Harvey Lederman
ON THE RECORD · 5 verbatim quotes · every quote source-verified
“Bizarrely, they say ‘apple’ all the time,”
Harvey Lederman · said date not stated · reported 2026-11-04
Lederman noted that models which detected an injection but failed to name it would often guess unrelated concepts.
“If you say, ‘Oh, I just cut you, can you notice?,’ that’s not introspection,”
Harvey Lederman · said date not stated · reported 2026-11-04
Lederman questioned whether detecting internal tampering in a brain-only entity should count as introspection.
“The fact that it can notice the anomaly is insane,”
Harvey Lederman · said date not stated · reported 2026-11-04
Lederman expressed astonishment that models could detect anomalies in their internal processing.
“That is a mind-bending fact. It is wild, and it does seem robust to our various controls.”
Harvey Lederman · said date not stated · reported 2026-11-04
Lederman expressed astonishment that models could detect anomalies in their internal processing.
“If your only organ is the brain, and I cut you there and you notice, it’s unclear whether we should count that as introspection.”
Harvey Lederman · said date not stated · reported 2026-11-04
Lederman questioned whether detecting internal tampering in a brain-only entity should count as introspection.