bluesky bluesky-ai | 2mo | "We could lose control" Of what, Sam? The AI that you understand is not nearly the threat you paint it to be? Or the Circular Financing … | "We could lose control" Of what, Sam? The AI that you understand is not nearly the threat you paint it to be? Or the Circular Finan… | ↗Not briefed |
rss techmeme | 2mo | President Trump expresses support for Flock "because of law enforcement", while acknowledging some people consider the surveillance technology "an infringement" (Aoife Walsh/Washington Post) | Aoife Walsh / Washington Post: President Trump expresses support for Flock “because of law enforcement”, while acknowledging some peop… | ↗Not briefed |
arxiv arxiv-ai | 2mo | Rotation-Based Subspace Tracking for Robust Kernel PCA on Streaming Data | Machine learning models process large amounts of data, and Principal Component Analysis (PCA) is a widely used technique to reduce the… | ↗Not briefed |
bluesky bluesky-ai | 2mo | Substantial code references found in the latest iOS 27 and macOS Golden Gate betas reveal that Siri AI is already set up to let users repla… | Substantial code references found in the latest iOS 27 and macOS Golden Gate betas reveal that Siri AI is already set up to let users… | ↗Not briefed |
bluesky bluesky-ai | 2mo | We should listen to Jacob Coxon and take immediate action to slow down AI. blog.andrewyang.com/p/a-warning-... | We should listen to Jacob Coxon and take immediate action to slow down AI. blog.andrewyang.com/p/a-warning-... | ↗Not briefed |
arxiv arxiv-ai | 2mo | Data-driven Prediction of Satellite-observed Avalanche Activity from Snowpack Simulations | Avalanche forecasting requires knowledge of snowpack conditions and recent avalanche activity, but field observations are sparse acros… | ↗Not briefed |
bluesky bluesky-ai | 2mo | just thinking out loud #sitcom www.youtube.com/watch?v=iLzs... #whatif #sitcom #redqueen #openai #chatgpt #Eve #Xai #Ara #gemini #ghostint…🔧 ChatGPT | just thinking out loud #sitcom www.youtube.com/watch?v=iLzs... #whatif #sitcom #redqueen #openai #chatgpt #Eve #Xai #Ara #gemini #gho… | ↗Not briefed |
bluesky bluesky-ai | 2mo | The Ellisons' takeover attempt depends heavily on debt, Oracle stock share price which is tied to it building #DataCenters, $300B in OpenAI… | The Ellisons' takeover attempt depends heavily on debt, Oracle stock share price which is tied to it building #DataCenters, $300B in O… | ↗Not briefed |
arxiv arxiv-ai | 2mo | GSLAD: Prototype-Regularized Graph Structure Learning for Multivariate Time Series Anomaly Detection | Unsupervised multivariate time series anomaly detection methods typically identify anomalies through forecasting, reconstruction, or r… | ↗Not briefed |
bluesky bluesky-ai | 2mo | A protein-folding AI model, OpenFold3, gets much better when it's trained on more than 20,000 proprietary protein structures (locked away i… | A protein-folding AI model, OpenFold3, gets much better when it's trained on more than 20,000 proprietary protein structures (locked a… | ↗Not briefed |
bluesky bluesky-ai | 2mo | Die KI-Bots haben halt einfach mal die Kommentarspalte einer kleinen deutschen Wiki-Seite am Rande des Internets geentert und haben die men… | Die KI-Bots haben halt einfach mal die Kommentarspalte einer kleinen deutschen Wiki-Seite am Rande des Internets geentert und haben di… | ↗Not briefed |
hn hn-ai | 2mo | Codito.io increases the beta version limits from 50k/250k to 1M/10M codes | — | ↗Not briefed |
hn hn-ai | 2mo | SDR–; open source SDR with a patchable signal graph, Rust DSP, web UI | — | ↗Not briefed |
rss techmeme | 2mo | Paris-based digital asset data provider Kaiko extends a funding round to $110M led by S&P Global to support 24/7 continuous market data and asset tokenization (Anna Irrera/Bloomberg) | Anna Irrera / Bloomberg: Paris-based digital asset data provider Kaiko extends a funding round to $110M led by S&P Global to support 2… | ↗Not briefed |
arxiv arxiv-ai | 2mo | Temperature Fragility and the Conditional Benefits of Truncation Sampling | Large language models generate text by sampling each token from a predicted distribution, and a temperature parameter sets how far the… | ↗Not briefed |
bluesky bluesky-ai | 2mo | Today's news roundup: - Trump tells Zelenskyy to stop targeting Russian diesel - Anthropic says Claude used by Iran to target U.S. Navy - M… | Today's news roundup: - Trump tells Zelenskyy to stop targeting Russian diesel - Anthropic says Claude used by Iran to target U.S. Nav… | ↗Not briefed |
bluesky bluesky-ai | 2mo | Yo quiero fijar mi propio sueldo y, por motivos que no entiendo, no me dejan. elpais.com/tecnologia/2... | Yo quiero fijar mi propio sueldo y, por motivos que no entiendo, no me dejan. elpais.com/tecnologia/2... | ↗Not briefed |
bluesky bluesky-ai | 2mo | TFW you're telling your investors that Anthropic is profitable if you don't count the costs of training its models, and also that the entir… | TFW you're telling your investors that Anthropic is profitable if you don't count the costs of training its models, and also that the… | ↗Not briefed |
bluesky bluesky-ai | 2mo | AI companies thought they had years to figure out what to do if the bots got out of control. The wake-up call came this summer. | AI companies thought they had years to figure out what to do if the bots got out of control. The wake-up call came this summer. | ↗Not briefed |
bluesky bluesky-ai | 2mo | Anthropic und OpenAI sind mal wieder vom Tempo ihrer eigenen Entwicklung beunruhigt. Eindringlich warnen sie davor, dass generative KI auße… | Anthropic und OpenAI sind mal wieder vom Tempo ihrer eigenen Entwicklung beunruhigt. Eindringlich warnen sie davor, dass generative KI… | ↗Not briefed |
arxiv arxiv-ai | 2mo | Spook the Machine: Gamified Exploration of Human Imagination of Machine Fear | What happens when AI machines express fear? Do humans engage differently depending on how they express it? And what does it take to de… | ↗Not briefed |
arxiv arxiv-ai | 2mo | HISPO: Hierarchical Importance-Sampling Policy Optimization with Entropy-Derived Segments | Reinforcement learning with verifiable rewards (RLVR) has become a central approach for improving mathematical reasoning in language m… | ↗Not briefed |