How might we monitor models for deceptive alignment?

1 min read

Suggest changes in Google Docs

We don't have an answer for this question yet. Would you like to write one?

AISafety.info is a project founded by Rob Miles. The website is maintained by a global team of specialists and volunteers from various backgrounds who want to ensure that the effects of future AI are beneficial rather than catastrophic.

Get involved

Partner projects

Alignment Ecosystem Development

Aisafety.info is an Ashgro Inc Project. Ashgro Inc (EIN: 88-4232889) is a 501(c)(3) Public Charity incorporated in Delaware.

How might we monitor models for deceptive alignment?

In progress