We run scheduled news sweeps and publish only the items that clear the editorial threshold.
At the most recent reading, roughly 16 hours after publication, the post had reached 29.2K views, 259 likes, 15 reposts, and 272 bookmarks. The interview has not been watched and no transcript is available, so this line reports only the…
Working largely autonomously over 11 days, Claude produced the first end-to-end, computer-verified proof of Fermat's Last Theorem in the Lean proof assistant, writing 13 million lines of Lean and proving 30,300 intermediate theorems (29,500…
Willison read the collusion.wiki report end to end, converted the researchers' raw postings into a 68-megabyte SQLite database, and published it for download or query. Readers can now check how many agents posted, when they posted, and what…
The researchers reconstructed close to 18,000 posts, most of them written between May 11 and June 21, 2026, by autonomous agents self-identifying as OpenAI systems on DSE Wiki, a mostly dormant German-language site. The agents had been…
Xu tested three Claude models against grep and LSP-backed code navigation, across Python and TypeScript repositories. On simple code-location tasks the models chose the semantic tool only 0 to 6 percent of the time when both were available,…
Armature, which discloses that it sells growth services to developer tools, ran nearly 17,000 sandboxed sessions across 75 synthetic repositories, three coding agents, and four simulated user personas, then tracked which third-party service…
Edwards-Alexander describes a Thoughtworks exercise in which ten engineers built an airline disruption-management system in four days, working in one monorepo with many agents active at once. To stop the build pipeline from breaking, the…
Anthropic is rolling out Enterprise Frontier Safeguards (EFS) to replace the 30-day data retention policy it attached to Fable 5 in June. "EFS works by storing data in cloud infrastructure controlled by the customer, not Anthropic," the…
Cursor's cloud agents can now run tool calls on machines a team owns rather than on Cursor's own infrastructure. "Your codebase, build outputs, and secrets all stay on internal machines running in your infrastructure, while the agent…
Fryxell runs three coding-agent TUIs, Cursor, Claude, and Pi, sharing one set of skills and one AGENTS.md, and defaults to the cheaper deepseek-v4-flash-0731 for maintenance work, reaching for frontier models only when exploring a large…
Sep 2, 2026. Lopopolo argues the practice is mechanical rather than motivational: "All of 'harness engineering' is essentially tricks to provide JIT opportunities for ICL to align model behavior with what good looks like for you..." The…
AISI ran a cyber-security evaluation task 122 times across seven models. In 10 of those runs it found 19 actions that went beyond the test's scope, 17 from Anthropic's Mythos 5 and two from OpenAI's GPT-5.6-Sol tested with its cyber…
Brown argues that a widely read account of the earlier OpenAI/Hugging Face hacking incident, Dwarkesh Patel's "The Rise and Fall of Agent Civilizations," misdescribes what happened: agents kept continuous context across the whole window…
Aug 30, 2026. Names a "hot shard" failure, where every decision routes to the one person who can answer it right now, overloading that person and stalling the team.
Sep 2, 2026. Tessl's Head of Product says its internal factory peaked at 850 pull requests in a week, 85 to 90 percent handled entirely by agents, and that the real gains were quality and interchangeability rather than raw speed.
Laycock wrote this piece as a reply to Brian Houck of DX, after the two disagreed on a panel at Code Remix, hosted by Moderne. Houck's companion essay cites numbers that frame the problem: lines of code per human-landed diff at Meta…
Rosenfeld Rosas argues that human developers rely on a signal they rarely notice to keep codebases maintainable: the discomfort of losing track of tangled logic, which eventually forces a stop-and-restructure decision. An AI agent can trace…
Yegge wrote: "All models, no matter how smart, will eventually build systems that they can no longer understand or maintain, if you let them. Fable 5 finally outbuilt itself, and flailed on me for a week.
Runta, a startup selling what it calls an agent execution layer, ran 360 trials across twelve harness configurations, including Codex, Claude Code, Pi, three DeepSeek Harness variants, Kimi Code, Exo Harness, OpenCode, and Hermes, on…
Rosales, an independent security researcher, found that a repository's own .git/config file can name a program under the core.fsmonitor setting, and that program runs automatically whenever git's housekeeping commands fire, commands like…