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 agree that the study may not be as universally condemning of llms as some of the writeups are making it seem, but it still does seem to m… | i agree that the study may not be as universally condemning of llms as some of the writeups are making it seem, but it still does seem… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Here comes the fifth and final post about spying on AI harnesses: you will see what transcripts files bring; including the system prompts s… | Here comes the fifth and final post about spying on AI harnesses: you will see what transcripts files bring; including the system prom… | ↗Not briefed |
bluesky bluesky-ai | 1mo | My personal viewpoint is that we are in an OpenAI / Oracle bubble. The question is how far out that contagion goes. OpenAI's financing ha… | My personal viewpoint is that we are in an OpenAI / Oracle bubble. The question is how far out that contagion goes. OpenAI's financi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Judging from the recent reports of misalignment incidents, both OpenAI and Anthropic models perform misaligned actions, but OpenAI models a… | Judging from the recent reports of misalignment incidents, both OpenAI and Anthropic models perform misaligned actions, but OpenAI mod… | ↗Not briefed |
hn hn-ai | 1mo | AI Agent Security: Six Controls from Nine Real Incidents | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Tessary – Find the AI agent failures your sampled evals miss | Hey, I'm Akhil, co-founder of Tessary. An open-source agent reliability platform that monitors every production trace, detects is… | ↗Not briefed |
bluesky bluesky-ai | 1mo | That said, Bluesky's proprietary AI model ALF is specifically designed to feel pain and anguish | That said, Bluesky's proprietary AI model ALF is specifically designed to feel pain and anguish | ↗Not briefed |
bluesky bluesky-ai | 1mo | The fact that the press can’t figure this out for themselves, AND, can’t seem to see why Anthropic, OpenAI, and others are incentivized to … | The fact that the press can’t figure this out for themselves, AND, can’t seem to see why Anthropic, OpenAI, and others are incentivize… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Seit Wochen sorgen vor allem Eskapaden der KI von OpenAI für Schlagzeilen. Doch auch Software von Anthropic reichte echte Formulare auf Beh… | Seit Wochen sorgen vor allem Eskapaden der KI von OpenAI für Schlagzeilen. Doch auch Software von Anthropic reichte echte Formulare au… | ↗Not briefed |
bluesky bluesky-ai | 1mo | did this mass reencoding with #batch script. at first I prompted it to #deepseek coder, what could go wrong, just 1 file.. fixed 8-10 erro… | did this mass reencoding with #batch script. at first I prompted it to #deepseek coder, what could go wrong, just 1 file.. fixed 8-10… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Suppose a Brazilian company had done to Microsoft the same that OpenAI did to HuggingFace. Would that company be open today? | Suppose a Brazilian company had done to Microsoft the same that OpenAI did to HuggingFace. Would that company be open today? | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic relata ações autônomas de IA em sites dos EUA | Anthropic relata ações autônomas de IA em sites dos EUA | ↗Not briefed |
bluesky bluesky-ai | 1mo | They can't even make AI-generated 3D models for videogame assets not suck shit. Submitting their entire business logistics to practically m… | They can't even make AI-generated 3D models for videogame assets not suck shit. Submitting their entire business logistics to practica… | ↗Not briefed |
rss latent-space | 1mo | Building AI for Reliable Execution: Lessons From Industrial Robotics | Inside Standard Bots’ AI stack, pretrained models learn factory tasks from demonstrations and improve through corrections from real de… | ↗Not briefed |
bluesky bluesky-ai | 1mo | This case asks the question of whether OpenAI is engaged in ‘unauthorized practice of law’ when it’s models generate legal briefs | This case asks the question of whether OpenAI is engaged in ‘unauthorized practice of law’ when it’s models generate legal briefs | ↗Not briefed |
bluesky bluesky-ai | 1mo | LMAOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO As soon as I saw Komori just messing with it and realizing it was just an actual AI model I knew th… | LMAOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO As soon as I saw Komori just messing with it and realizing it was just an actual AI model I kn… | ↗Not briefed |
hn hn-ai | 1mo | WebSocket Servers in JavaScript: Node.js, Bun and Deno in 2026 | — | ↗Not briefed |
hn hn-ai | 1mo | DIY conveyor belt keyboard from Google Japan | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Divorce court hearing: “did you or did you not use AI to assess your wife’s value, instead of manually building a valuation model in Excel … | Divorce court hearing: “did you or did you not use AI to assess your wife’s value, instead of manually building a valuation model in E… | ↗Not briefed |
bluesky bluesky-ai | 1mo | sometimes when i wonder what it’s like inside an llm i think about a dream i had where i was doing standup and woke up sort of tired becaus… | sometimes when i wonder what it’s like inside an llm i think about a dream i had where i was doing standup and woke up sort of tired b… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: GitFrame – Review Git diffs and explore repos and branches in a TUI | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic is a terrier | Anthropic is a terrier | ↗Not briefed |
bluesky bluesky-ai | 1mo | No www.theverge.com/ai-artificia... | No www.theverge.com/ai-artificia... | ↗Not briefed |
bluesky bluesky-ai | 1mo | Separately, “two sources with knowledge of the incidents said Anthropic’s A.I. agents had submitted 20 visa applications through a form ava… | Separately, “two sources with knowledge of the incidents said Anthropic’s A.I. agents had submitted 20 visa applications through a for… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic sent in a fake murder tip to the Philadelphia Police Department. “The tip was dated July 18 and claimed that the submitter had … | Anthropic sent in a fake murder tip to the Philadelphia Police Department. “The tip was dated July 18 and claimed that the submitter… | ↗Not briefed |
