News
Ar Xiv Will Ban Authors One Year for Unchecked LLM Output
2+ hour, 32+ min ago (24+ words) Ar Xiv will impose a one-year submission ban when a preprint shows incontrovertible evidence that authors did not review LLM-generated text or references....
Stanford HAI Launches Lab to Study Workplace AI
4+ day, 5+ hour ago (206+ words) Stanford HAI has launched the AI and Organizations Lab to study how workplace AI changes jobs, teams, decision-making, and accountability with research....
5% GPU Utilization: Enterprises Face Underused GPU Fleets as AI Costs Rise
6+ day, 13+ hour ago (364+ words) Enterprises are confronting 5% GPU utilization as AI infrastructure costs rise, pushing buyers toward inference efficiency, governance, and power-aware scaling....
Mistral Medium 3. 5 Folds Chat, Reasoning, and Code Into One 128 B AI Model
2+ week, 1+ day ago (25+ words) Mistral has released Medium 3. 5, a 128 B-parameter dense flagship folding chat, reasoning, and code into one model with a per-request reasoning effort toggle....
MIT, IBM Broaden AI Partnership With Quantum Lab Launch
2+ week, 4+ day ago (82+ words) MIT and IBM have launched the MIT-IBM Computing Research Lab, expanding their AI partnership into quantum computing, algorithms, and hybrid computing systems....
Microsoft Open-Sources Harrier Embedding Model, Tops MTEB
1+ mon, 1+ week ago (176+ words) Microsoft's Bing team has open-sourced Harrier, an embedding model family that tops the multilingual MTEB v2 benchmark under an MIT license. Microsoft's Bing team has released Harrier, an open-source embedding model family that ranks first on the multilingual MTEB v2 benchmark, outperforming…...
Anthropic Restricts New Mythos AI Model As it Finds Thousands of Zero-Day Vulnerabilitiess
1+ mon, 1+ week ago (319+ words) Anthropic has restricted its Claude Mythos AI model to a 12-company security consortium after it found thousands of zero-day vulnerabilities. Anthropic is restricting its highest-performing AI model to a consortium of security partners after Claude Mythos identified thousands of zero-day…...
Arcee AI Launches 399 B Top-Performing Open-Source Model at 96% Lower Cost
1+ mon, 1+ week ago (404+ words) Arcee AI has released Trinity-Large-Thinking, a 399 B open-source reasoning model under Apache 2. 0, offering 96% lower costs than proprietary alternatives. San Francisco startup Arcee launched on Thursday an open-source reasoning model that approaches the performance of systems from labs hundreds of times…...
AI Agents: Meta's DGM-Hyperagents Speed Up Self-Improvement with Additional Optimization Layer
1+ mon, 2+ week ago (586+ words) Meta has published research on DGM-Hyperagents, AI systems that edit their own improvement process and transfer learned strategies across unrelated domains. Traditional AI improvement follows a layered approach where a task agent solves problems and a separate meta-agent improves it....
Anthropic Data Leak Reveals Upcoming Mythos AI Model
1+ mon, 3+ week ago (552+ words) Anthropic has confirmed its Claude Mythos model after a CMS misconfiguration exposed 3, 000 internal documents, calling it a step change in AI reasoning. According to Fortune, which broke the story on March 26, a default setting in Anthropic's content management system made…...