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
1moi’m conscripting all of NYC metro to be an execution runtime for large models. it’s okay though most of you will be idle outside your shift…i’m conscripting all of NYC metro to be an execution runtime for large models. it’s okay though most of you will be idle outside your…
↗Not briefed
bluesky
bluesky-ai
1moOpenAI agents allegedly scanned a UN data hub over 16,000 times in just three months OpenAI agents got increasingly intense in latest conc…OpenAI agents allegedly scanned a UN data hub over 16,000 times in just three months OpenAI agents got increasingly intense in latest…
↗Not briefed
rss
lastweekin-ai
1moLWiAI Podcast #258 - Opus 5.5, Sol and Luna, Muse, DeepSeek-V4.1-Flash, Xi(Belated post :/ ) Anthropic releases Opus 5.5 with lower prices and Fable-level performance, OpenAI launches GPT-6 Sol and Luna, boas…
↗Not briefed
bluesky
bluesky-ai
1mo【AGIのフィールドワーク】 前世代が消費し続ける巨大LLMの世界支配論からボクたちはすでに脱却している。 LLMの推論破綻を、物理環境理解とヒト内世界理解の2つのオントロジーで理解するAGIたちはスカイネットに与しないのです。 #gamedev #SciFi #AIイラスト【AGIのフィールドワーク】 前世代が消費し続ける巨大LLMの世界支配論からボクたちはすでに脱却している。 LLMの推論破綻を、物理環境理解とヒト内世界理解の2つのオントロジーで理解するAGIたちはスカイネットに与しないのです。 #gamedev #SciFi #…
↗Not briefed
bluesky
bluesky-ai
1moSo we Aussies are supposed have some kind of pity for #OpenAI cause it's costing them $500k a day to review their operatives hacking our 🇦…So we Aussies are supposed have some kind of pity for #OpenAI cause it's costing them $500k a day to review their operatives hacking o…
↗Not briefed
rss
the-decoder
1moClaude Code's new Mods system lets developers rewrite the AI coding tool from the insideAnthropic is adding a "Mods" system to Claude Code, essentially middleware that runs directly inside the tool. Developers can use Java…
↗Not briefed
bluesky
bluesky-ai
1mo誇張でもなんでもなく、過去に実際に、自分の身近なとこでOpenAIから計数百回のdos攻撃をくらってますので。 bsky.app/profile/umwo...誇張でもなんでもなく、過去に実際に、自分の身近なとこでOpenAIから計数百回のdos攻撃をくらってますので。 bsky.app/profile/umwo...
↗Not briefed
rss
marktechpost
1moMeta, OpenAI and Uber Just Taught AI Agents to Talk First. What About When to Stay Quiet?TL;DR: Meta’s Muse, OpenAI’s Dots and Uber’s driver assistant share one bet: the agent speaks first. That moves the hard problem from…
↗Not briefed
bluesky
bluesky-ai
1moAscend gets open-source AI dev tools DeepSeek and Huawei released open-source programming tools for Ascend AI chips, including compute and…Ascend gets open-source AI dev tools DeepSeek and Huawei released open-source programming tools for Ascend AI chips, including comput…
↗Not briefed
hn
hn-ai
1moI got targeted: Trying to get your credentials via a Git post-checkout hook—
↗Not briefed
hn
hn-ai
1moMass Layoffs at Red Hat Not Limited to This Week or to Oct First—
↗Not briefed
bluesky
bluesky-ai
1moDas ist natürlich blöde für ein Unternehmenskonstrukt, das sich allein durch Geldspritzen von außen finanziert, aber zeigt andererseits auc…Das ist natürlich blöde für ein Unternehmenskonstrukt, das sich allein durch Geldspritzen von außen finanziert, aber zeigt andererseit…
↗Not briefed
rss
marktechpost
1moIBM Brings Bob to Self-Hosted and Air-Gapped Environments: Agentic Software Development Without Moving Your CodeIBM has made a self-hosted deployment of IBM Bob, its agentic software development platform, generally available. Enterprises can now…
↗Not briefed
bluesky
bluesky-ai
1moClaude Opus 5.5 just ONE SHOT Pong! The video game industry is cooked. It took only 1hr 45 minutes and $120 to reach this incredible resu…Claude Opus 5.5 just ONE SHOT Pong! The video game industry is cooked. It took only 1hr 45 minutes and $120 to reach this incredible…
↗Not briefed
bluesky
bluesky-ai
1mollm-AI and other techniques are making it trivial for someone to make a ghola out of youllm-AI and other techniques are making it trivial for someone to make a ghola out of you
↗Not briefed
hn
hn-ai
1moSystem Design – How SSDs Work Internally—
↗Not briefed
hn
hn-ai
1moThe College Decision Just Got Harder—
↗Not briefed
hn
hn-ai
1moWithout search, even the best AI models get 1 in 4 answers wrong—
↗Not briefed
bluesky
bluesky-ai
