Digital Consciousness Model Paper: Evidence Against 2024 LLM Consciousness Is Not Decisive

burny_tech · x · 2026-09-27

A new paper from Rethink Priorities and collaborators, "Initial results of the Digital Consciousness Model" (DCM), proposes the first systematic, probabilistic framework for assessing evidence of consciousness in AI systems, aggregating multiple leading consciousness theories rather than picking one. Initial results: the evidence is against 2024 LLMs being conscious, but it is not decisive—and it's far weaker than the evidence against consciousness in simpler AI systems. The authors use it to argue that 99.9%+ confidence LLMs aren't conscious is unjustified, with the lead author leaning toward octopi being conscious and current LLMs probably not.

Original post →

More from AGI Musings

AGI Musings channel →