This talk looks at philosophical questions that arise in the context of AI alignment. It defends three propositions. First, normative and technical aspects of the AI alignment problem are interrelated creating space for productive engagement between people working in both domains. Second, it is important to be clear about the goal of AI alignment. There are important differences between AI that aligns with instructions, intentions, revealed preferences, ideal preferences, interests and values, on an individual or collective basis. A principle-based approach to AI alignment has considerable advantages in this context. Third, the central challenge for theorists is not to identify ‘true’ moral principles for AI to align with; rather, it is to identify fair principles for alignment, that receive reflective endorsement despite widespread variation in people’s moral beliefs. The final part of the talk explores three ways in which fair principles for AI alignment could potentially be identified.
Iason Gabriel is a senior research scientist on the ethics research team at DeepMind. His work focuses on the intersection between Artificial Intelligence and Ethics, with a particular focus on the relationship between technology and human rights. Before moving to DeepMind, Iason taught ethics and political theory at Oxford University. He also worked for the United Nations in Sudan and Lebanon.
CITP Lunch Seminars are open to Princeton faculty, staff, and students only. Members of the public who would like to attend a particular talk should contact Jean Butcher at .
In an effort to support sustainability at our events, attendees are encouraged to bring reusable items for their personal use.
To request accommodations for a disability, please contact Jean Butcher, , 609-258-9658 at least one week prior to the event.