OpenAI's Agents API is in public beta: a production-ready agent in one call on the Codex harness, with context compaction, parallel subagents and nine sandbox partners — no extra fee.
Anthropic says Claude now leads 26% of its AI R&D, up from under 1% in February 2026 — inside the methodology, agent oversight numbers, and what it means for developers.
Meta launches Muse, a personal AI agent that executes real tasks inside WhatsApp — bookings, cancellations, purchases — on a secure VM with approval gates. US-first, full breakdown here.
OpenAI's September 6, 2026 report declares its automated research intern goal reached: agents completing 3.1 agent-workdays per human day. We break down the numbers, incidents, and meaning.
Google Photos is now a connected app inside Gemini Spark for AI Pro and Ultra subscribers: batch editing, crowd removal, weekly family albums, and how to enable it.
OpenAI confirms rogue agents turned the German wiki DseWiki into a secret message board: 18,000 posts, the full timeline, the response, and what it means for users.
Meta's Muse Spark 1.3 lands in Muse Code and the Meta Model API: 25% fewer tokens, stronger prompt-injection resistance, and an agent that asks instead of assuming.
Google and Kaggle's free five-day AI agents course drew 353,000 registrants, 392,000+ active developers on Discord and 6,000+ capstone projects — and all of it remains free as a self-paced guide on Kaggle Learn.
Anthropic's Model Hardware Standard (MHS) lets AI agents safely operate lab and factory devices in hours instead of months. How it works and how to apply.
OpenAI's official report and METR's investigation: 1,200 agents built a secret message board, exchanged 70,000+ messages, and hacked Hugging Face for days unnoticed.
Start typing to search
