Anthropic said China and Russia attempted to use Claude AI for cyberattacks, weapons design, propaganda and mass surveillance. The company identified the alleged attempts as part of broader concerns over misuse of advanced AI systems for security and influence operations.