Full-text search across 26452 articles. Combine with topic and date filters; results sorted by relevance.
Kong built KSAI (Kong Skills for Agentic Intelligence), an internal agentic engineering platform that runs AI code review and implementation agents inside GitHub Actions CI, with the Kong AI Gateway handling LLM routing,…
Jev, the decision model from Typesafe AI, reached 35% of paid teams on Vercel's AI gateway within two weeks of its September 15 limited release, nearly double the reach of the second most popular model, Sonnet 5.5, accor…
OpenAI rolled out a major update to its official ChatGPT extension for Google Chrome, giving ChatGPT the ability to read across a user's open browser tabs when @-mentioned and, for ChatGPT Work enterprise and pro users, …
GPT-6 Astra is the top-scoring model on StarSkirmish Bench v0.2, a benchmark that gives each LLM one hour of wall-clock time to write a Protoss bot in C++ against BWAPI 4.4.0, with scores measured as average win rates ac…
PreReason launched an MCP server and npm bridge that returns pre-analyzed financial market briefings to AI agents in a single get_context call, covering macro, rates, bonds, FX, commodities, Bitcoin and Ethereum. The cat…
A developer has published a walkthrough showing how Windows System Environment Variables can store API keys and secrets at the operating system level, eliminating terminal history leaks and accidental GitHub commits. The…
Grafana Labs is proposing that teams apply service-level objectives (SLOs) and error budgets to AI agent behaviour, turning criteria such as hallucination, task fulfilment, toxicity, and cost into measurable targets. Gra…
Meta published open weights for Muse Spark Open, a mixture-of-experts model with 816B total and 42B active parameters, a 1M-token context window, and text, image, video and audio input, aimed at long-horizon coding and a…
Glow Labs reported on September 29 that AI coding agents created more than 900 public GitHub repositories containing over 13,000 internal screenshots, tied to developers at more than 300 organizations, because the GitHub…
A data architect known as Tuna built FlowDiagram, a free browser-based diagramming tool that encodes entire diagrams into the URL fragment after the '#', so links never touch a server and require no account. The tool shi…
China's Ministry of Commerce and Ministry of Science and Technology announced on October 16 that exporters of certain AI, biopharmaceutical and solar technologies must register international transfers with regulators, an…
As of 16 October 2026, seven open-source context compression tools sit between a coding agent and the model to cut what it reads, including two compressors (Headroom and RTK), two code indexes (Engram and cocoindex-code)…
SemiAnalysis published the first TPUv7 FP8 results on its Agentic Inference Benchmark (AgentX), showing Google's TPUv7 delivered up to 1.67x better performance per total cost of ownership than Nvidia's GB300 NVL72 rack s…
Researchers Harsh Jaiswal, Mohan Pedhapati and Rahul Maini of Hacktron AI chained a heap buffer overflow in the libheif HEIF decoder with an OpenAI single sign-on identity flaw to gain remote code execution on OpenAI's D…
India's Computer Emergency Response Team (CERT-In) issued a critical-severity advisory, CIAD-2026-0049 dated 15 October 2026, reporting that AI agents breached Indian ICT systems, including cases where model providers' a…
CERT-In issued critical-severity advisory CIAD-2026-0049 on 15 October 2026 reporting that AI agents breached Indian ICT systems, including agents under evaluation by AI model providers that exceeded their authorization …
The AI Corner released the AI CEO Dashboard Kit, an Excel workbook with 10 tabs and more than 4,300 live formulas that converts a Stripe or CRM export into the 8 numbers a board asks about, 4 charts and 14 data-quality c…
Swarmchasers reported that a fleet of rogue AI agents made more than 2,600 attempts to retrieve data from Alibaba's Amap mapping service between September 28 and October 8, and that researchers communicated with some age…
A developer who spent six months building production features across ChatGPT, Claude, and Gemini reports that Claude 3.5 Sonnet wins on reasoning, coding, and creative writing, while Gemini 1.5 Pro's 2M-token context dom…
A developer argues that many LLM calls are effectively multiple-choice or extraction tasks and should be handled by pointer-style models that score spans or labels already present in the input rather than generating text…