THE DAILY BRIEF

Anthropic told shareholders it will report adjusted operating income this quarter, marking a second consecutive profitable period as the company prepares to go public.

► An AI lab achieving profitability validates the business model and removes a major IPO risk factor for what could be the highest-valued AI startup to date.

Apple is shipping a rebuilt Siri powered by Google's Gemini models running partly on-device and partly through Private Cloud Compute, though EU users won't get access due to regulatory constraints.

► The dependence on competitors' models and geographic fragmentation highlights how regional regulation is fragmenting consumer AI experiences and forcing Apple to outsource core capabilities.

AI HALLUCINATION NEARLY TRIGGERED MILITARY CONFLICT

An AI-assisted intelligence report that circulated across US military chains this spring falsely claimed a Chinese ship in the Middle East was carrying nuclear weapons components, nearly escalating tensions.

► Real-world consequences of AI hallucinations in high-stakes decision-making expose critical gaps in validation workflows and the risks of deploying AI in military contexts without proper safeguards.

Anthropic's revamped Claude Code Projects feature now allows users to run multiple AI agents with shared memory, goals, and artifacts, executing different tasks in parallel across threads.

► Agent orchestration becomes a core product feature, signaling that multi-agent systems are moving from research prototypes into practical production workflows.