US federal agencies are reportedly bypassing a Trump-era ban to test Anthropic's advanced AI model, Mythos, for its hacking capabilities, according to Politico.
Why it matters
This article highlights the complex and evolving relationship between government agencies and advanced AI technologies. The reported circumvention of a ban suggests a perceived necessity or strategic advantage in utilizing cutting-edge AI, even with prior restrictions. It raises questions about national security, AI development, and the practical implementation of policy in the face of rapid technological advancement.
Some US government groups are testing a powerful AI from a company called Anthropic, even though there was a rule against it. They want to see if the AI can be used for hacking.
Anthropic is engaging with the Trump administration despite a Pentagon designation as a supply-chain risk, indicating ongoing dialogue between AI companies and government bodies.
Read on TechCrunch →MIT Technology Review's 'The Download' newsletter covers AI warfare's human illusion and a theory about our 'inner Neanderthal'.
Read on MIT Technology Review →Anthropic CEO Dario Amodei is set to meet with White House Chief of Staff Susie Wiles, signaling a potential resolution to a dispute between the AI startup and the Pentagon.
Read on Economic Times Tech →