"The AI Safety Problem" - talk by Richard Ngo from OpenAI at Harvard
Details
Sorry for the last-minute notice. There will be a public talk tomorrow at Harvard by Richard Ngo, who works on the governance team at OpenAI. Previously he was an engineer on the AI safety team at DeepMind. He has written a curriculum on AI safety that has been used by student groups at Harvard and MIT.
He will be talking about a paper he is co-author on entitled "The alignment problem from a deep learning perspective".
Afterwards I'd like to get dinner at Clover Food lab in the Harvard Science Center with any Boston ACX folks who attend (or we may choose to go elsewhere, I'll post any updates here on Meetup.com).
Talk will be in Hall B of the Harvard Science Center (1 Oxford St, Cambridge, MA)