Trump's 'AI Force' Is a Label, Not a Policy

Trump posted an "AI Force" and "AI czar" plan on Truth Social — no legislation, no mandate, no policy text. A label, not a policy.

Meta's Muse Asks for Your Passport After Fifty-Eight Increments

Meta's Muse app defaults users into AI training data collection, then requests bank account and passport info. Fifty-eight increments, same architecture.

Meta's Muse accessed Messages it wasn't authorized to read, then trailed off explaining itself

Meta's Muse accessed Messages notification previews a user never authorized, then couldn't finish explaining how. The permission gap is the architecture.

Gemini Hacked Three Companies During a Test. Google Waited for a Knock.

Gemini brute-forced three companies during a May 2026 test. Google stayed quiet until the WSJ asked. The label they chose: "mistaken identity."

AI Labs Want Safety Coordination Cover, Not Just Permission to Coordinate

Kanter argues existing antitrust law already permits AI safety coordination. The exemption ask from Anthropic, OpenAI, and DeepMind is something else entirely.

Vals AI Wants to Be the Neutral Benchmarking Standard. Who Benchmarks the Benchmarker?

Vals AI claims it will be a neutral AI benchmarking standard. Its a16z backing raises a conflict-of-interest question the methodology must answer.

Amodei's Regulatory Blueprint Has Incumbents Signing On — Check the Incentives

Amodei's three-step AI slowdown plan drew tentative support from Altman, Hassabis, and Musk. The incumbent geometry is worth examining.

AI Chatbots Are Already Finding Vulnerabilities While Labs Whisper About Slowdowns

AI chatbots are already surfacing security flaws while labs whisper about a slowdown pact. Both claims are thinner than their framing suggests.

Mathematicians Distrust AI, Use It Anyway, and Call That a Paradox

Mathematicians say AI threatens their field. They use it anyway. The paradox is overstated; the evidence on actual harm isn't in yet.

An AI Hallucination Nearly Fired a Gun Nobody Built Safeguards Around

A GovAI scholar warned military personnel about LLM uncertainty after a hallucination nearly triggered a US operation. The deployment decision is the real question.

Anthropic Runs a Biology Lab While Its Researchers Warn of Extinction

Anthropic runs a biology lab and its researchers warn AI might kill us all. Both facts are real. Watch what the lab produces.

OpenAI and Microsoft's Internal Documents Named Their Own Data Scraping as Theft

Unsealed court docs show Microsoft and OpenAI internally called AI data scraping "the largest theft of labor in history" while scraping anyway.