← All Issues
Monday, June 15, 2026

AI Daily

Your Automated Intelligence Briefing

Geopolitics

Unplugged Overnight: A Single US Letter Turns Frontier AI Into a Sovereignty Crisis

When Washington forced Anthropic to pull Fable 5 and Mythos 5 for foreign nationals, European and Indian leaders stopped treating it as a product outage. "A nation that depends on others for its technology is a nation that can be unplugged overnight," warned one French presidential candidate. From Mistral to India's new Varya model to a spike in decentralised-AI tokens, the scramble for sovereign alternatives is suddenly real.

Read full story →
AI Policy

The States Didn't Get the Memo: AI Regulation Moves Ahead Without Washington

Six months after Trump told states to stop regulating AI, more bills are being introduced than ever, many by Republicans. As the White House quietly negotiates a deal to preempt state laws, a coalition of attorneys general subpoenaed OpenAI on Friday, making the stakes concrete: the states are using the very authority Washington is trying to take away.

Read full story →
Hardware

Nvidia's RTX Spark Wants AI on Every Desk, Just as It Made Desks Unaffordable

Nvidia's new superchip crams data-center AI into a laptop and promises "unmetered intelligence to every desk." But it arrives in the middle of a memory shortage that Nvidia's own data-center boom helped cause, asking consumers to pay a premium its success inflated, for a future few of them asked for.

Read full story →
Industry

Meta Built Its AI Reputation on Openness. Its New Flagship Stays Locked.

Meta's AI chief says the old Llama playbook no longer works at the frontier. Its most capable model, Muse Spark, shipped closed over biological-risk findings the company says it cannot contain once weights are public. Open AI isn't dying, but it is being quietly pushed below the frontier.

Read full story →