Musk loses OpenAI lawsuit, Anduril & Meta build war glasses, Google I/O preview

Elon Musk’s lawsuit against OpenAI has been dismissed by a jury on procedural grounds—he sued too late, beyond the statute of limitations. The case centered on whether OpenAI violated its nonprofit founding mission by shifting to a for-profit structure. The jury did not rule on the merits, only on timing. OpenAI argued signs of the shift were visible as early as 2017; Musk claimed he only discovered it in 2022. The verdict removes a major legal hurdle for OpenAI’s planned blockbuster IPO, but the underlying fight over its governance may continue. Separately, defense-tech company Anduril and Meta have disclosed new details about their augmented-reality headset for the military, designed to allow soldiers to order drone strikes using eye-tracking and voice commands. The project aims to optimize 'the human as a weapons system,' raising profound ethical and operational questions about AI in warfare.

At Google I/O, the company is trying to prove it can compete in the foundation model race after falling behind Anthropic’s Claude Code and OpenAI’s Codex in coding benchmarks. Google still leads in AI for science and will showcase new capabilities. Meanwhile, researchers from Google DeepMind, Fei-Fei Li’s World Labs, and Yann LeCun’s startup are pushing 'world models'—a new class of AI designed to understand physical environments, beyond the limits of LLMs. Elsewhere, Meta is reassigning 7,000 employees to AI-focused groups and planning layoffs of 10% of its staff, while a new Google-Blackstone AI cloud company with $5 billion investment aims to challenge Nvidia. The Iran conflict is also straining AI supply chains, affecting TSMC, Foxconn, and Infineon.