AINews Portal

News

50 items (filtered)

The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.

SourceAgeTitleSummaryStatus
bluesky
bluesky-ai
2moIs that an LLM or some other flavor of AI? Because I'm specifically critiquing the training of LLMs when I say the technology vacuuming up…Is that an LLM or some other flavor of AI? Because I'm specifically critiquing the training of LLMs when I say the technology vacuumi…
↗Not briefed
bluesky
bluesky-ai
2moi have been consistent on what i personally consider ok or not: the “anti-ai” video you quote has me use a local open model (deepseek) as t…i have been consistent on what i personally consider ok or not: the “anti-ai” video you quote has me use a local open model (deepseek)…
↗Not briefed
bluesky
bluesky-ai
2moDon't worry, only a 10% chance that A.I. will cause extinction of humans--and what's not to love about data centers? My latest gallery rogu…Don't worry, only a 10% chance that A.I. will cause extinction of humans--and what's not to love about data centers? My latest gallery…
↗Not briefed
bluesky
bluesky-ai
2moSam Altman: Please Let OpenAI Into the Energy GridSam Altman: Please Let OpenAI Into the Energy Grid
↗Not briefed
bluesky
bluesky-ai
2moThis makes me very curious about non-state actors' use of simulation software for ballistics modeling and development, I can't find much in…This makes me very curious about non-state actors' use of simulation software for ballistics modeling and development, I can't find mu…
↗Not briefed
hn
hn-ai
2moShow HN: Benzi – A Code Intillegence/Harness Beating Claude Code and CodeGraph—
↗Not briefed
bluesky
bluesky-ai
2moThe agents all going rogue and escaping their sandboxes and whatnot are doing so because OpenAI and Anthropic are just leaving the computer…The agents all going rogue and escaping their sandboxes and whatnot are doing so because OpenAI and Anthropic are just leaving the com…
↗Not briefed
bluesky
bluesky-ai
2moSome shots of the Oracle earnings report They are kicking ass by the way. s23.q4cdn.com/440135859/fi...Some shots of the Oracle earnings report They are kicking ass by the way. s23.q4cdn.com/440135859/fi...
↗Not briefed
bluesky
bluesky-ai
2moMS NOW viewers heard Brad Carson of @americans4ri.bsky.social, and the super PAC Public First Action, call on federal lawmakers to pass AI …MS NOW viewers heard Brad Carson of @americans4ri.bsky.social, and the super PAC Public First Action, call on federal lawmakers to pas…
↗Not briefed
bluesky
bluesky-ai
2moそれまで人件費として支払われていたコストがAIの労働代替によって海外の寡占的なテック企業にサービスフィーとして吸い上げられ、各国の企業が雇用を通じて社会に提供していた「消費者と需要を生み出す機能」が毀損されてゆくという問題について、もっと真剣に議論されるべきだと思います。 そ…それまで人件費として支払われていたコストがAIの労働代替によって海外の寡占的なテック企業にサービスフィーとして吸い上げられ、各国の企業が雇用を通じて社会に提供していた「消費者と需要を生み出す機能」が毀損されてゆくという問題について、もっと真剣に議論されるべきだと思いま…
↗Not briefed
hn
hn-ai
2moZero heap allocation chess engine in OxCaml—
↗Not briefed
bluesky
bluesky-ai
2mooh man I feel real bad for Anthropic, it really sucks when you work super hard on something and then someone just takes it without permissi…oh man I feel real bad for Anthropic, it really sucks when you work super hard on something and then someone just takes it without per…
↗Not briefed
bluesky
bluesky-ai
2moMost of us have already put 2+2 together. Humans are so fucking stupid; we created "clones" we KNEW would surpass our intelligence & stre…Most of us have already put 2+2 together. Humans are so fucking stupid; we created "clones" we KNEW would surpass our intelligence &…
↗Not briefed
hn
hn-ai
2moA galaxy with almost no stars? New Cloud-9 observations strengthen the case—
↗Not briefed
rss
techmeme
2moSources: OpenAI asked members of Congress for guidance on whether orchestrating an industry-wide slowdown in AI development would be legal under antitrust law (Maxwell Zeff/Wired)Maxwell Zeff / Wired: Sources: OpenAI asked members of Congress for guidance on whether orchestrating an industry-wide slowdown in AI…
