A study from Ludwig Maximilian University of Munich asked nearly 1,100 participants to judge characters described as either humans or AI systems, with varying levels of responsiveness to their surroundings. The researchers found that while responsiveness boosted perceived consciousness in both cases, a sharp gap remained: even the most responsive AI was rated as less conscious than the least responsive human.
The pattern changed when participants were asked about 'awareness' rather than 'consciousness.' For awareness, judgments of humans and AI followed largely similar patterns. The authors suggest that awareness is seen as a neutral term tied to observable behavior, while consciousness carries implications about subjective inner experience.
The researchers caution that hyped language about AI—crediting it with intentions or moral conscience—does not necessarily shape everyday judgments. But the study shows that the boundary between humans and machines depends heavily on which words are used. The findings were published in the journal Cognition.