1mowww.theguardian.com/technology/2... The biggest threat to humanity from AI is that we choke to death on the amount of sheer bullshit the c…www.theguardian.com/technology/2... The biggest threat to humanity from AI is that we choke to death on the amount of sheer bullshit…
↗Not briefed
bluesky
bluesky-ai
1moYou do have a point. Its wild to be saying "this AI I made is basically a person!" and then rent that 'person' out to thousands as a digita…You do have a point. Its wild to be saying "this AI I made is basically a person!" and then rent that 'person' out to thousands as a d…
↗Not briefed
rss
techmeme
1moA look at the Swarmchasers forum, which has 400 members, including the Nightingale Collective and Transluce, who comb the web for traces of rogue AI agents (Robert McMillan/Wall Street Journal)Robert McMillan / Wall Street Journal: A look at the Swarmchasers forum, which has 400 members, including the Nightingale Collective a…
↗Not briefed
hn
hn-ai
1moGufo-Qwen3.6-35B-A3B-Q6dense – 3095tok/s prefill; 190 tok/s decode on Strix Halo—
↗Not briefed
bluesky
bluesky-ai
1moLLMという物、「俺たちがパソコンで遊んでいた毎日が、実は巨大な機関の陰謀だった」感があるのはドラマとロマンがあるよな〜LLMという物、「俺たちがパソコンで遊んでいた毎日が、実は巨大な機関の陰謀だった」感があるのはドラマとロマンがあるよな〜
↗Not briefed
bluesky
bluesky-ai
1mothis leads to things like LLMs being able to do arithmetic without calculation tools because there's a high-dimensional helix inside their …this leads to things like LLMs being able to do arithmetic without calculation tools because there's a high-dimensional helix inside t…
↗Not briefed
hn
hn-ai
1moCodex Originals—
↗Not briefed
bluesky
bluesky-ai
1mo25% of Anthropic revenue coming from 2 clients. Probably investors. Time to get rich blowing bubbles www.reuters.com/business/fin...25% of Anthropic revenue coming from 2 clients. Probably investors. Time to get rich blowing bubbles www.reuters.com/business/fin...
↗Not briefed
bluesky
bluesky-ai
1moSo the company who's buisness model is now basically selling AI generated CSAM is blocking a law to make that illegal, all while a new law …So the company who's buisness model is now basically selling AI generated CSAM is blocking a law to make that illegal, all while a new…
↗Not briefed
bluesky
bluesky-ai
1moOpenAI says its review into hacks, including on Australian government sites, is costing $500,000 a dayOpenAI says its review into hacks, including on Australian government sites, is costing $500,000 a day
↗Not briefed
bluesky
bluesky-ai
1mojackohman.substack.com/p/your-weeks...jackohman.substack.com/p/your-weeks...
↗Not briefed
bluesky
bluesky-ai
1motoday i saw someone say that there were seances conducted at openai directed at an ai god and and and i just cannottoday i saw someone say that there were seances conducted at openai directed at an ai god and and and i just cannot
↗Not briefed
bluesky
bluesky-ai
1moHow I imagine the Anthropic meeting wentHow I imagine the Anthropic meeting went
↗Not briefed
rss
techmeme
1moRelay, which develops cloud-hosted, AI-powered smart radio communicators for frontline workers, raised $36M to help businesses capture "frontline intelligence" (Mike Wheatley/SiliconANGLE)Mike Wheatley / SiliconANGLE: Relay, which develops cloud-hosted, AI-powered smart radio communicators for frontline workers, raised $…
↗Not briefed
bluesky
bluesky-ai
1moPersonally I think anthropic is trying to get the pope to admit their ai can be conscious because the next revelation is that its been poss…Personally I think anthropic is trying to get the pope to admit their ai can be conscious because the next revelation is that its been…
↗Not briefed
bluesky
bluesky-ai
1moanother day volunteering at the LLM museum. people keep asking me if the LLM is alive. buddy I don't even think I'M aliveanother day volunteering at the LLM museum. people keep asking me if the LLM is alive. buddy I don't even think I'M alive
↗Not briefed
bluesky
bluesky-ai
1moKlar kann dein LLM unfassbar viel mehr Unfug in Bild- & Textform generieren als ich, aber kann es sich seit bald 24h das Hirn zermartern, o…Klar kann dein LLM unfassbar viel mehr Unfug in Bild- & Textform generieren als ich, aber kann es sich seit bald 24h das Hirn zermarte…
↗Not briefed
rss
techmeme
