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 | OpenAI scraps release of new model over safety concerns in internal testing | OpenAI scraps release of new model over safety concerns in internal testing | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic’s newly public IPO filing provides a detailed look at the San Francisco AI company’s extraordinary growth and plans for the futur… | Anthropic’s newly public IPO filing provides a detailed look at the San Francisco AI company’s extraordinary growth and plans for the… | ↗Not briefed |
rss bing-news-ai | 1mo | ChatGPT can’t hack your future: Why grit and emotional intelligence still win in the AI era🔧 ChatGPT | AI is already reshaping the college admissions process. According to Rim, one of AI's most immediate effects on college admissions is… | ↗Not briefed |
bluesky bluesky-ai | 1mo | @gruber.foo points to this Zuckerberg profile. At the core is a fawning overview of Meta's AI efforts and the idea that the tech is likely … | @gruber.foo points to this Zuckerberg profile. At the core is a fawning overview of Meta's AI efforts and the idea that the tech is li… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAIが「GPT-6.1 Astra」の一般公開を中止、安全基準を満たさないため https://gigazine.net/news/20260929-openai-abandons-plan-to-release-gpt-6-1-astra/ | OpenAIが「GPT-6.1 Astra」の一般公開を中止、安全基準を満たさないため https://gigazine.net/news/20260929-openai-abandons-plan-to-release-gpt-6-1-astra/ | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Caffold – The same agent workspace on desktop, foldable, tablet, phone | Hello, Caffold is an open-source, self-hosted workspace for coding agents. You can continue the same work across desktop, foldable, ta… | ↗Not briefed |
bluesky bluesky-ai | 1mo | weird that i've had to like, draft a businessy stock reply to all of this garbage. but that's the world we live in now, thanks openai | weird that i've had to like, draft a businessy stock reply to all of this garbage. but that's the world we live in now, thanks openai | ↗Not briefed |
bluesky bluesky-ai | 1mo | 米フロリダ州の司法長官が、OpenAI社に対して、安全性が確認されるまで、生成AIの開発を中断すること、未成年者の利用の禁止を求める差し止め請求を州裁判所に申請したそうですね。 安全対策ガードレールの設置や、生成AIが人間であるかのように装う仕組みの禁止なども求めているそうです… | 米フロリダ州の司法長官が、OpenAI社に対して、安全性が確認されるまで、生成AIの開発を中断すること、未成年者の利用の禁止を求める差し止め請求を州裁判所に申請したそうですね。 安全対策ガードレールの設置や、生成AIが人間であるかのように装う仕組みの禁止なども求めてい… | ↗Not briefed |
hn hn-ai | 1mo | OpenAI scraps release of new AI model over safety concerns | — | ↗Not briefed |
hn hn-ai | 1mo | OpenAI apologises for Medicare hack and reveals extent of attack | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Apprise v2.0 – Self-hosted notifications across 160 services | — | ↗Not briefed |
rss techmeme | 1mo | Q&A with Instinct founder Noah Shinn on Instinct's business model, privacy, safety and security, securing compute, Big Tech competition, and more (Patrick O'Shaughnessy/Colossus) | Patrick O'Shaughnessy / Colossus: Q&A with Instinct founder Noah Shinn on Instinct's business model, privacy, safety and security, sec… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Someone just showed me today an entirely LLM-generated email ostensibly from my work email account, but it ended in .br. So please check an… | Someone just showed me today an entirely LLM-generated email ostensibly from my work email account, but it ended in .br. So please che… | ↗Not briefed |
hn hn-ai | 1mo | Run Decision Models on vLLM and Red Hat AI Using DiffusionGemma🔧 vLLM | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | “We’re losing billions and we’re making a product that’s an existential threat to humanity” is a hell of sales pitch www.ft.com/content/c76… | “We’re losing billions and we’re making a product that’s an existential threat to humanity” is a hell of sales pitch www.ft.com/conten… | ↗Not briefed |
bluesky bluesky-ai | 1mo | One has to wonder if the selective leaks are in return for coverage that hypes the IPO. Reuters reports, "Anthropic is making a massive be… | One has to wonder if the selective leaks are in return for coverage that hypes the IPO. Reuters reports, "Anthropic is making a massi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | like, inference is really quite cheap so with anthropic charging businesses as much for tokens as they do i imagine it helps right the ship… | like, inference is really quite cheap so with anthropic charging businesses as much for tokens as they do i imagine it helps right the… | ↗Not briefed |
