Security researchers at Microsoft, OpenAI’s largest investor, detected suspicious data extraction from the firm’s API in the fall of 2024, Bloomberg reported.