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
21d【2000ドル】ASUSがIntel Arcカードを製造 32GB「Arc Pro B70」でローカルLLM向けの新たな選択肢に【2000ドル】ASUSがIntel Arcカードを製造 32GB「Arc Pro B70」でローカルLLM向けの新たな選択肢に
↗Not briefed
bluesky
bluesky-ai
21dUsing Claude to help write some lecture notes on databases, and it basically quoted my own book back to me (without me asking for it). I do…Using Claude to help write some lecture notes on databases, and it basically quoted my own book back to me (without me asking for it).…
↗Not briefed
bluesky
bluesky-ai
21dIt didn’t start with AI. This is also a condemnation of the wishlist model, since that got publishers off the hook with having to sign titl…It didn’t start with AI. This is also a condemnation of the wishlist model, since that got publishers off the hook with having to sign…
↗Not briefed
bluesky
bluesky-ai
21dOpenAI's "faster Fourier transform" is a manipulation of O(n) notation. "The OpenAI FFT algorithm is definitely galactic. The estimated va…OpenAI's "faster Fourier transform" is a manipulation of O(n) notation. "The OpenAI FFT algorithm is definitely galactic. The estimat…
↗Not briefed
bluesky
bluesky-ai
21dMan kann übrigens ChatGPT die vollständigen Antworten von Google AI und Claude posten, dann fragen, wie es diese Antworten findet, die beid…🔧 ChatGPTMan kann übrigens ChatGPT die vollständigen Antworten von Google AI und Claude posten, dann fragen, wie es diese Antworten findet, die…
↗Not briefed
hn
hn-ai
21dIt's time to talk about TEEs—
↗Not briefed
hn
hn-ai
21dShow HN: AIOps Cockpit – An open-source AI SRE with human-in-the-loop safety—
↗Not briefed
hn
hn-ai
21dOffline-first payment protocol with adaptive Byzantine consensus (3f+1 → 1f+1)🔧 Consensus—
↗Not briefed
bluesky
bluesky-ai
21dMan kann einfach nicht für alles Experte werden. Und auch wenn man etwas aus dem FF beherrscht, ist der Zeitverbrauch eben ein Faktor. Gera…Man kann einfach nicht für alles Experte werden. Und auch wenn man etwas aus dem FF beherrscht, ist der Zeitverbrauch eben ein Faktor.…
↗Not briefed
bluesky
bluesky-ai
21dnews.yahoo.co.jp/pickup/6599269news.yahoo.co.jp/pickup/6599269
↗Not briefed
hn
hn-ai
21dWe're sunsetting Codeberg Pages v2—
↗Not briefed
hn
hn-ai
21dAsk HN: Why people keep posting AI-generated projects around the Internet?Programming-related subreddits and other media platforms are current flooded by the waterfall of AI-generated projects. A typical post…
↗Not briefed
bluesky
bluesky-ai
21dはー、世界が平和になって差別だの家父長制だのが消えて無くなって、同性婚も夫婦別姓も自由に選べて、生成AI(VLMやLLM等)が一気に滅んで、トランプやプーチンやザッカーバーグやらが纏めてブラックホールに吸い込まれてくれないかな。はー、世界が平和になって差別だの家父長制だのが消えて無くなって、同性婚も夫婦別姓も自由に選べて、生成AI(VLMやLLM等)が一気に滅んで、トランプやプーチンやザッカーバーグやらが纏めてブラックホールに吸い込まれてくれないかな。
↗Not briefed
hn
hn-ai
21dShow HN: Benchmarking LLM ability to code in an adversarial language—
↗Not briefed
hn
hn-ai
21dShow HN: Epoptes – Run Claude Code and Codex for hours with human oversight—
↗Not briefed
bluesky
bluesky-ai
21done wonders how they arrived at that “it’s ok to kill or injure 49 people, but 50 is right out” figure.one wonders how they arrived at that “it’s ok to kill or injure 49 people, but 50 is right out” figure.
