AI security is entering a new stage as enterprise organizations look beyond protecting models from prompt injection, unauthorized access, and data exposure.
Microsoft has launched Frontier Company, making a $2.5 billion bet that the next competitive battleground in artificial intelligence will not be foundation models, but helping enterprises put those models to work.
Microsoft has shut down EvilTokens, a cybercrime service that used AI to help attackers break into e-mail accounts, study their victims, and build convincing financial scams.
OpenAI has announced what it says is an AI-generated solution to the Navier-Stokes existence and smoothness problem, one of mathematics' seven Millennium Prize Problems, while Anthropic has reported separate advances in research mathematics and formal proof generation.