hn hn-ai | 1mo | Ask HN: Is strong anti-AI sentiment psyops | Is there any work being done to find out whether or not the strong anti-AI sentiment in U.S. workforce isn’t psyops?<p>There hasn’t be… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Never has anyone ever wanted something to fail as much as Ed Zitron has wanted Anthropic to fail. Churchill had a mild preference by compar… | Never has anyone ever wanted something to fail as much as Ed Zitron has wanted Anthropic to fail. Churchill had a mild preference by c… | ↗Not briefed |
bluesky bluesky-ai | 1mo | You would have a problem getting the first Jewish LLM because you can’t put the whole thing in a mikveh. | You would have a problem getting the first Jewish LLM because you can’t put the whole thing in a mikveh. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Like it's hard for me to get too upset with the AI skeptics when Anthropic is releasing nonsense like this in total earnestness bsky.app/p… | Like it's hard for me to get too upset with the AI skeptics when Anthropic is releasing nonsense like this in total earnestness bsky.… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Hugging Face saw a flurry of deal interest before Nvidia agreed to buy the open-source platform for roughly $13 billion, CNBC has learned.🔧 Hugging Face | Hugging Face saw a flurry of deal interest before Nvidia agreed to buy the open-source platform for roughly $13 billion, CNBC has lear… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI holds off on releasing new model over safety concerns, saying it "didn't quite meet the bar" | OpenAI holds off on releasing new model over safety concerns, saying it "didn't quite meet the bar" | ↗Not briefed |
bluesky bluesky-ai | 1mo | nothing really new or interesting in here, sadly. all the numbers are for 2025 and are dwarfed by everything that has happened since | nothing really new or interesting in here, sadly. all the numbers are for 2025 and are dwarfed by everything that has happened since | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI scraps release of new AI model after it showed 'higher levels of deception' | OpenAI scraps release of new AI model after it showed 'higher levels of deception' | ↗Not briefed |
rss latent-space | 1mo | [AINews] AMD buys World Labs for $8.2B, as Atlas solves sparse reconstruction problem for robotics, design and more | Congrats team! | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Rename.Tools – Open-source bulk file renaming in the browser | Hi HN, I built Rename.Tools after needing to rename a large batch of images. This is something I only do occasionally, so I couldn’t j… | ↗Not briefed |
bluesky bluesky-ai | 1mo | here's Anthropic this month talking about disrupting ShinyHunters' use of Claude specifically (much more detail in the full thing than this… | here's Anthropic this month talking about disrupting ShinyHunters' use of Claude specifically (much more detail in the full thing than… | ↗Not briefed |
rss latent-space | 1mo | [AINews] Opus 5.5 is good at explainer videos | a rare feature of a capability | ↗Not briefed |
bluesky bluesky-ai | 1mo | Seems like a “my bad” isn’t really sufficient here. | Seems like a “my bad” isn’t really sufficient here. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Another day working in AI Safety at a frontier lab. Everyone asks me if they can use the model without it hacking into the UN. Buddy, they … | Another day working in AI Safety at a frontier lab. Everyone asks me if they can use the model without it hacking into the UN. Buddy,… | ↗Not briefed |
bluesky bluesky-ai | 1mo | an OpenAi blog post pretending their tech can “interpret and analyse” information lmao masquerading as mea culpa: openai.com/index/how-we... | an OpenAi blog post pretending their tech can “interpret and analyse” information lmao masquerading as mea culpa: openai.com/index/how… | ↗Not briefed |
bluesky bluesky-ai | 1mo | From a technical standpoint, unless you have a zero-day retention policy, a BAA/DPA, or another agreement that precludes your data from bei… | From a technical standpoint, unless you have a zero-day retention policy, a BAA/DPA, or another agreement that precludes your data fro… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Yes, yes we are pushing back. | Yes, yes we are pushing back. | ↗Not briefed |
