Back to all research
RESEARCH / 01
AI safety
We research foundational technologies that help AI and agents operate safely and reliably, and take part in the real world with trust.
OUR EXPLORATION
Research direction
As AI takes part in everyday tasks and real work, safety must be present in every action. We research how AI can act in line with people's intentions, helping them understand their options and retain control over important decisions, so greater capability brings greater peace of mind. Our research asks whether suggestions are clear, actions are appropriate, and people can step in and change direction when a situation exceeds the AI's understanding.
POSSIBILITIES IN EVERYDAY LIFE01 02 03
Applications we are exploring
Studying how businesses can define task boundaries and responsibilities when working with AI.
Exploring how AI can seek human help when uncertain instead of acting prematurely.
CONTACT
Contact usDiscuss this research area
We welcome conversations about business partnerships, technology collaboration, and joint research.