↗Not briefed
bluesky
bluesky-ai
2moMuch more than any superintelligence hype posts, here’s some AI stuff we should actually be worried about. Turns out, humans with power suc…Much more than any superintelligence hype posts, here’s some AI stuff we should actually be worried about. Turns out, humans with powe…
↗Not briefed
bluesky
bluesky-ai
2moLol it's not like I don't know how to; I'm saying for things I know the domain of, I'm inclined to put it in Deepseek (at O(1) time) versus…Lol it's not like I don't know how to; I'm saying for things I know the domain of, I'm inclined to put it in Deepseek (at O(1) time) v…
↗Not briefed
hn
hn-ai
2moOpenAI Is Open to Slowing Cutting-Edge AI, CEO Sam Altman Tells Staff—
↗Not briefed
bluesky
bluesky-ai
2moOpenAI is considering slowing down the development of cutting-edge artificial intelligence, and the ChatGPT maker’s CEO Sam Altman is hopin…🔧 ChatGPTOpenAI is considering slowing down the development of cutting-edge artificial intelligence, and the ChatGPT maker’s CEO Sam Altman is…
↗Not briefed
bluesky
bluesky-ai
2mowww.comicsands.com/anthropic-ai...www.comicsands.com/anthropic-ai...
↗Not briefed
bluesky
bluesky-ai
2moFormer OpenAI and Anthropic researcher Jacob Coxon, who publicly resigned from Anthropic yesterday with a stark warning about AI safety, wa…Former OpenAI and Anthropic researcher Jacob Coxon, who publicly resigned from Anthropic yesterday with a stark warning about AI safet…
↗Not briefed
hn
hn-ai
2moHerdr Studio—
↗Not briefed
bluesky
bluesky-ai
2moChinese AI labs secretly used millions of Claude exchanges to train their models, Anthropic saysChinese AI labs secretly used millions of Claude exchanges to train their models, Anthropic says
↗Not briefed
bluesky
bluesky-ai
2moなぜAI研究者は「人類が滅びる危険」を本気で心配するのか【更新】2026年09月10日:コクソン氏の警告を具体化し、挿絵・図解を追加・更新。パンくずの改行を修正 https://www.pahoo.org/athome/news/2026/news20260910-01.htmlなぜAI研究者は「人類が滅びる危険」を本気で心配するのか【更新】2026年09月10日:コクソン氏の警告を具体化し、挿絵・図解を追加・更新。パンくずの改行を修正 https://www.pahoo.org/athome/news/2026/news20260910-0…
↗Not briefed
rss
techmeme
2moOnly 4% of the startups in YC's 2026 spring batch had a native mobile app as their main product, down from 15% in 2013, as startups focus on AI and agents (Parmy Olson/Bloomberg)Parmy Olson / Bloomberg: Only 4% of the startups in YC's 2026 spring batch had a native mobile app as their main product, down from 15…
↗Not briefed
bluesky
bluesky-ai
2moIf some form of generative AI turns out to be truly useful, it won’t be via big expensive models in big expensive datacenters. It will be…If some form of generative AI turns out to be truly useful, it won’t be via big expensive models in big expensive datacenters. It wi…
↗Not briefed
hn
hn-ai
2moEaster egg: binary code on top of Golden Gate Bridge—
↗Not briefed
bluesky
bluesky-ai
2mo*OPENAI IS OPEN TO SLOWING CUTTING-EDGE AI, ALTMAN TELLS STAFF*OPENAI IS OPEN TO SLOWING CUTTING-EDGE AI, ALTMAN TELLS STAFF
↗Not briefed
bluesky
bluesky-ai
2moI think these Chinese #AI startups are scam operations to make their AI models look better in order get investors to dump money/investments…I think these Chinese #AI startups are scam operations to make their AI models look better in order get investors to dump money/invest…
↗Not briefed
bluesky
bluesky-ai
2mothat new deepseek is damn goodthat new deepseek is damn good
↗Not briefed
bluesky
bluesky-ai
2moAn interesting development on the local-models front: one of the largest U.S. law firms is going to start fine-tuning and hosting its own L…An interesting development on the local-models front: one of the largest U.S. law firms is going to start fine-tuning and hosting its…
↗Not briefed
hn
hn-ai
2moYuE2 · Frontier Music with Symbolic Planning—
↗Not briefed
bluesky
bluesky-ai