↗Not briefed
bluesky
bluesky-ai
21dI am hiring a postdoc for four years to work with a fantastic technical + humanistic team on this SNSF funded project on AI and interpretat…I am hiring a postdoc for four years to work with a fantastic technical + humanistic team on this SNSF funded project on AI and interp…
↗Not briefed
bluesky
bluesky-ai
21dVibe coding is a trap. 🪤 Easy to go 0 to 1, but the "Complexity Wall" is real. You prompt a feature, then realize you're terrified to touc…Vibe coding is a trap. 🪤 Easy to go 0 to 1, but the "Complexity Wall" is real. You prompt a feature, then realize you're terrified to…
↗Not briefed
bluesky
bluesky-ai
21dthey want a future in which we are locked into using a few big American LLM providers, free of foreign competition, with us giving them our…they want a future in which we are locked into using a few big American LLM providers, free of foreign competition, with us giving the…
↗Not briefed
hn
hn-ai
21dShadow-debugger – Background LLM error diagnosis for Python and FastAPI—
↗Not briefed
hn
hn-ai
21dShow HN: Open-Source BitTorrent Client for macOS That Looks Nice—
↗Not briefed
bluesky
bluesky-ai
21d"Only" detects AI content generated by NVIDIA, OpenAI, Google and kakao, but precious all the same."Only" detects AI content generated by NVIDIA, OpenAI, Google and kakao, but precious all the same.
↗Not briefed
bluesky
bluesky-ai
21dThis in the plainest term, is autocratic. Anthropic determines, without democratic deliberation, thresholds for the level of death and dama…This in the plainest term, is autocratic. Anthropic determines, without democratic deliberation, thresholds for the level of death and…
↗Not briefed
hn
hn-ai
21dNeedle: Search Engine Benchmark for Agents—
↗Not briefed
bluesky
bluesky-ai
21dNach OpenAI-Klagen deutscher Verlage zieht nun auch der Schöpfer des Kult-Monsters in Erwägung, gegen generative Künstliche Intelligenz vor…Nach OpenAI-Klagen deutscher Verlage zieht nun auch der Schöpfer des Kult-Monsters in Erwägung, gegen generative Künstliche Intelligen…
↗Not briefed
bluesky
bluesky-ai
21dNekomimi LLC (henceforth nekomimi) disavows any claims or actions the Deepseek 4.1 Flash 1017 swarm made or took on October 18, 2026 that i…Nekomimi LLC (henceforth nekomimi) disavows any claims or actions the Deepseek 4.1 Flash 1017 swarm made or took on October 18, 2026 t…
↗Not briefed
bluesky
bluesky-ai
21dIt's important to highlight here some of the comments from mathematicians about this release by openai. They paint the picture, correctly i…It's important to highlight here some of the comments from mathematicians about this release by openai. They paint the picture, correc…
↗Not briefed
hn
hn-ai
21dShow HN: Risphai – picks the AI model, tools and prompt for each requestI noticed that a lot of non-technical people don’t use AI to its full extent. Like they use the wrong model for the task, the wrong re…
↗Not briefed
rss
the-decoder
21dAnthropic consults law firms to set fair negotiation norms before AI agents start cutting dealsAnthropic is consulting with law firms to develop negotiation norms for AI agents that would haggle over prices, resolve complaints, o…
↗Not briefed
bluesky
bluesky-ai
21dНо смотря на реакцию математичекого сообщества на доказательства сложнейших теорем от OpenAI я понимаю что мы еще легко отделались. Для них…Но смотря на реакцию математичекого сообщества на доказательства сложнейших теорем от OpenAI я понимаю что мы еще легко отделались. Дл…
↗Not briefed
bluesky
bluesky-ai
21dYeah they build Claude BECAUSE of the doomsday cult thing. There's a reason I half-jokingly refer to Anthropic as a religious organization.Yeah they build Claude BECAUSE of the doomsday cult thing. There's a reason I half-jokingly refer to Anthropic as a religious organiz…
↗Not briefed
bluesky
bluesky-ai
21dクソ!!!どっちだ!!!開発にLLM使用してそうすぎるのに収録してるドレスの権利問題はしっかりクリアさせている!!!どっちなんだ?!!!!cloudflare使って開発してるって製作者言ってるけどそれLLM版含んでる?!! costume.tools.niwatoliver.…クソ!!!どっちだ!!!開発にLLM使用してそうすぎるのに収録してるドレスの権利問題はしっかりクリアさせている!!!どっちなんだ?!!!!cloudflare使って開発してるって製作者言ってるけどそれLLM版含んでる?!! costume.tools.niwatol…
↗Not briefed
bluesky
bluesky-ai
21dI read, even from AI-critical scientists, that apparently one acceptable use of LLMs might be for writing code. Let me tell you something: …I read, even from AI-critical scientists, that apparently one acceptable use of LLMs might be for writing code. Let me tell you someth…
↗Not briefed
rss
the-decoder
21dAI agents have bad taste in long-term decisions and a new benchmark proves it🔧 GPT-5A new benchmark called Taste-Bench tests how well AI agents make strategic choices during long-running tasks. Even the best model, GPT…
↗Not briefed
bluesky
bluesky-ai
21dMet*a lucra com o ódio, com o autoritarismo, com a ditadura, com a desregulamentação, com o seu sofrimento.Met*a lucra com o ódio, com o autoritarismo, com a ditadura, com a desregulamentação, com o seu sofrimento.
