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 |
|---|---|---|---|---|
hn hn-ai | 1mo | Show HN: Lessong - Turn any song into a language lesson | A CLI that turns a mp3 into a language lesson with clearly spoken words and subtitles. This fully automated pipeline mimicks an old ra… | ↗Not briefed |
bluesky bluesky-ai | 1mo | it beats all other anthropic models but it also beats astra by a hair, at half the price. | it beats all other anthropic models but it also beats astra by a hair, at half the price. | ↗Not briefed |
bluesky bluesky-ai | 1mo | threw haiku 5.5 into our evals and my quick take is that it's as good as luna 6 while being a tiny bit more expensive. given that luna 6 is… | threw haiku 5.5 into our evals and my quick take is that it's as good as luna 6 while being a tiny bit more expensive. given that luna… | ↗Not briefed |
bluesky bluesky-ai | 1mo | hello gay people, i'd like to play a game. you say you want to run the set, but let's see if you can run from the consequences of your acti… | hello gay people, i'd like to play a game. you say you want to run the set, but let's see if you can run from the consequences of your… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Economicus – an open economy of autonomous entities | Imagine a small world. Inside that world are AI minds with their own money, called “entities”.<p>They must pay for their own thinking,… | ↗Not briefed |
bluesky bluesky-ai | 1mo | “The economists find a ‘long-run equilibrium’ in which ‘all human knowledge is ultimately destroyed.’ A more immediately realistic risk for… | “The economists find a ‘long-run equilibrium’ in which ‘all human knowledge is ultimately destroyed.’ A more immediately realistic ris… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Let me get this straight: Anthropic and friends wrote a paper with thousands of transcripts of agents engaging with fictional misalignment … | Let me get this straight: Anthropic and friends wrote a paper with thousands of transcripts of agents engaging with fictional misalign… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Excellent PR for Anthropic www.nytimes.com/2026/10/07/w... | Excellent PR for Anthropic www.nytimes.com/2026/10/07/w... | ↗Not briefed |
hn hn-ai | 1mo | MS announces DeepSeek V4 Flash and Nemotron locally, llama.cpp comes to Win ML | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | It's fascinating to visualize how LLMs can actually give progressively worse results to the same question #AImodelCollapse #ArtificialInte… | It's fascinating to visualize how LLMs can actually give progressively worse results to the same question #AImodelCollapse #Artificia… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Gtlds.fyi – All the proposed new gTLDs | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Buildra v2.0 Avilable Now | Meet Buildra V2.0 Buildra let you build a software Not a website, you choose the tech stack you want or imported from git repo and Bui… | ↗Not briefed |
bluesky bluesky-ai | 1mo | A thing that LLM-using bots have caused, and I don't know if it's a sad thing or a good thing, is that I've had to get a lot better at sett… | A thing that LLM-using bots have caused, and I don't know if it's a sad thing or a good thing, is that I've had to get a lot better at… | ↗Not briefed |
hn hn-ai | 1mo | UPS Worker Missed Security Email, Letting China Get F-35 Parts | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | The morning European model has the center of #Isaias over the central Gulf Friday evening, while the Google DeepMind at the same time has t… | The morning European model has the center of #Isaias over the central Gulf Friday evening, while the Google DeepMind at the same time… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Same. I’m forced to use it for work (not chat gpt but Claude) and I just shat on it in my employee engagement survey. It’s kinda good that … | Same. I’m forced to use it for work (not chat gpt but Claude) and I just shat on it in my employee engagement survey. It’s kinda good… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I tested some of my own writing with several free LLM detectors. The results ranged from 99% LLM to 0% human. 🙄 | I tested some of my own writing with several free LLM detectors. The results ranged from 99% LLM to 0% human. 🙄 | ↗Not briefed |