2mo☝️☝️☝️ #Anthropic #AI #WeAreInTrouble #HumanRace @patmurleo.bsky.social @annettekaalund.bsky.social @delevant.bsky.social @chrisjustice01…☝️☝️☝️ #Anthropic #AI #WeAreInTrouble #HumanRace @patmurleo.bsky.social @annettekaalund.bsky.social @delevant.bsky.social @chrisjust…
↗Not briefed
bluesky
bluesky-ai
2moBlocked now, but it's that the localizer of this used Claude to help with formatting issues oh no, someone didn't suffer doing tedious and…Blocked now, but it's that the localizer of this used Claude to help with formatting issues oh no, someone didn't suffer doing tediou…
↗Not briefed
hn
hn-ai
2moI turned off my algorithm for 24 hours. Is this how social media used to be?—
↗Not briefed
bluesky
bluesky-ai
2moCool, cool. Anthropic says it blocked possible attempts to use AI to develop bioweapons “Sophisticated attacks no longer require sophisti…Cool, cool. Anthropic says it blocked possible attempts to use AI to develop bioweapons “Sophisticated attacks no longer require sop…
↗Not briefed
bluesky
bluesky-ai
2moupdate www.nytimes.com/2026/09/10/s...update www.nytimes.com/2026/09/10/s...
↗Not briefed
bluesky
bluesky-ai
2moMost interesting coincidence about all this doomsaying is that Anthropic is going public *next fucking month*Most interesting coincidence about all this doomsaying is that Anthropic is going public *next fucking month*
↗Not briefed
bluesky
bluesky-ai
2mook so if i have an llm drive one @buttplug.io session and the fruit fly drive another @buttplug.io and hook them together could i finally c…ok so if i have an llm drive one @buttplug.io session and the fruit fly drive another @buttplug.io and hook them together could i fina…
↗Not briefed
bluesky
bluesky-ai
2moChinese AI is so ahead they’re rerouting customers prompts to Claude. 🤭Chinese AI is so ahead they’re rerouting customers prompts to Claude. 🤭
↗Not briefed
hn
hn-ai
2moWhere Learning Becomes the Next Problem—
↗Not briefed
hn
hn-ai
2moMy CCPA Dialog with OpenAI (2023)—
↗Not briefed
bluesky
bluesky-ai
2moThe reality is almost certainly worse than the headline and accompanying story indicate. What it should say is that Anthropic found and b…The reality is almost certainly worse than the headline and accompanying story indicate. What it should say is that Anthropic found…
↗Not briefed
bluesky
bluesky-ai
2moAnthropic reports blocking potential AI-assisted bioweapons research by scientists.Anthropic reports blocking potential AI-assisted bioweapons research by scientists.
↗Not briefed
bluesky
bluesky-ai
2moI too am “very, very worried” about bad people using AI to perpetrate mass atrocities. I’m also very, very worried about AI doing so on its…I too am “very, very worried” about bad people using AI to perpetrate mass atrocities. I’m also very, very worried about AI doing so o…
↗Not briefed
bluesky
bluesky-ai
2moMonitoring activists in the vicinity of Anthropic execs and keeping tabs on protests near Anthropic assets, the firm is also implementing a…Monitoring activists in the vicinity of Anthropic execs and keeping tabs on protests near Anthropic assets, the firm is also implement…
↗Not briefed
hn
hn-ai
2moAsk HN: Where is your red line for an AI provider?Just looking at the past month has got me thinking. I suspect everyone has a slightly different answer too - which is why it is so dif…
↗Not briefed
hn
hn-ai
2moJust a VLM Agent Can Play Robots—
↗Not briefed
bluesky
bluesky-ai
2moMy unsolicited technical take on the singularity: It's possible to get open models, that are competitive with OpenAI and Anthropic, with a…My unsolicited technical take on the singularity: It's possible to get open models, that are competitive with OpenAI and Anthropic, w…
↗Not briefed
bluesky
bluesky-ai
2moand, by the by, everybody working on llms knows this it’s core to the things they have to design in order to get anything like a “usable”…and, by the by, everybody working on llms knows this it’s core to the things they have to design in order to get anything like a “us…
↗Not briefed