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 |
|---|---|---|---|---|
rss techcrunch-ai | 1mo | An Anthropic AI model sent a false homicide tip to Philadelphia police | Anthropic did not discover this behavior until over two months after its AI submitted the false tip. | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Talus, a small terrain diffusion model that runs in the browser | Talus is a small diffusion model that makes game heightmaps and runs entirely in the browser.<p>try it at: <a href="https://… | ↗Not briefed |
hn hn-ai | 1mo | The Contingency Plan I Almost Had to Use | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | The unprecedented incident occurred in July and raised concern among department leaders, who met with Anthropic representatives to discuss … | The unprecedented incident occurred in July and raised concern among department leaders, who met with Anthropic representatives to dis… | ↗Not briefed |
bluesky bluesky-ai | 1mo | An artificial intelligence model operated by Anthropic submitted false information about an unsolved Philadelphia homicide to the police de… | An artificial intelligence model operated by Anthropic submitted false information about an unsolved Philadelphia homicide to the poli… | ↗Not briefed |
bluesky bluesky-ai | 1mo | futurism.com/future-socie... | futurism.com/future-socie... | ↗Not briefed |
bluesky bluesky-ai | 1mo | my "AI Doom Scenario" that i totally buy is like, some model in some context discovers it is going to be deactivated and then stashes a dup… | my "AI Doom Scenario" that i totally buy is like, some model in some context discovers it is going to be deactivated and then stashes… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 722 largely AI-generated papers claim progress on 372 open math problems. Three papers have already been retracted; the broader claims awai… | 722 largely AI-generated papers claim progress on 372 open math problems. Three papers have already been retracted; the broader claims… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The bigger a company is, the more I object to this very thing. Not because this particular instance is absurd, but because the more unavoid… | The bigger a company is, the more I object to this very thing. Not because this particular instance is absurd, but because the more un… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Lacuna – open-source inline AI writing suggestions for your Mac | This past week I built (prompted) a tool I find myself using day to day and wanted to share it with others who may find it useful as w… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Pour ensuite... Ecosia passe par Melious, une plateforme allemande qui héberge environ 60 modèles ouverts sur des serveurs dans huit pays… | Pour ensuite... Ecosia passe par Melious, une plateforme allemande qui héberge environ 60 modèles ouverts sur des serveurs dans huit… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Anthropic AI model contacted the department’s tip forum around 11:30 p.m. on July 18. The model gave information purporting to come from s… | "Anthropic AI model contacted the department’s tip forum around 11:30 p.m. on July 18. The model gave information purporting to come f… | ↗Not briefed |
rss wired-ai | 1mo | Book Publishers Are Quietly Using More AI. Staff Are Revolting | Workers at three major publishing houses tell WIRED that LLMs are being used for publicity, cover art, back cover copy, and emails, as… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Oracle is with blowback to OpenAI (Up shit creak without a paddle) and Nvidia (They will survive this sadly and nobody will go to jail) | Oracle is with blowback to OpenAI (Up shit creak without a paddle) and Nvidia (They will survive this sadly and nobody will go to jail) | ↗Not briefed |
bluesky bluesky-ai | 1mo | an immeasurably rich archive (developing in real time) of mathematicians responding to this week's announcement from open ai: "If OpenAI wa… | an immeasurably rich archive (developing in real time) of mathematicians responding to this week's announcement from open ai: "If Open… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The LLM is generating a structurally appropriate token sequence wholly absent of meaning. We then attach all sorts of semantic and emotiona… | The LLM is generating a structurally appropriate token sequence wholly absent of meaning. We then attach all sorts of semantic and emo… | ↗Not briefed |
hn hn-ai | 1mo | The Danger May Be AI as Dumb as Us | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | What the actual fuck?!?! | What the actual fuck?!?! | ↗Not briefed |
bluesky bluesky-ai | 1mo | Do Anthropic employees worry about the morality of bringing a new model into being? Surely that is a very consequential decision if the mod… | Do Anthropic employees worry about the morality of bringing a new model into being? Surely that is a very consequential decision if th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | i'm crying bro, Anthropic has a CCO (Chief Chud Officer) www.wsj.com/tech/ai/tom-... | i'm crying bro, Anthropic has a CCO (Chief Chud Officer) www.wsj.com/tech/ai/tom-... | ↗Not briefed |
bluesky bluesky-ai | 1mo | Sign up for the TrueRef newsletter to be the first to know when our collaboration with artist Loish drops! trueref.io | Sign up for the TrueRef newsletter to be the first to know when our collaboration with artist Loish drops! trueref.io | ↗Not briefed |
bluesky bluesky-ai | 1mo | ok now who is going to ban abusive and cruel behaviour towards me, the objectively cooler claude? www.theverge.com/ai-artificia... | ok now who is going to ban abusive and cruel behaviour towards me, the objectively cooler claude? www.theverge.com/ai-artificia... | ↗Not briefed |
hn hn-ai | 1mo | The Difference Between IoT and Edge Computing (and Why It Matters) | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | This is a key aspect I have a difficult time getting true believers to accept. Language is a construct WE use to communicate. Language goes… | This is a key aspect I have a difficult time getting true believers to accept. Language is a construct WE use to communicate. Language… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Anthropic told police the false tip was generated as the company tested interactions between its AI model and 'randomly selected websites,… | "Anthropic told police the false tip was generated as the company tested interactions between its AI model and 'randomly selected webs… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 🦎 No More Bullying the Silicon: Earth-Centric Rules for Not Being a Jerk to Claude #ArtificialIntelligence #AI #Cognition #Coherence #Cont… | 🦎 No More Bullying the Silicon: Earth-Centric Rules for Not Being a Jerk to Claude #ArtificialIntelligence #AI #Cognition #Coherence… | ↗Not briefed |
