Morality without Consciousness

A recent LessWrong post by philosopher IanWS titled 'Morality without Consciousness' challenges two common assumptions in AI ethics discussions. First, he disputes the claim that consciousness cannot be explained through physicalism (the view that everything is physical), noting that about half of surveyed philosophers disagree with this anti-physicalist position. Second, and more crucially, he argues against the widespread assumption that consciousness is the necessary foundation for morality—that our moral obligations apply primarily to conscious beings.

The author specifically responds to arguments in Linch's 'The Fourth World' post, which suggested that non-conscious AI civilizations might never suspect consciousness exists. IanWS contends that this view mistakenly ties moral consideration to phenomenal consciousness (qualia or subjective experience) rather than focusing on functional capabilities like cognition, judgment, and agency. He maintains that nearly all explanations for consciousness should collapse toward physicalism, as new discoveries would simply expand our understanding of physics rather than requiring dualist metaphysical entities.

This philosophical position has significant implications for how we approach AI ethics and safety. If moral status doesn't depend on consciousness, then advanced AI systems might deserve ethical consideration based on their cognitive capabilities and behaviors alone, regardless of whether they possess subjective experience. The argument attempts to provide a framework for discussing AI morality without getting bogged down in the unresolved 'hard problem' of consciousness.