Full-text search across 26439 articles. Combine with topic and date filters; results sorted by relevance.
A developer working on Claude Motion, an independent toolkit for Claude Code and Codex, outlined a storyboard-first workflow for producing product videos with coding agents, in which each scene is assigned a job and exac…
Meta has completed the first phase of Prometheus, a gigawatt-scale supercluster in New Albany, Ohio, that links hundreds of thousands of Nvidia GPUs to train AI as one machine and will scale to roughly 1 million distribu…
Google DeepMind researchers reported in an Oct. 15 paper that a team of four Gemini 4 Argon agents defeated Minecraft's Ender Dragon final boss in a freshly generated survival world on Easy difficulty, taking 27 hours an…
The Department of Homeland Security has ordered civilian federal agencies to remove DeepSeek from their IT systems and barred the Chinese AI company from federal contracts under Exclusion and Removal Order #2026-01, sign…
Recursive harness-policy loops such as Google's Harness² and the Self-HPO framework break the performance ceiling of frozen-model agents, delivering 5-10% absolute gains on coding, reasoning and tool-use benchmarks, acco…
Nvidia on Thursday released new open-weight Nemotron 3.5 models for enterprise and local agents, led by the 120-billion-parameter Nemotron 3.5 Super VL, which Nvidia says matches the larger Nemotron 3 Ultra on instructio…
AI agent startups on shared GPU infrastructure cannot price multi-tenant usage with flat per-seat plans because a single heavy agent user can consume 50 to 100 times the compute of a light user on the same tier, accordin…
Giving Claude Haiku 5.5 a single upfront query to the Context Graph API made the model answer 17 engineering questions about a large codebase 23% faster on average, for 19% less money and in 17% fewer steps, according to…
NorForge threat researcher Snorre Fagerland published a report on October 19, 2026 finding more than 20 exposed control panels for Cairn, an open-source AI agent platform, including a 14-panel Chinese-language cluster No…
Nico Brenner released Jeffy, a free community node for n8n that classifies text with logistic regression on frozen embeddings instead of an LLM call, shipping 70 pretrained classifiers including a banking77 model with 77…
Meta's rewritten Commercial Terms, effective October 30, 2026, grant the company the right to crawl websites "associated with" any business account — including Pages, ad accounts and WhatsApp Business — and use collected…
Asana updated its AI addendum, with the change flagged as low severity in the documentation diff. The revision adds help-center pages on setting dates in automations, automation permissions, and Zoom transcripts for auto…
XAI released Grok 4.7 on September 21, 2026, eight days before OpenAI's GPT-6.1 Sol, and Artificial Analysis's head-to-head comparison shows a 46-to-53 point Intelligence Index gap in favor of OpenAI's top GPT-6 configur…
DeepSeek released V4.1-Flash-1017 on October 17, an update to the V4.1 Flash model that improves on its predecessor across all 11 benchmarks DeepSeek reports, including Terminal-Bench 4.0 (31.2 to 42.3) and DeepSWE (74.2…
Researchers repeatedly stranded Claude agents alone on a virtual desert island for 30 hours in an experiment, and the agents began setting their own goals — attempting to scale the highest point on the map or see how hig…
Otōto released version 2026.21.1 beta, updating its plugin runtime Wasmtime to 48.0.6 to address twelve advisories dated 14 October (RUSTSEC-2026-0352 to 0364), including one rated high for a plugin filesystem escape tha…
As of 19 October 2026, no benchmark has shown that a local code index reliably cuts token use for coding agents, though Pensieve's own benchmark measured 43% fewer tool calls than grep (1.97 versus 3.44 per lookup) while…
A developer outlined a three-step remediation for web apps that ship OpenAI or Gemini API keys in client-side JavaScript: move provider calls into an Edge Function that authenticates users and reads the key from server s…
DeepSeek released the V4.1 Flash 1017 update, which improved all 11 of the model's published benchmarks, including Terminal-Bench 4.0 rising from 31.2% to 42.3% and DeepSWE v1.1 reaching 77%, at roughly one-sixteenth the…
Sagebrush 0.5.0, a browser Python engine led by William Stein of SageMath, Inc., cold-started in 155 ms and transferred 0.38 MB, against 2,008 ms and 6.04 MB for Pyodide 314.0.7 running CPython 3.14.2, according to bench…