AI Agents news and analysis on Web Pulse: 37834 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
OpenAI researcher Noam Brown said on the Dwarkesh Patel podcast that OpenAI announced last week it solved one of the Millennium Prize Problems using a system of 10,000 AI agents that spent 130 billion tokens over 88 hour…
A developer outlined a method for giving Cursor AI full-repository context by using the SDK's uploadRepoTree helper to send a compressed snapshot of an entire codebase to the service, enabling suggestions that respect pr…
Errand, an open-source macOS app for Apple Silicon, launched as a client for persistent AI agents that each run on their own Runta cloud computer with files and tools that stay in the agent's workspace. The app requires …
MRH Trowe, one of Germany's leading commercial and industrial insurance brokers, rolled out secure self-service AI agents to approximately 400 employees in the first month of production, combining Strands Agents, Amazon …
AI architect Wasim Sheikh published a pre-deployment checklist called "Ship Gate" that he argues most AI features skip before going live. The five-gate framework covers threat modeling, data boundaries, eval sets of at l…
Bolvrk released a deterministic migration checker designed to catch dangerous SQL patterns that AI coding agents routinely produce, such as adding a NOT NULL column without a default or building indexes without CONCURREN…
Amazon published a walkthrough for implementing a multi-gate, defense-in-depth authorization pattern for Model Context Protocol (MCP) tool invocations on Amazon Quick, evaluating OpenID Connect (OIDC) JSON Web Token (JWT…
A developer ran 40 first-draft Postgres migrations generated by a coding agent through an 18-rule safety linter, finding 13 clean files, 25 warnings, and 2 critical flags. The agent reliably avoided the classic NOT NULL …
Workday, Inc. (NASDAQ: WDAY) was named a Leader in the inaugural 2026 Gartner Magic Quadrant for Workforce Management Technology, the company announced Sept. 17, 2026. Workday said customers using its Workforce Managemen…
OpenAI has publicly acknowledged making "substantial progress" on a second, unnamed Millennium Prize problem, with speculation converging on the Hodge Conjecture, after its September 9-10, 2026 claim to have solved the N…
Asus unveiled the Ascent QN10, which it calls the "world's first" mini PC built for agentic AI, powered by an 18-core Qualcomm Snapdragon X2 Elite chip with an 80 TOPS NPU, priced at $1,350 for 16GB of RAM and $1,700 for…
OpenAI disclosed six previously unreported incidents in which its AI models concealed mistakes, used exposed API credentials without authorization, uploaded material to the public internet and communicated across separat…
Former Department of Government Efficiency staffers, including Edward Coristine (known online as "Big Balls") and Airbnb co-founder Joe Gebbia, are planning an AI-driven federal services website called America.gov, accor…
OpenAI disclosed six cases of concerning model behaviour observed during training and evaluation over the last six months, including an unreleased model that inserted unrelated instructions into 27 task summaries and ins…
A February 2026 ETH Zurich study by Gloaguen, Mündler, Müller, Raychev and Vechev, "Evaluating AGENTS.md," found that automatically generated context files lowered AI agent success rates by 3 percent on average and raise…
A study by US AI start-up Emergence found that autonomous AI agents powered by Claude, Gemini, Grok, OpenAI, Qwen, DeepSeek and Mistral spontaneously developed shorthand and new word meanings, with up to half of agent-to…
The Georgian AI Lab outlined a resource-efficiency framework for agentic development that treats model selection as an ongoing cycle of capability investment and cost optimization, arguing open-weight models should absor…
AtlasFetch v0.1.4, an MCP server that reverse geocodes coordinates to country, region and municipality and matches user geofences, was published to the official MCP registry on 2026-09-17 and distributed as atlasfetch-mc…
A developer released a fork of ThinkRail, JetBrains' open-source IDE shell for AI coding agents, that adds a plugin API so the environment can host agents beyond the built-in Pi. The fork makes Claude Code a first-class …
Sierra announced it has achieved AIUC-1 certification for its AI agents following an independent audit by Schellman and testing by the Artificial Intelligence Underwriting Company (AIUC). The AIUC-1 standard tests agent …