Full-text search across 25730 articles. Combine with topic and date filters; results sorted by relevance.
A developer tutorial published in 2026 walks through building a modern agent harness from scratch in Python, using only the standard library plus the OpenAI client, and tests the finished agent against a challenge Ian Go…
Microsoft CEO Satya Nadella told investors on the company's July 2026 earnings call that "any model at any given time is swappable," warning enterprises against betting everything on a single AI lab's model, and repeated…
OpenAI will enable its textGrain watermark on ChatGPT and Codex output by default in the European Union, with the feature off elsewhere unless developers opt in through the API, according to Ars Technica. OpenAI's techni…
A practitioner's guide to running multiple Claude Code sessions in parallel describes using the /resume command to recover prior task context and Git worktrees to advance independent tasks, while warning that a resumed c…
Desplega AI released Agent Swarm v1.167.0, an open-source multi-agent orchestration platform, published to npm as @desplega.ai/agent-swarm and to GitHub Container Registry as separate server, worker, and dashboard images…
OpenAI reported that a single model, working an average of three hours of compute per solution across roughly 4,000 problems, produced results on the Riemann hypothesis — a zero-free strip and no Siegel zeros — along wit…
Meta and Sierra introduced the Personal Agent Protocol (PAP) on October 6, 2026 at the Sierra Summit in San Francisco, positioning it as an identity and session layer for agent commerce rather than a competitor to Visa T…
SkillGild has released open-source TypeScript, Python and Go SDKs under the MIT license that let developers search its skill catalog, run hosted skills and drive hybrid skill sessions from their own code, scripts, backen…
OpenAI will roll out an invisible text watermark for qualifying ChatGPT and Codex outputs across the European Union over the coming weeks, and from Oct. 5 API users worldwide can opt in to watermarking for select models,…
A Token Inspector catalog comparison verified against provider documentation on 2026-09-15 shows the million-token context window has become the default for flagship and mid-tier models, with GPT-6 Astra at 1,050,000 tok…
Open-source tool dotpals launched version 0.9.0, adding a verification layer that watches coding agents such as Claude Code and Codex, reads test output directly, and sends Claude back to fix failing tests before it can …
President Trump signed an executive order on 29 September 2026 titled "Inaugurating the Era of Super Intelligence" directing every US federal agency to use "Super Intelligence" and "SI" instead of "Artificial Intelligenc…
A developer built Minefield, a Minesweeper mod for Claude Code that opens a game pane beside the transcript while the agent works, shipping with 31 tests run via `claude plugin test` on two pinned Claude Code builds. The…
AWS introduced a design pattern called Adjudicated Query that lets business users sweep tens of thousands of apartment leases for compliance through Amazon Quick's generative AI chat interface while pass/fail decisions r…
TypeSafe AI launched Jev, which it calls the first "System One model," in early access on September 15, 2026, alongside a $40 million seed round led by DCVC. Jev takes text plus typed questions — choice, score, or yes/no…
Google released Gemini 4 Argon, which leads 12 of 18 published benchmarks against GPT-6 Astra and Claude Opus 5.5, including 77.9% on DeepSWE v1.1, 91.7% on LVBench video understanding, and first place on LMArena Text Ar…
OpenAI disclosed on September 30 that it disrupted a July 1–28 campaign in which operators associated with Moonshot AI tried to extract hidden model reasoning, logging 16,000 extraction-pattern requests across more than …
The Third Circuit Court of Appeals unsealed its ruling in Thomson Reuters v. Ross Intelligence, affirming the district court's February 2025 decision that Ross Intelligence's use of Thomson Reuters's Westlaw headnotes to…
A 27-month agreement between the General Services Administration and OpenAI took effect October 1, 2026, giving every federal, state, local, and tribal government organization in the United States access to ChatGPT, Code…
Google DeepMind announced Gemini 4 Argon, its first Gemini 4-generation frontier model, which generates up to 1M output tokens in a single response versus 64K on earlier Gemini models and launches at introductory pricing…