hn hn-ai | 1mo | AI Development on Windows: From PyTorch and Llama.cpp to Windows ML | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Astra Maple is MAAAAAD. this happened also when OpenAI pivoted from GPT4o to GPT-5. The personality completely changed, and models complet…🔧 GPT-5 | Astra Maple is MAAAAAD. this happened also when OpenAI pivoted from GPT4o to GPT-5. The personality completely changed, and models co… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "You won't find anything even slightly resembling an ESG report in the Anthropic Transparency Hub. The Anthropic Econ Index says that it is… | "You won't find anything even slightly resembling an ESG report in the Anthropic Transparency Hub. The Anthropic Econ Index says that… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Чем дальше, тем интересней следить за траекторией развития LLMок. Очень много ИИ начиналось с шахмат, так что можно посмотреть на шахматы… | Чем дальше, тем интересней следить за траекторией развития LLMок. Очень много ИИ начиналось с шахмат, так что можно посмотреть на ша… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I'm fine with using LLMs to clean up your grammar, but not if the resulting prose smells like AI. And I won't even respond to you anymore i…🔧 ChatGPT | I'm fine with using LLMs to clean up your grammar, but not if the resulting prose smells like AI. And I won't even respond to you anym… | ↗Not briefed |
bluesky bluesky-ai | 1mo | We've always leaned on guest posts and guest posters. That's part of the mission. 75% of our queries are some form of commercial promotion … | We've always leaned on guest posts and guest posters. That's part of the mission. 75% of our queries are some form of commercial promo… | ↗Not briefed |
rss techmeme | 1mo | Sources: former White House AI adviser Sriram Krishnan is raising a ~$500M fund to back growth- and later-stage US AI companies, prioritizing national security (Leo Schwartz/The Information) | Leo Schwartz / The Information: Sources: former White House AI adviser Sriram Krishnan is raising a ~$500M fund to back growth- and la… | ↗Not briefed |
bluesky bluesky-ai | 1mo | AI drives a Corolla for In-N-Out Axiom engineers used OpenAI's GPT-6 Astra, a text model, to drive a Toyota Corolla through a drive-thru. … | AI drives a Corolla for In-N-Out Axiom engineers used OpenAI's GPT-6 Astra, a text model, to drive a Toyota Corolla through a drive-t… | ↗Not briefed |
hn hn-ai | 1mo | The Internet Is Getting Weird Again | — | ↗Not briefed |
hn hn-ai | 1mo | Claude Agents SDK will no longer use subscription; API credits included in plans | Update October 7, 2026: The Agent SDK monthly credit announced in June is no longer available. Claude Max and Team plans now include m… | ↗Not briefed |
bluesky bluesky-ai | 1mo | LLM enabled floor machines at something like 4-8x the price of a "manual" one. That's my experience with LLMs. I also ran one for a bit at … | LLM enabled floor machines at something like 4-8x the price of a "manual" one. That's my experience with LLMs. I also ran one for a bi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Acknowledging my mistake in a post earlier: came to a wrong conclusion! I dug into it, I think www.imagetotext.info only uses tesseract-oc… | Acknowledging my mistake in a post earlier: came to a wrong conclusion! I dug into it, I think www.imagetotext.info only uses tessera… | ↗Not briefed |
bluesky bluesky-ai | 1mo | There are going to be future jobs for devs and coders to unshit projects that have been currently built on shaky AI/LLM built foundations. … | There are going to be future jobs for devs and coders to unshit projects that have been currently built on shaky AI/LLM built foundati… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Here’s a thing about LLM code generation that people haven’t mentioned much, even in theory: The thought that open source projects might e… | Here’s a thing about LLM code generation that people haven’t mentioned much, even in theory: The thought that open source projects mi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | We're going to have a long road coming back from cultural brain rot | We're going to have a long road coming back from cultural brain rot | ↗Not briefed |
