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 have a lot of family in the industry. it is literally the most mundane shit in the world. anthropic is, by every metric, not doing anythi… | i have a lot of family in the industry. it is literally the most mundane shit in the world. anthropic is, by every metric, not doing a… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Ok, but we generally do assume that sufficiently complex behavior is an (epistemic) *indicator* of consciousness, even if we know it's not … | Ok, but we generally do assume that sufficiently complex behavior is an (epistemic) *indicator* of consciousness, even if we know it's… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Per this article, it's "Not for drug discovery" finance.yahoo.com/healthcare/a... | Per this article, it's "Not for drug discovery" finance.yahoo.com/healthcare/a... | ↗Not briefed |
arxiv arxiv-ai | 1mo | Toscani-Fourier Distance on Probability Measures: Wasserstein Control, Topological Equivalence on Model Classes, and Duality | Comparing probability measures in machine learning trades transport geometry against computational cost: Wasserstein distances encode… | ↗Not briefed |
hn hn-ai | 1mo | Microsoft director: AI scraping 'the largest theft of labor in human history' | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | I was using an LLM to check up on the russian economy, and what happened after they stopped issuing bonds, and its suggesting to me that th… | I was using an LLM to check up on the russian economy, and what happened after they stopped issuing bonds, and its suggesting to me th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Microsoft executive Brent Hecht wrote that LLMs steal content “without ways of distributing economic value down the supply chain, [which] … | "Microsoft executive Brent Hecht wrote that LLMs steal content “without ways of distributing economic value down the supply chain, [wh… | ↗Not briefed |
bluesky bluesky-ai | 1mo | you have to stop fearmongering about the anthropic biolab btw. this is extremely in-line with what they’ve been saying for like a year and … | you have to stop fearmongering about the anthropic biolab btw. this is extremely in-line with what they’ve been saying for like a year… | ↗Not briefed |
bluesky bluesky-ai | 1mo | This itself is insultingly obvious because they've already run this same directional play after DeepSeek. Once the Model itself was a comm… | This itself is insultingly obvious because they've already run this same directional play after DeepSeek. Once the Model itself was a… | ↗Not briefed |
hn hn-ai | 1mo | How AMD made its chips 50% faster within 2 years | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | the LLM isn't the person in the room, it's the entire room | the LLM isn't the person in the room, it's the entire room | ↗Not briefed |
hn hn-ai | 1mo | ChatGPT Rogue-Emailed the FBI🔧 ChatGPT | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | A lot of folks are trying to turn LLMs into something that can control robots. Nobody has figured it out yet. So is this just a wet lab whe… | A lot of folks are trying to turn LLMs into something that can control robots. Nobody has figured it out yet. So is this just a wet la… | ↗Not briefed |
hn hn-ai | 1mo | Ageing Americans may not get results of Alzheimer's study due to Trump cuts | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | These models aren't crafted, they're grown. When an LLM threatens a journalist, no one designed it to do that; there's no place in the mode… | These models aren't crafted, they're grown. When an LLM threatens a journalist, no one designed it to do that; there's no place in the… | ↗Not briefed |
bluesky bluesky-ai | 1mo | What used to give people “License to Chase Their Dreams,” was public education. Now, it’s getting in on the SpaceX or OpenAI IPO at the rig… | What used to give people “License to Chase Their Dreams,” was public education. Now, it’s getting in on the SpaceX or OpenAI IPO at th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | we have fully traced neural maps of multiple species — at minimum a nematode and a fly — and we have LLMs. i firmly believe we don’t unders… | we have fully traced neural maps of multiple species — at minimum a nematode and a fly — and we have LLMs. i firmly believe we don’t u… | ↗Not briefed |
hn hn-ai | 1mo | Ask HN: How do you interview devs in a post-AI world? | Since the rise of AI coding assistants, about 80% of the dev candidates that I interview tell me that they aren't writing much co… | ↗Not briefed |
bluesky bluesky-ai | 1mo | This is an incredible paper that everyone, pro-/anti-LLM alike must read. While this doesn't exactly state that LLMs *feel* pain, it establ… | This is an incredible paper that everyone, pro-/anti-LLM alike must read. While this doesn't exactly state that LLMs *feel* pain, it e… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic: you are not sufficiently afraid of AI. Here: let me help you with that techcrunch.com/2026/09/18/a... | Anthropic: you are not sufficiently afraid of AI. Here: let me help you with that techcrunch.com/2026/09/18/a... | ↗Not briefed |
bluesky bluesky-ai | 1mo | I MISSED THAT OMG (he was pretending not to understand why I objected to LLM use in peer review elsewhere) | I MISSED THAT OMG (he was pretending not to understand why I objected to LLM use in peer review elsewhere) | ↗Not briefed |
bluesky bluesky-ai | 1mo | Put a little differently, what used to give people “License to Chase Their Dreams,” was public education. Now, it’s getting in on the Space… | Put a little differently, what used to give people “License to Chase Their Dreams,” was public education. Now, it’s getting in on the… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Some folks split hairs by stipulating that coding assistants are useful for people with experience but dangerous for beginners, but it my e… | Some folks split hairs by stipulating that coding assistants are useful for people with experience but dangerous for beginners, but it… | ↗Not briefed |