bluesky bluesky-ai | 1mo | it's fine to report gross revenue here since Anthropic is the principal. they do include the distribution share as a separate expense line … | it's fine to report gross revenue here since Anthropic is the principal. they do include the distribution share as a separate expense… | ↗Not briefed |
bluesky bluesky-ai | 1mo | but there is no use case for publicly available data enshittified by OpenAi tech. | but there is no use case for publicly available data enshittified by OpenAi tech. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Open AI and Anthropic making big public shows of not releasing products because they’re ’too dangerous’ is the most ridiculous thing I’ve e… | Open AI and Anthropic making big public shows of not releasing products because they’re ’too dangerous’ is the most ridiculous thing I… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Lost $42 billion last year but worth $2 trillion in their IPO. OK sure! Anthropic's IPO prospectus shows sweeping AI vision, surging cos… | Lost $42 billion last year but worth $2 trillion in their IPO. OK sure! Anthropic's IPO prospectus shows sweeping AI vision, surgin… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I think AI can be good and useful, and can be profitable. My long term thinking is we're still gonna end up with a lot of good enough disti… | I think AI can be good and useful, and can be profitable. My long term thinking is we're still gonna end up with a lot of good enough… | ↗Not briefed |
bluesky bluesky-ai | 1mo | ChatGPT-maker OpenAI canceled the planned launch of a new artificial intelligence model after it was found to take actions beyond the instr…🔧 ChatGPT | ChatGPT-maker OpenAI canceled the planned launch of a new artificial intelligence model after it was found to take actions beyond the… | ↗Not briefed |
bluesky bluesky-ai | 1mo | seriously, SpaceX showed the way and Anthropic is following it. Selective leaks of the "confidential" prospectus to gin up enthusiasm for t… | seriously, SpaceX showed the way and Anthropic is following it. Selective leaks of the "confidential" prospectus to gin up enthusiasm… | ↗Not briefed |
hn hn-ai | 1mo | Every 4-piece checkers endgame, solved (6.4M positions, 1.8MB) url | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | "During the testing phase for the new model, it showed high levels of what the company saw as deception, or a willingness to mislead users … | "During the testing phase for the new model, it showed high levels of what the company saw as deception, or a willingness to mislead u… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Claude was blocking every single request I made. Took me a few days to realize it was because I had instructions asking it to explain its t… | Claude was blocking every single request I made. Took me a few days to realize it was because I had instructions asking it to explain… | ↗Not briefed |
hn hn-ai | 1mo | Einsum Trees: An Abstraction for Optimizing the Execution of Tensor Expressions [video] | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | The hard part here is deciding on how to do the fissioning. You want to maintain conversation flow and history, but that's hard to do autom… | The hard part here is deciding on how to do the fissioning. You want to maintain conversation flow and history, but that's hard to do… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Does anthropic place a penalty on subagents?! maybe as concurrency goes up usage does too? | Does anthropic place a penalty on subagents?! maybe as concurrency goes up usage does too? | ↗Not briefed |
bluesky bluesky-ai | 1mo | The decomp/recomp scene sucks so bad, man. Full of the most devout AI shills you'll ever meet, kinda guys who would happily put their grand… | The decomp/recomp scene sucks so bad, man. Full of the most devout AI shills you'll ever meet, kinda guys who would happily put their… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic warns AI may pose 'existential risks to humanity' in IPO filing - www.reuters.com/business/fin... | Anthropic warns AI may pose 'existential risks to humanity' in IPO filing - www.reuters.com/business/fin... | ↗Not briefed |
bluesky bluesky-ai | 1mo | this is actually gender-validating because it assigns me the same amount of personhood as an LLM in the same way as misogyny is gender-val… | this is actually gender-validating because it assigns me the same amount of personhood as an LLM in the same way as misogyny is gende… | ↗Not briefed |