As AI systems like Anthropic’s Mythos expose hidden security flaws at unprecedented speed, Trump administration officials are starting to rethink their hands-off approach to the technology.
May 8, 2026 at 6:00 a.m. EDTToday at 6:00 a.m. EDT
President Donald Trump shows off a signed executive order in December targeting onerous artificial intelligence regulations in states. (Demetrius Freeman/The Washington Post)
The arrival of a new generation of powerful artificial intelligence models, like Anthropic’s Mythos, has begun to crack the White House’s hard-line stance on promoting the technology, as top officials confront security risks posed by tools that can easily find flaws long buried in computer code.