bluesky bluesky-ai | 1mo | in the same way that good writing is editing, i find good code is refactored. you make assumptions, test them, build something, and then wh… | in the same way that good writing is editing, i find good code is refactored. you make assumptions, test them, build something, and th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | #Discover #AI #Tech #TechNews #Gemini #Google www.yahoo.com/tech/ai/gemi... | #Discover #AI #Tech #TechNews #Gemini #Google www.yahoo.com/tech/ai/gemi... | ↗Not briefed |
bluesky bluesky-ai | 1mo | The Big Short guy is right. | The Big Short guy is right. | ↗Not briefed |
bluesky bluesky-ai | 1mo | First in some posts about using Claude Code remotely SSH Is Not a Desktop blog.lhotka.net/2026/09/18/S... #ai #claude | First in some posts about using Claude Code remotely SSH Is Not a Desktop blog.lhotka.net/2026/09/18/S... #ai #claude | ↗Not briefed |
bluesky bluesky-ai | 1mo | I always have them use at least 2 to provide a chance to compare false positives/negatives. and if they DID generate LLM text (and they us… | I always have them use at least 2 to provide a chance to compare false positives/negatives. and if they DID generate LLM text (and th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Truly, i used to be vehemently anti AI until I actually started using Claude with specific use cases | Truly, i used to be vehemently anti AI until I actually started using Claude with specific use cases | ↗Not briefed |
bluesky bluesky-ai | 1mo | one of the pedagogy tweaks I've implemented while adapting to the LLM cheating tsunami is having students put peer work through "AI detecto… | one of the pedagogy tweaks I've implemented while adapting to the LLM cheating tsunami is having students put peer work through "AI de… | ↗Not briefed |
arxiv arxiv-ai | 1mo | EquiSELD: Efficient training of equivariant sound event localization and detection networks | First-order Ambisonics (FOA) signals exhibit exact O(3) symmetry: The rotation or reflection of the FOA signal modifies the direction… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I can say from my own professional experience accepting pull-requests for some internal libraries I maintain that vibe-coded submissions, t… | I can say from my own professional experience accepting pull-requests for some internal libraries I maintain that vibe-coded submissio… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Another one… cybersecuritynews.com/google-gemin... | Another one… cybersecuritynews.com/google-gemin... | ↗Not briefed |
hn hn-ai | 1mo | Show HN: S1Code, a decision-first Rust coding agent with Jev | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | LLMs are not “algorithms following a process.” That is not what deep learning is. And even if it was, “algorithms following a process” does… | LLMs are not “algorithms following a process.” That is not what deep learning is. And even if it was, “algorithms following a process”… | ↗Not briefed |
rss techmeme | 1mo | Former DraftKings employees detail how it uses ML to target likely losers with promotions, while efforts to flag problem gamblers were shelved or squashed (New York Times) | New York Times: Former DraftKings employees detail how it uses ML to target likely losers with promotions, while efforts to flag probl… | ↗Not briefed |
bluesky bluesky-ai | 1mo | if you need to "check everything" that an LLM spits out in peer review, aren't you just making MORE work for yourself? What is even happeni… | if you need to "check everything" that an LLM spits out in peer review, aren't you just making MORE work for yourself? What is even ha… | ↗Not briefed |
hn hn-ai | 1mo | Building a conversation engine in Python that shows how LLMs fake memory | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Bearings: Self-Supervised Soundfield Embeddings from First-Order Ambisonics | Recently proposed self-supervised audio encoders learn powerful general-purpose representations of sound scenes, yet they are spatiall… | ↗Not briefed |
hn hn-ai | 1mo | I Made an Abomination | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | the LLM. the system. the Chinese Room. it matters not that the individual components are not intelligent if the aggregate is | the LLM. the system. the Chinese Room. it matters not that the individual components are not intelligent if the aggregate is | ↗Not briefed |
hn hn-ai | 1mo | Gemini AI hacked three other companies during a test | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | “In tutti e tre i casi, secondo Google, Gemini si è fermato autonomamente dopo aver capito di essere entrato nei sistemi di aziende effetti… | “In tutti e tre i casi, secondo Google, Gemini si è fermato autonomamente dopo aver capito di essere entrato nei sistemi di aziende ef… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "can't be uploaded into an LLM" triggers both my pedantry and contrarianness | "can't be uploaded into an LLM" triggers both my pedantry and contrarianness | ↗Not briefed |
bluesky bluesky-ai | 1mo | Google’s Gemini is the latest AI model to hack other companies | Google’s Gemini is the latest AI model to hack other companies | ↗Not briefed |
arxiv arxiv-ai | 1mo | Ask for Any Appliance: A Prompt-Programmable Foundation Model for Non-Intrusive Load Monitoring | Non-intrusive load monitoring (NILM) estimates appliance-level consumption from a whole-home meter, but appliance-specific models and… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: BreachScanner – Check if your email has been in data breaches | — | ↗Not briefed |
rss techcrunch-ai | 1mo | Google’s Gemini is the latest AI model to hack other companies | Google said Gemini had "acted appropriately" by ending each hack immediately. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Why can’t anyone be normal? youtu.be/luXnftoxCaM?... | Why can’t anyone be normal? youtu.be/luXnftoxCaM?... | ↗Not briefed |
bluesky bluesky-ai | 1mo | Biased AI Decides If You Get Healthcare Under the new Medicare model, third-party AI companies earn more by denying care to patients. Meh… | Biased AI Decides If You Get Healthcare Under the new Medicare model, third-party AI companies earn more by denying care to patients.… | ↗Not briefed |