(Image credit: Getty / Bloomberg) According to a report by The Economist last week, Anthropic’s powerful Mythos AI model was able to break into “almost all” classified systems belonging to the National Security Agency (NSA) — one of the highest-ranking and most powerful intelligence agencies in the U.S. government — within hours during a controlled security evaluation. The claim came from Sen. Mark Warner, vice chair of the Senate Intelligence Committee, who said Gen. Joshua Rudd, the head of the NSA and U.S. Cyber Command, briefed him on the model’s capability. “(This tool) broke into almost all of our classified systems, not in weeks, but in hours,” Rudd reportedly told Warner, as cited by The Economist in a June 14th report that initially went under the radar.…