Home Artificial Intelligence Frontier risk and preparedness

Frontier risk and preparedness

Frontier risk and preparedness

To attenuate these risks as AI models proceed to enhance, we’re constructing a brand new team called Preparedness. Led by Aleksander Madry, the Preparedness team will tightly connect capability assessment, evaluations, and internal red teaming for frontier models, from the models we develop within the near future to those with AGI-level capabilities. The team will help track, evaluate, forecast and protect against catastrophic risks spanning multiple categories including:

  • Individualized persuasion
  • Cybersecurity
  • Chemical, biological, radiological, and nuclear (CBRN) threats
  • Autonomous replication and adaptation (ARA)

The Preparedness team mission also includes developing and maintaining a Risk-Informed Development Policy (RDP). Our RDP will detail our approach to developing rigorous frontier model capability evaluations and monitoring, making a spectrum of protective actions, and establishing a governance structure for accountability and oversight across that development process. The RDP is supposed to enrich and extend our existing risk mitigation work, which contributes to the security and alignment of recent, highly capable systems, each before and after deployment.


Please enter your comment!
Please enter your name here