JUST IN: Anthropic found itself at the center of a technological and geopolitical debate after stating that its new Mythos model is too dangerous to release immediately, even as its annualized revenue growth rate reportedly climbed to $30 billion.

Meanwhile, investors and analysts debated whether the company is acting prudently or repeating a fear based marketing strategy, amidst tensions with OpenClaw, the race for AI agents, and the noise generated by the Iran-Israel truce.
Anthropic asserted that Mythos detected thousands of vulnerabilities in critical systems, including long standing flaws in OpenBSD, FFmpeg, and the Linux kernel.