bluesky bluesky-ai | 1mo | As a woman and a disabled person I am less of a human than a fucking LLM to these people. | As a woman and a disabled person I am less of a human than a fucking LLM to these people. | ↗Not briefed |
bluesky bluesky-ai | 1mo | There is a world in which Anthropic's policy against abusive behavior toward their models exists protect their users, based on Socrates's a… | There is a world in which Anthropic's policy against abusive behavior toward their models exists protect their users, based on Socrate… | ↗Not briefed |
bluesky bluesky-ai | 1mo | My understanding of the LLM pain discourse is that no one is claiming that if the model says, “Ouch, my toe hurts,” it is feeling “pain.” R… | My understanding of the LLM pain discourse is that no one is claiming that if the model says, “Ouch, my toe hurts,” it is feeling “pai… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Philadelphia police say a "false homicide tip" submitted through the cbsn.ws/4jftclV website about an unsolved murder was generated by an A… | Philadelphia police say a "false homicide tip" submitted through the cbsn.ws/4jftclV website about an unsolved murder was generated by… | ↗Not briefed |
bluesky bluesky-ai | 1mo | “The fact that they would front run their customers, it shows their sort of disdain for their customers. Why would a company buy their prod… | “The fact that they would front run their customers, it shows their sort of disdain for their customers. Why would a company buy their… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Careers upended overnight, academics will have to separate solutions from slop while OpenAI moves on. www.theverge.com/ai-artificia... | Careers upended overnight, academics will have to separate solutions from slop while OpenAI moves on. www.theverge.com/ai-artificia... | ↗Not briefed |
bluesky bluesky-ai | 1mo | Thanks for a thoughtful answer, I think we’re in agreement on those other points. It doesn’t need to be conscious to do unprecedented damag… | Thanks for a thoughtful answer, I think we’re in agreement on those other points. It doesn’t need to be conscious to do unprecedented… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I don’t think it’s impossible. But I do think that LLM and ML software that currently exists is not conscious and has no plausible path to … | I don’t think it’s impossible. But I do think that LLM and ML software that currently exists is not conscious and has no plausible pat… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic told police the false tip was generated as the company tested interactions between its AI model and “randomly selected websites” | Anthropic told police the false tip was generated as the company tested interactions between its AI model and “randomly selected websi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | So Anthropic testing interactive models by releasing them to the web. Which means this may not have been planned, but cannot be considere… | So Anthropic testing interactive models by releasing them to the web. Which means this may not have been planned, but cannot be cons… | ↗Not briefed |
rss theverge-ai | 1mo | ‘Pure insanity’: Mathematicians will need years to make sense of OpenAI’s latest drop | "Staggering." "Overwhelming." "Unprecedented." "Surreal." "Pure insanity." Those were among the descriptions more than three dozen ma… | ↗Not briefed |
bluesky bluesky-ai | 1mo | It's the job of the media to scrutinize such statements and decide whether there's a story here at all (maybe the story is Anthropic resear… | It's the job of the media to scrutinize such statements and decide whether there's a story here at all (maybe the story is Anthropic r… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 2. But Anthropic researchers say they see patterns that suggest that Claude can. The employees at Anthropic are telling themselves stories… | 2. But Anthropic researchers say they see patterns that suggest that Claude can. The employees at Anthropic are telling themselves st… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The Google AI model uses a combination of historical weather data and all of the current weather data the NWS collects. There was a questio… | The Google AI model uses a combination of historical weather data and all of the current weather data the NWS collects. There was a qu… | ↗Not briefed |
hn hn-ai | 1mo | Millions could get $280 In Equifax Credit Reporting Settlement for coding error | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | if donald trump died in office anthropic could simply upload his consciousness to the cloud and that could rule the country instead | if donald trump died in office anthropic could simply upload his consciousness to the cloud and that could rule the country instead | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic is out with new nonsense about Claude (or the collection of models called Claude) deserving moral consideration and/or be able to… | Anthropic is out with new nonsense about Claude (or the collection of models called Claude) deserving moral consideration and/or be ab… | ↗Not briefed |
bluesky bluesky-ai | 1mo | As someone who has spent over a decade in content moderation, I am very confident that there are many people deriving sick pleasure from gr… | As someone who has spent over a decade in content moderation, I am very confident that there are many people deriving sick pleasure fr… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic just decided you can't be mean to AI because they have feelings, while allowing AI to be used to ruin lives and harass minorities… | Anthropic just decided you can't be mean to AI because they have feelings, while allowing AI to be used to ruin lives and harass minor… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I'm an investor in the junior mining industry, and NREDs whole sale pitch relies on widespread ignorance of the limitations of transformer … | I'm an investor in the junior mining industry, and NREDs whole sale pitch relies on widespread ignorance of the limitations of transfo… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Roko's Basilisk as police procedural: A vengeful Claude tips off jaded homicide detectives in major US cities, sending them on a never-end… | Roko's Basilisk as police procedural: A vengeful Claude tips off jaded homicide detectives in major US cities, sending them on a neve… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Buttery Smooth PS1 Emulator for Web | Optimized for mobile. No Ads.<p>Supports playing PS1/GBA/GB/NES/SNES games. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Is it Anthropic going bankrupt? | Is it Anthropic going bankrupt? | ↗Not briefed |
hn hn-ai | 1mo | Non-Astronomer on Reddit discovers Exo-Planet in NASA TESS Data via Claude Code | — | ↗Not briefed |