DeepMind paper argues AI cannot instantiate consciousness due to abstraction fallacy

A new research publication from Google DeepMind argues that artificial intelligence systems cannot achieve consciousness because they rely on symbolic computation, which is an abstraction dependent on an external cognitive agent. The paper, titled "The Abstraction Fallacy: Why AI Can Simulate But Not Instantiate Consciousness," critiques computational functionalism—the hypothesis that subjective experience emerges from abstract causal topology regardless of physical substrate. The authors contend that this view mischaracterizes the relationship between physics and information. They propose that symbolic computation is not an intrinsic physical process but a mapmaker-dependent description requiring an active, experiencing agent to convert continuous physics into finite meaningful states. The framework distinguishes simulation (behavioral mimicry via vehicle causality) from instantiation (intrinsic physical constitution via content causality). The authors argue that algorithmic symbol manipulation is structurally incapable of instantiating experience, though this does not rely on biological exclusivity—if an artificial system were conscious, it would be due to its specific physical constitution, not its syntactic architecture. The paper offers a physically grounded refutation of computational functionalism, aiming to resolve uncertainty around AI sentience without requiring a complete theory of consciousness.