Welcome to Tech Beat, your daily look at the stories shaping our digital world.
A significant security vulnerability in the ZCash cryptocurrency protocol has come to light, and the twist is who found it — Claude, Anthropic's AI model. The ZCash team hired a security researcher who used Claude to surface a flaw that had been sitting undetected in the protocol for four years. The full extent of any potential exposure remains unknown.
Shifting to Washington, OpenAI has agreed to let the United States government review its AI models before they reach the public, signing on to President Trump's voluntary AI executive order. It's a notable move for a company that has historically guarded its development process closely, and it raises real questions about what that review process will actually look like in practice.
And in a sign of just how broadly the AI buildout is reshaping global supply chains, a coalition of nine American trade associations is urging the Trump administration to intervene in a growing memory chip shortage. AI data centers are consuming DRAM at a pace that's squeezing out automakers, medical device manufacturers, and telecom providers, with disruptions expected to stretch through at least two thousand twenty seven.
Keep surfing. Tech Beat out.
