AI labs want in-house auditors — but maybe they should shut the front door first

There may be a simpler and more effective fix for rogue agents, hiding in plain sight.

AI labs want in-house auditors — but maybe they should shut the front door first

TL;DR

  • A simpler fix for rogue agents may exist.
  • The fix is described as 'hiding in plain sight'.
  • This suggests an easily overlooked solution for managing AI.