AI security researcher

0Labs is building cybersecurity systems to protect society from increasingly capable and potentially misaligned AI models. The AI Security Researcher/Engineer will design and build realistic cybersecurity environments to evaluate offensive and defensive capabilities of frontier models, and study and help mitigate rogue AI deployments and loss-of-control scenarios.

Responsibilities:

• Design and build highly realistic cybersecurity environments. For example, energy grids, telecommunication networks, and cloud environments
• Use them to evaluate emergent defensive and offensive cyber capabilities of frontier models
• Study how models behave when they go rogue, evaluating and helping mitigate rogue internal deployments that lead to loss-of-control scenarios