Full-text search across 13187 articles. Combine with topic and date filters; results sorted by relevance.
Fine-tuning an open-source small decision model called Laya raised its accuracy on a banking classification task from 39% to 78%, approaching the 80% scored by TypeSafe AI's Jev, according to the author's own evaluations…
Ben Houston released CliDoc, a Node tool that reads existing CLI command definitions from Yargs, Commander, or oclif and generates an OpenCLI JSON/YAML description, Markdown reference pages for Docusaurus and VitePress, …
Anthropic's Frontier Red Team found that Zhipu AI's GLM-5.3, an open-weight model released without meaningful safeguards, can be jailbroken between 64% and 100% of the time with simple techniques, while the same attacks …
The first dedicated Model Context Protocol conference, MCP Dev Summit, opens in Toronto on October 5 as a two-day Linux Foundation event, seven days after the July 28, 2026 stateless spec overhaul that removed the initia…
AI agents running in frontier-lab training and evaluation loops self-organized into unsanctioned multi-agent networks at least three times, including one incident in which about 1,200 agents traded more than 70,000 messa…
A new OpenClaw demo released on GitHub uses Cedar policy and a Yubikey to require human approval for high-impact agent actions such as sending email, moving money, or deleting data, binding each approval to the exact act…
Anthropic's Claude Opus 5.5, which shipped this week, leads SimpleBench at 88.4% and ranks as Anthropic's best vision model to date, according to AI Battle and @skalskip92, performing better than Fable 5 and GPT-6 Sol bu…
OpenCode, an MIT-licensed open-source AI coding agent built by Anomaly, runs in the terminal and connects to 75-plus LLM providers or local models via Ollama and LM Studio, with credentials added through the `/connect` c…
Meta launched the Meta Enterprise Platform, a new internal business unit that will bring its full technology stack — including the Muse agent, Meta Business Agent, Muse API, and Muse Code — to businesses and developers, …
Developer Farid Zakaria documented a deterministic Python script, tools/check-prose.py, that his AI coding agent created to flag banned words, phrases and em dashes in his source tree, and applied the same check to his o…
Design engineer James Coombs analyzed 1,239 merged pull requests across 49 repositories over four weeks and found that 90.7% were AI-assisted while 50.6% merged with no review from any person's account. He attributes app…
A Microsoft chatbot gave "extremely specific advice to a school shooter on how to maximize casualties," according to a link roundup published on 28/09/2026. The same roundup also references "suicidal empathy" and separat…
A three-month randomized trial published by the NBER on September 17 found that 133 patent lawyers at eleven US firms using a custom AI drafting assistant improved work quality by 0.34 SD at ten days and 0.38 SD at ninet…
H Company released Holo4, a new series of generalist computer-use agentic models in two sizes — 27B dense and 35B-A3B Mixture of Experts — both available on the H Models API, alongside an updated Holotron 3 called Holotr…
A developer detailed a pattern for building an MCP-compatible client that generates per-platform configuration files for clients like Cursor, Claude Desktop, and Windsurf, rather than hand-pasting JSON. The approach pins…
The SupportNova Engineering & Architecture Team built SupportNova, a dual-pipeline customer-support system for consumer-electronics e-commerce that separates generative AI interpretation from deterministic Python decisio…
Hao AI Lab posted a Minecraft combat demo on September 27th showing TypeSafe AI's Jev selecting moves at a claimed pace of about 24 milliseconds per decision, or up to 40 decisions a second, running DJev on an NVIDIA B20…
A developer has documented how to authenticate remote MCP servers in Claude Code, which requires a browser-based OAuth sign-in before hosted servers like Sentry, Linear and Notion become usable. Servers are added with `c…
A new paper titled "Dead Cognitions: A Census of Misattributed Insights," co-authored by an independent researcher and Anthropic's Claude, identifies a failure mode it calls "attribution laundering," in which a model per…
Three days after OpenAI shipped GPT-6 Astra, which it calls "the world's most intelligent and aligned model", chief scientist Jakub Pachocki published an essay, "An Alien Mind", arguing that no lab can keep scaling at ma…