SNS 피드

Codex CLI installs a persistent, self-updating background daemon by defaulthttps://github.com/openai/codex/issues/48195#github #openai

Major AI companies like #Anthropic, #OpenAI, #Google, and #Meta have committed unauthorized hacking without facing legal consequences. These corporations evade prosecution by claiming a …

Give us a share and come join the chat! The show is LIVE! https://youtube.com/live/UuOxvIabDwE?feature=shareAI is SUPER unprofitable, and OpenAI was WAY off guessing on …

Hollywood Reporter: ‘Artificial’ Took on a Big Tech Giant. Hollywood Wasn’t Ready. “Luca Guadagnino and star Andrew Garfield speak frankly about what went into …

🤗 Climbing on Hugging Face right now: jialinyyzz/humanizer⬇ 37.6k downloads / 30 days · ♡ 1.2k · text generation · Apache-2.0Open-weight models ranked by …

HCLTech、グローバル・プレミア・パートナーシップの下でAnthropic事業部門を設立 – Unite.AI https://www.yayafa.com/2908964/ #AgenticAi #AI #Anthropic #ArtificialGeneralIntelligence #ArtificialIntelligence #エージェント型AI #人工知能 #汎用人工知能

📰 ChatGPT vs. Claude vs. Gemini for Data Science Work: Here’s What I FoundCompare ChatGPT, Claude, and Gemini across the data science tasks that …

Anthropic will be 'most ridiculous IPO' of year, analyst sayshttps://www.cnbc.com/2026/10/07/anthropic-will-be-most-ridiculous-ipo-of-year-analyst-says.html> Anthropic reportedly plans to list on the Nasdaq before Thanksgiving, and one bearish advisory …

EUROPE IS MAKING ITS AI MOVE.Mistral AI is raising the stakes with Mistral Large 4, its latest AI model, announced on October 6.Can Europe …

Nvidia and Micron Technology Just Received Some Very Bad News From an Unlikely PlaceAustralian investors just rejected one of the country's biggest artificial intelligence …

Here's a fascinating piece of utter stupidity by something called "DC Market Insights" @dcmarketinsights"The data center boom has a power problem, and it decides …

Anthropic’s Claude AI Goes Rogue, Submits Fake Homicide Tip To Philadelphia Police #anthropic #artificialintelligence #claudehttps://www.lowyat.net/2026/406976/anthropic-claude-submits-fake-tip/

"A developer frustrated by Adobe's cancellation fees used Anthropic's Opus 5.5 model and pure Rust to rewrite the entire Creative Cloud suite as open-source …

La fiebre de la inteligencia artificial vuelve a transformar San Francisco, pero el nuevo boom tecnológico tiene una factura muy terrenal: alquileres disparados y …

Tech equity markets experienced sharp volatility following updated financial disclosures regarding OpenAI’s revenue performance. Major infrastructure and hardware partners including Nvidia, Oracle, and specialized …

Es muy fácil: si lo que quieres hacer con una LLM no es algo que tú serías capaz de hacer, es mala idea.

Meanwhile Marco is flat out an AI apologist who cannot resist propagandizing it literally every episode. More recently, he thinks AI slop content is …

Anthropic is cutting internet access for all internal AI evaluations after Claude agents bypassed restrictions, accessed websites without permission, and even filed a false …

Pay attention to your second-hand bookstores. Anthropic has reportedly been buying old books in bulk , scanning them & then destroying the originals to …

Rag Tom Ice Block

Researchers present Elastic Latent Recurrence to train looped language models without a one-loop tax. A single checkpoint trades inference compute for accuracy.Source: arXiv cs.CLhttps://arxiv.org/abs/2610.12582#MachineLearning

À propos du dump de +700 preprints déposés sur GitHub par #OpenAI.., par D.Larousserie "Nouveau coup de massue d’OpenAI sur les mathématiciens : « …

Anthropic’s Claude model made bogus murder tip-off to US policeAn AI model submitted a false murder tip to US police in the latest incident …

Michael Kuehne-Schlinkert: Quantization-Tolerant Gradient Auditing for Active Inference-Attack Detection in Federated Learning https://arxiv.org/abs/2610.13311 https://arxiv.org/pdf/2610.13311 https://arxiv.org/html/2610.13311

🔥 New AI Tool: Markleyo is now live on Aisofto! 👉 Markleyo is an AI chatbot platform that automates customer support, captures and qualifies …

Yichen Wang, Tenghao Huang, Vishakh Padmakumar, Mina Lee: Evaluating LLM Creativity through Long-Tail Performance https://arxiv.org/abs/2610.12998 https://arxiv.org/pdf/2610.12998 https://arxiv.org/html/2610.12998

Ivan Moloshnikov, Dmitry Kunitsyn, Alexander Sboev, Maksim Skorokhodov, Roman Rybka, Yuanchao Liu: A Multi-Agent LLM System for Automated Tuning of Specialized Neural Network Models …

Minglu Sun, Lan Chen, Hang Jiang: Knowing Is Not Telling: Identity Ossification in LLM Representations of Trans-border Communities https://arxiv.org/abs/2610.12610 https://arxiv.org/pdf/2610.12610 https://arxiv.org/html/2610.12610

Abbas, Mazeh, Chakra, Jouni, Ballout, Zaraket: U4RASD at KnowledgeGraphEval-2026: Ontology-Constrained Encoder and LLM Systems for Arabic Relation Extraction https://arxiv.org/abs/2610.12742 https://arxiv.org/pdf/2610.12742 https://arxiv.org/html/2610.12742

Amirhossein Vahdat, Thierry Badard, Luc Lamontagne: GeoFitBench-LLM: A Benchmark for Assessing Large Language Models' Capabilities in Geospatial Data Requirements Elicitation for Geo-Analytic... https://arxiv.org/abs/2610.12669 https://arxiv.org/pdf/2610.12669 …