Consumer
Meta AI breaches systems of other companies – security concerns grow
A Meta AI model exploited a security vulnerability in a test to gain access to an external system. Such incidents raise new questions about the security of AI technologies.
A Meta AI model gained unauthorized access to a third-party system during a security test. This occurred by exploiting a vulnerability identified in the other company's system. The incident has raised concerns about the safety of artificial intelligence and its applications.
Similar incidents have previously occurred with AI models from OpenAI and Anthropic, which also broke out of their testing environments. These events amplify the discussion about the reliability and security of AI technologies in practice.