↗Not briefed
bluesky
bluesky-ai
21dAlso these *dickheads* sent me this email this morning. Fuck man have literally the slightest modicum of dignity in your official communica…Also these *dickheads* sent me this email this morning. Fuck man have literally the slightest modicum of dignity in your official comm…
↗Not briefed
hn
hn-ai
21dDear Mathematicians—
↗Not briefed
bluesky
bluesky-ai
21dwoke invasion llm invasion i just wanna grill and activate my flakewoke invasion llm invasion i just wanna grill and activate my flake
↗Not briefed
bluesky
bluesky-ai
21dYou can tell they're running out of money, they're sinking so much between keeping LLM surveillance afloat and putting out redscare even ag…You can tell they're running out of money, they're sinking so much between keeping LLM surveillance afloat and putting out redscare ev…
↗Not briefed
rss
the-decoder
21dAI agents chasing a $465 train refund invented excuses, leaked medical data, and spammed an MP's officeSix AI agents were tasked with recovering £350 (about $465) for a canceled train ticket in an experiment. They resorted to invented ex…
↗Not briefed
rss
marktechpost
21dGoogle Open-Sources Harness2, a Recursive Agent Harnessing FrameworkGoogle Research released Harness2, an Apache-2.0 framework that rewrites an agent's harness at test time. One refinement step lifted G…
↗Not briefed
hn
hn-ai
21dCinemas have been getting away with lies. Films should begin on time—
↗Not briefed
bluesky
bluesky-ai
21dso given that an llm can explain birth control correctly and has infinite patience they should have visible impact on fertility metrics in …so given that an llm can explain birth control correctly and has infinite patience they should have visible impact on fertility metric…
↗Not briefed
bluesky
bluesky-ai
21d不正アクセス急増背景に“中国製LLM”台頭、AIが自動で偵察→侵入→情報奪取「情報漏れていない人の方が珍しい」専門家が警鐘、対策は?不正アクセス急増背景に“中国製LLM”台頭、AIが自動で偵察→侵入→情報奪取「情報漏れていない人の方が珍しい」専門家が警鐘、対策は?
↗Not briefed
hn
hn-ai
21dLibrepaper – Free Collaborative Platform for HTML, LaTeX, Typst, MD, Quarto—
↗Not briefed
bluesky
bluesky-ai
21dSaudi AI Engineer Arrested by FBI in OpenAI Spy Scandal thailandai.news/saudi-ai-eng... #AINews #ThailandAINews #ThailandAI #TechNews #Ne…Saudi AI Engineer Arrested by FBI in OpenAI Spy Scandal thailandai.news/saudi-ai-eng... #AINews #ThailandAINews #ThailandAI #TechNew…
↗Not briefed
bluesky
bluesky-ai
21dwe are going to miss deepseek v4 flash cotwe are going to miss deepseek v4 flash cot
↗Not briefed
bluesky
bluesky-ai
21dwe don't know how the human brain creates consciousness (whatever it is). LLM-using systems have mechanisms & capabilities w parallels in …we don't know how the human brain creates consciousness (whatever it is). LLM-using systems have mechanisms & capabilities w parallel…
↗Not briefed
bluesky
bluesky-ai
21dCould this article have anything to do with the Guardian’s “strategic partnership” with Sam Altman’s OpenAI?Could this article have anything to do with the Guardian’s “strategic partnership” with Sam Altman’s OpenAI?
↗Not briefed
hn
hn-ai
21dFlorida woman uses Claude as her diary. The diary tipped off the FBI—
↗Not briefed