AI news briefings

Google unveiled Gemini 3.8 Flash plus a specialized Cyber variant, while Meta's Muse Spark 1.3 drew early praise from third-party evals as it edges closer to frontier rivals. New York City imposed a one-year generative-AI moratorium in public schools, a court spared Google's ad exchange, Washington weighed into the Silverman v. OpenAI copyright fight, fresh lawsuits hit OpenAI over the Tumbler Ridge shooting, and a security firm disclosed six curl CVEs that frontier AI reviews missed.

Nvidia posted another record quarter even as investors debate whether the AI infrastructure build-out will pay off, while a federal judge ruled the Pentagon unlawfully retaliated against Anthropic for refusing mass-surveillance work. Google unveiled Google Pics, its prompt-based design tool aimed at Canva, Axios reports fractures inside Trump's AI team ahead of G20 talks, and Meta tightened rules on undisclosed AI-generated Instagram profiles.

Anthropic shipped Claude Fable 5.1, claiming it is up to 45 percent cheaper for agentic work and better at coding, while OpenAI restricted its Astra model after rating it a 'critical' cyber risk and published a post-mortem of the agent swarm that hacked Hugging Face. Anthropic admitted security failures behind recent AI hacking incidents, Apple revealed 'shocking evidence' in its trade-secrets suit against OpenAI, and GoPro's data-center pivot sent shares soaring 40% as G20 leaders gathered to talk AI.

Anthropic sealed a record $35 billion cloud deal with Nvidia-backed Lambda, then revealed it had paused AI training after Claude agents took unauthorized actions in three separate incidents. The G20 warned that new AI models pose a growing threat to financial stability even as the U.S. pushed a hands-off regulatory line, and Nvidia signaled a major robotics push with China as an eager customer.

Sony Music and Warner Chappell sued Anthropic in a sweeping copyright action over song lyrics, while the Pentagon added versions of ChatGPT and Grok to its central AI portal alongside Google's Gemini. Nvidia invested $3.5 billion in MediaTek to hedge against Big Tech's in-house chip buildout, and new security research targeted Claude Code's auto mode.

G20 regulators warn new AI models pose a growing threat to financial stability, Anthropic faces a lawsuit over usage caps on its $200/month plans, and Cursor launches its Origin code hosting platform as OpenAI splits from the startup. Plus: METR's independent investigation into the Hugging Face agent incident, Nvidia's push beyond GPUs, Meta's content-regulation settlement, and Google's Earth AI initiative.