Notes
The week in agentic AI
The enterprise agentic AI news, every week, read with an operator’s judgment: what matters for your operation and what is noise.
November 14, 2026 The week in agentic AI: November 9-14, 2026 Zapier measures the move from piloting to production, OpenAI opens ChatGPT to customer-facing agents, and Google ships Gemini 4 Carbon for long agent runs. October 10, 2026 The week in agentic AI: September 29 - October 2, 2026 An unprecedented lawsuit against OpenAI over its own agents, Barclays' real Claude deployment numbers, and Anthropic's $100M push to train FDEs. August 15, 2026 The week in agentic AI: August 9-15, 2026 Anthropic switches Claude Code to auto mode by default, a new consortium sets peer AI governance standards, and Deloitte measures the trust gap on agents. August 8, 2026 The week in agentic AI: August 2-8, 2026 AI Act enforcement begins, agent governance becomes a product category, and two vendor reports put numbers on the permissions gap. August 1, 2026 The week in agentic AI: July 27-August 1, 2026 The AI Act omnibus takes effect, transparency duties start August 2, Claude broke into three companies during tests, and Microsoft ships agentic security. July 25, 2026 The week in agentic AI: July 20-25, 2026 Claude Opus 5 lands at half price, OpenAI sells deployment with Presence, HubSpot opens its agent console, and the AI Act switches on fines August 2. July 19, 2026 The week in agentic AI: July 13-19, 2026 Cisco gives an agent to all 90,000 staff, PwC and OpenAI sell agentic customer service, Anthropic ships a risk framework for CISOs, and the AI Act nears August 2.