GOOD: Hegseth Says US and China Agreed to Keep Talking About AI Guardrails.

Trump and President Xi Jinping discussed guardrails on AI and how to prevent bad actors from exploiting the most powerful AI models, such as Anthropic’s Mythos, which has exposed major software security vulnerabilities.

Hegseth said the two countries agreed to continue talking as technology develops.

“Guardrail conversations are productive between two strong countries,” Hegseth said, “but it’s also our job to run the fastest, and certainly at the War Department, we’re trying to do everything we can to maintain that.”

If we learned anything from The Terminator or WarGames, it’s to keep a human in the loop.