Protect AI users by designing and refining safeguards that keep powerful models honest, safe, and aligned with real-world risks and ethical challenges.
Protect AI users by designing and refining safeguards that keep powerful models honest, safe, and aligned with real-world risks and ethical challenges.