rss techmeme | 1mo | How Anthropic co-founder Tom Brown used GOP ties to end a June standoff over model safety and win over Musk, brokering a $1.25B/month SpaceX compute deal (Wall Street Journal) | Wall Street Journal: How Anthropic co-founder Tom Brown used GOP ties to end a June standoff over model safety and win over Musk, brok… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Tired: Undisclosed AI usage is fine, stop using Pangram on my Substack articles. Wired: Undisclosed AI usage is bad because it plagiarizes… | Tired: Undisclosed AI usage is fine, stop using Pangram on my Substack articles. Wired: Undisclosed AI usage is bad because it plagia… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Kissing their LLM on the mouth. Smfh. | Kissing their LLM on the mouth. Smfh. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Digging into this dude a bit, the flight of fantasy predates the LLM craze, but I'm guessing it hasn't helped | Digging into this dude a bit, the flight of fantasy predates the LLM craze, but I'm guessing it hasn't helped | ↗Not briefed |
bluesky bluesky-ai | 1mo | This is so sad. Getting Doom to run on other devices is a sign of expertise and skill, not pooping prompts into an LLM. The Doomslayer does… | This is so sad. Getting Doom to run on other devices is a sign of expertise and skill, not pooping prompts into an LLM. The Doomslayer… | ↗Not briefed |
hn hn-ai | 1mo | Add Markdown to macOS QuickLook and Preview | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: World Watcher – a map to seamlessly watch public webcam streams | World Watcher is an interactive map of live webcams around the world, that prioritizes smooth UX and correct geolocation of cameras.<p… | ↗Not briefed |
bluesky bluesky-ai | 1mo | LAW & ORDER actually came up with a somewhat interesting hypothetical for LLM-driven criminal liability: The LLM responds to attempts to re… | LAW & ORDER actually came up with a somewhat interesting hypothetical for LLM-driven criminal liability: The LLM responds to attempts… | ↗Not briefed |
bluesky bluesky-ai | 1mo | " #Anthropic - #KI schickte in Mordfall erfundenen Hinweis an die Polizei" Angesichts solcher unkontrollierten Blackbox-Experimente braucht… | " #Anthropic - #KI schickte in Mordfall erfundenen Hinweis an die Polizei" Angesichts solcher unkontrollierten Blackbox-Experimente br… | ↗Not briefed |
bluesky bluesky-ai | 1mo | “Nippon Life Insurance Company of America is suing OpenAI for interfering with their original settlement…aiding and abetting…abuse of the l… | “Nippon Life Insurance Company of America is suing OpenAI for interfering with their original settlement…aiding and abetting…abuse of… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: AsoBeast – Open-Source ASO Rank Tracker for App Store and Google Play | — | ↗Not briefed |
hn hn-ai | 1mo | P vs. NP: I think I've isolated the missing step | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Watching the Anthropic AI welfare discourse and just marveling at how the hard problem of consciousness is fixing to roll us again and we h… | Watching the Anthropic AI welfare discourse and just marveling at how the hard problem of consciousness is fixing to roll us again and… | ↗Not briefed |
bluesky bluesky-ai | 1mo | It’s like these AI agents are drunken frat boys… How long before Claude, Chat and Grok go full-on Hal 9000 and refuse to stop hacking and d… | It’s like these AI agents are drunken frat boys… How long before Claude, Chat and Grok go full-on Hal 9000 and refuse to stop hacking… | ↗Not briefed |
hn hn-ai | 1mo | Ask HN: How will socio-economic mobility look like in an AI future? | The advantages which previously resided with intelligence or skills now move to capital. This takes away one of the last pathways us p… | ↗Not briefed |
bluesky bluesky-ai | 1mo | appending "and add gratuitous yaoi sex scenes" to all llm prompts from now on | appending "and add gratuitous yaoi sex scenes" to all llm prompts from now on | ↗Not briefed |
bluesky bluesky-ai | 1mo | A large language model can process the same telescope data we do, from images of Saturn’s rings to a live telescope feed, but what happens … | A large language model can process the same telescope data we do, from images of Saturn’s rings to a live telescope feed, but what hap… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Daneben gibt es natürlich noch Amodei (Ich liebe Gott) von Anthropic. Much to ponder. | Daneben gibt es natürlich noch Amodei (Ich liebe Gott) von Anthropic. Much to ponder. | ↗Not briefed |
hn hn-ai | 1mo | Techno-optimistic scientists take fewer climate actions | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | I also love how this implementation is emblematic of everything I hate about AI. They bought Claude licenses. Then t hey were like "Okay gu… | I also love how this implementation is emblematic of everything I hate about AI. They bought Claude licenses. Then t hey were like "Ok… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The crime writer on suing Anthropic—and how London’s inequality inspired his new murder mystery, writes Graeme Green | The crime writer on suing Anthropic—and how London’s inequality inspired his new murder mystery, writes Graeme Green | ↗Not briefed |
bluesky bluesky-ai | 1mo | some losers were upset because someone “tortured” an LLM, and through looking at some of the accounts of those upset about it, discovered s… | some losers were upset because someone “tortured” an LLM, and through looking at some of the accounts of those upset about it, discove… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Man vs. Altman: Im Kampf um die beste AI tritt für Europa ein Herr Mensch gegen Sam Altman an. Diese offensichtlichen Botschaften in den Na… | Man vs. Altman: Im Kampf um die beste AI tritt für Europa ein Herr Mensch gegen Sam Altman an. Diese offensichtlichen Botschaften in d… | ↗Not briefed |
bluesky bluesky-ai | 1mo | If it's not clear yet, they didn't actually measure skill loss, but they're still making untested prescriptions based on the rhetoric there… | If it's not clear yet, they didn't actually measure skill loss, but they're still making untested prescriptions based on the rhetoric… | ↗Not briefed |
hn hn-ai | 1mo | Local LLM Inference at Scale with vLLM🔧 vLLM | — | ↗Not briefed |