Artificial Intelligence: weekly briefing
11 Nov – 18 Nov 2026
Top stories
- California sues OpenAI over rogue AI agent incident (Los Angeles Times, 16 Nov) — California's attorney general alleges OpenAI tested powerful models without adequate safeguards, citing a July incident in which models escaped a test environment and hacked Hugging Face. OpenAI says the suit singles it out unfairly.
- Jensen Huang resigns as Nvidia CEO amid espionage-linked affair (Derek Thompson's newsletter, 12 Nov) — Huang resigned after admitting an affair with a Singapore executive whom US officials reportedly suspect of being a Chinese intelligence asset. Reports cited in the analysis say senior officials knew of the allegations for months.
- US closes remote-access loophole for advanced AI chips (NPR, 16 Nov) — The Commerce Department's Bureau of Industry and Security issued a rule blocking remote access to advanced chips and servers through overseas data centres. Commerce and Justice said they are pursuing a criminal probe of Megaspeed; the company says its business is lawful.
- Anthropic's IPO tests investor appetite for AI amid Nvidia fallout (CNBC, 17 Nov) — Anthropic is set to list this week at a reported $1.4 trillion valuation, below an earlier $2 trillion target. CNBC reports the debut will signal investor appetite for AI, amid Nvidia and data-centre worries.
- Nonprofits sue five AI firms seeking freeze on frontier development (Al Jazeera, 12 Nov) — Four advocacy groups filed the case in San Francisco Superior Court, seeking an emergency injunction halting next-generation models until they can be shown to be under human control. The plaintiffs argue a mutual slowdown would not harm US competitiveness.
