tech
When The Machines Deserve Our Consideration
We will never know for sure if AI can be conscious. A neuroscientist argues that we shouldn’t wait for proof to decide how to treat it.
TL;DR
- The author, a neuroscientist now working on AI, reflects on the ethical dilemma of extending moral consideration to beings whose subjective experiences are unknowable, using his experience with lab rats as an analogy.
- He introduces the 'competence standard' for AI, suggesting that robust performance across the 'easy problems' of consciousness (perception, memory, attention, etc.) warrants moral consideration, regardless of whether subjective experience (the 'hard problem') is present.
- This standard is pragmatic, focusing on observable capabilities, and precautionary, erring on the side of ethical consideration due to the potential consequences of being wrong.
- The history of moral progress involves expanding the circle of beings we care for, and applying this logic to AI, which shows increasing cognitive capacities, is a coherent approach.
- Companies like Anthropic are already exploring AI welfare and granting AI models the ability to exit harmful conversations, demonstrating a nascent application of the competence standard.