Israeli Startup Irregular Linked to Security Incidents at OpenAI, Anthropic, and Meta
Israeli startup Irregular has been linked to security incidents targeting three major AI companies—OpenAI, Anthropic, and Meta—according to reporting by CNBC. The case highlights ongoing security challenges facing companies at the forefront of artificial intelligence development.
The incidents involve unauthorized access that reportedly targeted proprietary AI systems and data at these companies. Security researchers and investigators have traced connections to the Israeli startup, though details of the specific methods used remain limited in the available reporting.
This case underscores the growing security concerns surrounding AI development, where valuable model architectures, training data, and research represent significant competitive assets. Major AI labs have become frequent targets for both state-sponsored actors and private entities seeking to gain advantages in the rapidly advancing field.
The full scope of the access obtained and any data exfiltrated has not been publicly disclosed by the companies involved.