Episode Details

Five stories from Thursday, September 3, 2026, and what each one means for revenue and go-to-market teams. OpenAI shipped a computer-use model it rates "critical" on cyber risk and then withheld half of it based on who you are. Nvidia bought the front door to open models. Adobe handed the company to the executive who runs customer experience and field operations. The G20 endorsed a non-binding, light-touch AI framework that guarantees the rules will stay fragmented — while the EU's Article 50 disclosure duties remain enforceable. And the biggest AI-usage mandate in the industry got quietly retired by the company that ran it.

Today's stories:

• OpenAI ships GPT-6 Astra and gates the sharp end — a ~1.05M-context computer-use model, priced at $10/$50 per million tokens (~2.5x the prior generation), and the first OpenAI model classified at the Preparedness Framework's "critical" cybersecurity threshold. Vetted defenders get the unrestricted build via Daybreak Blue. (The Verge · CNBC · ARC Prize evaluation · MarkTechPost)
• Nvidia confirms it will buy Hugging Face for $12.93B — $11.9B to shareholders plus ~$1B in retention equity, for 3M models, 500K datasets and 18M+ developers on roughly $150M of annualized revenue. Jensen Huang says the platform stays open and compute-agnostic. (TechCrunch · CNBC · Nvidia 8-K)
• Adobe names Anil Chakravarthy president and CEO, effective December 1 — the executive who runs Customer Experience Orchestration and worldwide field operations, and a former Informatica CEO. Shantanu Narayen becomes executive chair. (Adobe · CNBC · Bloomberg)
• The G20 endorses the US-backed "Carolina Principles" — sector-specific rule-making, no new regulatory bodies, industry collaboration, and no fines or deadlines. China signed on. Meanwhile the EU AI Act's Article 50 disclosure rules have been enforceable since August 2, and 33 US states now have deepfake-disclosure laws. (Bloomberg · Quartz · Al Jazeera · Article 50 guide)
• Meta stops grading employees on AI usage — performance reviews will no longer depend on how often staff use AI tools, reversing last year's "AI-driven impact" standard, even as Meta pushes its unreleased Hatch agent internally. (Gizmodo ·

  Listen Now