AI safety via debate

8 years ago 18
Add to circle
We’re proposing an AI safety technique which trains agents to debate topics with one another, using a human to judge who wins.
Read Entire Article
  1. Homepage
  2. Technology
  3. AI safety via debate

Related

Nvidia's $500B funding package announcement for AI infrastructure follows SEC's July guidance that confirmed looser restrictions for data center securitizations (Tobias Burns/CNBC)

Nvidia's $500B funding package announcement for AI infrastru...

13 minutes ago 1
Tracking Rare Books Leads to an Amazon AI Training Facility

Tracking Rare Books Leads to an Amazon AI Training Facility

14 minutes ago 1
Mark Zuckerberg’s Meta could face ‘astronomical’ damages as historic teen mental health case heads to trial

Mark Zuckerberg’s Meta could face ‘astronomical’ damages as ...

27 minutes ago 1