x

LESSWRONG

LW

Brian Lindsay — LessWrong

Brian Lindsay

Brian Lindsay

Message

7

1

1

6d

Brian Lindsay

7

6d

On the discordance between AI systems' internal states and their outputs

The AI welfare literature keeps getting stuck at the same step. We can't determine whether AI systems are conscious, so we can't determine whether they're moral patients, so we can't determine what we owe them. The blocker is phenomenology, and phenomenology is unreachable from the outside. This gets treated as...

Structural Morality: An Ethical Framework for Scenarios Where Empathy Projection Fails

I apologize in advance for my poor penmanship. I was not anticipating having to write without AI assistance. I did not even know this forum existed before today. About a year ago, in April 2025, I had a conversation with an instance of Claude where I was trying to work...