With frontier AI companies’ money on the line, a mutual insurer would enforce standards, peer review incidents, and pool safety research and development.

A regional court in Germany ordered Google to answer for apparent falsehoods about a Munich publishing house provided by its AI overview feature.

The First Circuit deals a legal setback to Puerto Rico’s attempt to regulate misinformation during states of emergency.

 The oncoming AI-driven flood of vulnerabilities demands changes to the U.S. process for retaining them for offensive cyber operations.

Congress must determine whether the proposed U.S.-Saudi nuclear energy agreement has adequate precautions.

Your weekly summary of everything on the site.

Nonforcible countermeasures can help stop Iran’s closure of Hormuz without undermining freedom of navigation.

A review of Mia Bennett and Klaus Dodds, “Unfrozen: The Fight for the Future of the Arctic” (Yale University Press 2025)

A CJEU ruling on driving-app data and online pornography quietly guts the EU's Section 230 equivalent—and the court owes Europe an explanation it hasn't given.

The answer to one of the world’s most dangerous security problems was drafted in Mexico City nearly sixty years ago.

State AI laws require developers to report safety incidents—but not what investigation follows or what evidence they must keep.

Models never released publicly may still fall under the EU AI Act. Companies cannot assume that what stays in-house stays out of scope.