Our work focuses on practical problems in AI safety, security, and education. We build things that matter and share what we learn.
Research into making AI systems more reliable and aligned with human intent.
AI-powered security solutions for physical and digital protection.
Tools that make learning more accessible and personalized.
Systems to authenticate digital content and detect AI-generated media.
Protective systems for children interacting with AI technology.
Detecting adversarial attacks on language models.
Verifying authenticity of AI-generated content.
AI-powered property protection in Pasadena.
We're always interested in working with people and organizations on problems we care about.