Introduction to Existential Risks from AI

​Experts predict AI systems could surpass human abilities by 2030. We don't yet know how to make them safe.

​Join Cambridge AI Safety Hub (caish.org) for a talk on AI existential risk.

​What we'll cover:

• ​How capable could AI systems become, and when?

• ​Why is it so hard to align AI to what we actually want?

• ​How does misalignment present an existential threat?

• ​How can you spend your career on this?

​Spaces are limited.

Note the talk is in-person and it will not be streamed.