Don’t Trust AI’s Self-Reports
1:26:07–1:28:37 · 150s
Cartmill argues LLMs will always give fluent answers, so we should borrow animal-cognition methods to test real understanding beyond verbal output.
1:26:07–1:28:37 · 150s
Cartmill argues LLMs will always give fluent answers, so we should borrow animal-cognition methods to test real understanding beyond verbal output.
We use cookies to understand how you use our platform and to improve your experience. Click "Accept All" to consent, or "Decline non-essential" to opt out of non-essential cookies. Read our Privacy Policy.