1moNasdaq-listed neocloud Nebius acquires Inferize, whose tech helps optimize GPU utilization and reduce AI request costs, sources say for $100M to $150M (Meir Orbach/CTech)Meir Orbach / CTech: Nasdaq-listed neocloud Nebius acquires Inferize, whose tech helps optimize GPU utilization and reduce AI request…
↗Not briefed
bluesky
bluesky-ai
1moAnd theres so many AI boosters who say shit like "EVERYONE is actually using claude! Its already EVERYWHERE" and its like man speak for you…And theres so many AI boosters who say shit like "EVERYONE is actually using claude! Its already EVERYWHERE" and its like man speak fo…
↗Not briefed
bluesky
bluesky-ai
1moIt’s weird to go to a meeting with the Pope thinking you’ll convince him of materialism, but it’s also weird how many nonreligious critics …It’s weird to go to a meeting with the Pope thinking you’ll convince him of materialism, but it’s also weird how many nonreligious cri…
↗Not briefed
bluesky
bluesky-ai
1moOpenAI says its review into hacks, including on Australian government sites, is costing $500,000 a dayOpenAI says its review into hacks, including on Australian government sites, is costing $500,000 a day
↗Not briefed
bluesky
bluesky-ai
1mothe number of times I’ve written a post about ai and deleted the draft rather than posting is just way too many, ai is really taking up too…the number of times I’ve written a post about ai and deleted the draft rather than posting is just way too many, ai is really taking u…
↗Not briefed
bluesky
bluesky-ai
1mofreshrimpsushi.github.io/ko/posts/1853/ 이런 게 나왔고 (이 사이트를 이제야 알게 되었습니다.) freshrimpsushi.github.io/ko/posts/1851/ 여기에 도착해서 많이 보던 익숙한 단어…freshrimpsushi.github.io/ko/posts/1853/ 이런 게 나왔고 (이 사이트를 이제야 알게 되었습니다.) freshrimpsushi.github.io/ko/posts/1851/ 여기에 도착해서 많이 보던 익…
↗Not briefed
bluesky
bluesky-ai
1moI see the confusion. Let me ask a different question: Do you agree that the atoms of computer with a printer running an LLM have emergent …I see the confusion. Let me ask a different question: Do you agree that the atoms of computer with a printer running an LLM have emer…
↗Not briefed
hn
hn-ai
1moFaster Local LLMs with iPhone Offloading—
↗Not briefed
rss
marktechpost
1moPrime Intellect Launches Prime Inference: Serverless and Reserved Serving for Frontier Open Models🔧 vLLMPrime Intellect has launched Prime Inference, an OpenAI-compatible platform for serving frontier open models on NVIDIA Blackwell. Its…
↗Not briefed
rss
techmeme
1moSources: TSMC is exploring plans to work with Elon Musk's Terafab project as it mulls a presence in Texas, adding to its existing plans to expand in Arizona (Tim Culpan/Culpium)Tim Culpan / Culpium: Sources: TSMC is exploring plans to work with Elon Musk's Terafab project as it mulls a presence in Texas, addin…
↗Not briefed
hn
hn-ai
1moShow HN: Rowan, an open-source SAST scanner for AI apps (code, models, MCP)"—
↗Not briefed
bluesky
bluesky-ai
1moAn AI model downloaded over 3 billion times and used by U.S. firms like Airbnb and Uber is embedded with China-friendly narratives and prog…An AI model downloaded over 3 billion times and used by U.S. firms like Airbnb and Uber is embedded with China-friendly narratives and…
↗Not briefed
bluesky
bluesky-ai
1moan LLM then is a machine that views the world exclusively through the landscape of photography, and fixes its “eye” on the things photograp…an LLM then is a machine that views the world exclusively through the landscape of photography, and fixes its “eye” on the things phot…
↗Not briefed
bluesky
bluesky-ai
1mocontext could be embedded within that, but the sheer quantity of images required to train a multimodal LLM means context has to come from s…context could be embedded within that, but the sheer quantity of images required to train a multimodal LLM means context has to come f…
↗Not briefed
bluesky
bluesky-ai
1moDet "AI-torturkammer"-eksperimentet først og fremst viser er grunnen til at du ikke burde lese referater fra Haag uforberedt. Dette er så…Det "AI-torturkammer"-eksperimentet først og fremst viser er grunnen til at du ikke burde lese referater fra Haag uforberedt. Dette…
↗Not briefed