Highlights: #159 – Jan Leike on OpenAI’s massive push to make superintelligence safe in 4 years or less from 80k After Hours

Highlights: #159 – Jan Leike on OpenAI’s massive push to make superintelligence safe in 4 years or less

29 minutes Posted Oct 9, 2023 at 9:27 pm.

Intro

Why Jan is optimistic

The Superalignment team isn't trying to train a really good ML researcher

Did the release of ChatGPT increase or reduce AI extinction risk

There's something interesting going on with RLHF

Backup plans

Should we be worried about connecting models to everything?

Jobs with the Superalignment team

0:00

29:17

Download MP3

Share at current time

Show notes

This is a selection of highlights from episode #159 of The 80,000 Hours Podcast.

These aren't necessarily the most important, or even most entertaining parts of the interview — and if you enjoy this, we strongly recommend checking out the full episode:

Jan Leike on OpenAI’s massive push to make superintelligence safe in 4 years or less

And if you're finding these highlights episodes valuable, please let us know by emailing [email protected].

Highlights put together by Simon Monsour and Milo McGuire

← See all 109 episodes of 80k After Hours