Will taking annual MRIs of the smartest alignment researchers turn out alignment-relevant by 2033?
7%
chance

Get
Ṁ1,000
to start trading!
Sort by:

@yasmeen this one's for you

@ArnavBansal ha thanks! MRI alone wouldn’t be sufficient and assumes incomplete a priori knowledge with brain activity and any possible alignment output (based on what we know now)

I don't think MRIs are granular enough to even begin to implement the plan of "simulate the process of AI alignment researchers conducting such research".

© Manifold Markets, Inc.Terms + Mana-only TermsPrivacyRules