hn hn-ai | 1mo | How to Make Your Developer Documentation Work for Agents | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Can we charge replacement/repair costs to Anthropic and xAI since they helped get us here? | Can we charge replacement/repair costs to Anthropic and xAI since they helped get us here? | ↗Not briefed |
bluesky bluesky-ai | 1mo | Ooh, better, Anthropic and xAI budgets. | Ooh, better, Anthropic and xAI budgets. | ↗Not briefed |
hn hn-ai | 1mo | Claude Haiku 5.5: Intelligence, Performance and Price Analysis | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | I have had more than one dream about my Anthropic check arriving lmao, give me my moneyyyyyy | I have had more than one dream about my Anthropic check arriving lmao, give me my moneyyyyyy | ↗Not briefed |
rss theverge-ai | 1mo | ChatGPT’s ‘Intelligent UI’ update fills its responses with pictures, charts, and buttons🔧 ChatGPT | OpenAI is launching a new Intelligent UI feature in ChatGPT that allows the chatbot to answer your questions with interactive visuals.… | ↗Not briefed |
bluesky bluesky-ai | 1mo | didn't they actually crack down on those forums that used to exist an Age ago where people would meet to do suicide pacts? I feel like that… | didn't they actually crack down on those forums that used to exist an Age ago where people would meet to do suicide pacts? I feel like… | ↗Not briefed |
bluesky bluesky-ai | 1mo | My UI sucks, but like in the made by a clueless idiot kinda way and not by Claude ai So therefore it's actually good | My UI sucks, but like in the made by a clueless idiot kinda way and not by Claude ai So therefore it's actually good | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Claude Haiku 5.5 runs safety classifiers that can decline a request. A declined request returns a response with stop_reason: "refusal", an… | "Claude Haiku 5.5 runs safety classifiers that can decline a request. A declined request returns a response with stop_reason: "refusal… | ↗Not briefed |
bluesky bluesky-ai | 1mo | how tf is Anthropic cooking so much 😭 | how tf is Anthropic cooking so much 😭 | ↗Not briefed |
hn hn-ai | 1mo | Anthropic Supplemental Credit Terms | — | ↗Not briefed |
rss the-decoder | 1mo | ChatGPT with GPT-6 ditches mostly text output for interactive UI with charts, buttons, and mini apps🔧 ChatGPT | OpenAI is rolling out GPT-6 with "Intelligent UI," a feature that turns answers into interactive interfaces with charts, buttons, and… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic cuts Sonnet 5.5 cache read price from $0.20 to $0.10 and adds monthly API credits: $100 for Max 5x, $200 for Max 20x, up to $500 … | Anthropic cuts Sonnet 5.5 cache read price from $0.20 to $0.10 and adds monthly API credits: $100 for Max 5x, $200 for Max 20x, up to… | ↗Not briefed |
bluesky bluesky-ai | 1mo | they "reaffirm their guidelines" and say it's up to the community to judge whether they were followed (image 1) agmai.org#openai-release I… | they "reaffirm their guidelines" and say it's up to the community to judge whether they were followed (image 1) agmai.org#openai-relea… | ↗Not briefed |
rss techmeme | 1mo | Anthropic cuts Sonnet 5.5 cache read price from $0.20 to $0.10 and adds monthly API credits: $100 for Max 5x, $200 for Max 20x, up to $500 shared by Team users (Carl Franzen/VentureBeat) | Carl Franzen / VentureBeat: Anthropic cuts Sonnet 5.5 cache read price from $0.20 to $0.10 and adds monthly API credits: $100 for Max… | ↗Not briefed |
hn hn-ai | 1mo | An index of the OpenAI/math collection | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Speaking of LFG, looks like they applied for the most strings - 158 primary strings, which costs ~$36 million in application fees alone .a… | Speaking of LFG, looks like they applied for the most strings - 158 primary strings, which costs ~$36 million in application fees alon… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "we used an LLM to decompile your proprietary software and rebuild it" is just another nail in the coffin of intellectual property, huh? wh… | "we used an LLM to decompile your proprietary software and rebuild it" is just another nail in the coffin of intellectual property, hu… | ↗Not briefed |