The researchers found that responsiveness influenced how people judged both humans and AI. Artificial intelligence is becoming increasingly ...
 |
| The researchers found that responsiveness influenced how people judged both humans and AI. |
New research reveals a surprising gap between how people judge artificial intelligence and human consciousness, even when both behave in exactly the same way.
The study, published in the journal Cognition, involved nearly 1,100 participants and examined how people attribute mental states to humans and artificial intelligence. Participants evaluated descriptions of agents displaying different levels of responsiveness to their environments. Some agents reacted to sounds, while others recognized emotional states or responded to changes around them. In different versions of the experiment, the same behaviors were attributed either to humans or to AI systems. Participants then evaluated how conscious or aware each agent appeared to be.
AI and Human Consciousness
The researchers found that responsiveness influenced how people judged both humans and AI. Agents that reacted more actively to their surroundings were generally considered more conscious and aware. However, a striking difference emerged when participants were asked specifically about consciousness. Even the most responsive AI agent was rated as less conscious than the least responsive human agent. In other words, identical behavior did not lead people to assign identical levels of consciousness to humans and machines. This suggests that people do not judge consciousness solely by observing what an entity can do. Their judgments are also influenced by whether they believe the entity is human or artificial.
One of the most interesting findings emerged when researchers asked participants about awareness rather than consciousness. Under this wording, judgments of humans and AI became much more similar. Participants appeared relatively comfortable describing AI as aware of its surroundings because awareness can refer to the ability to detect information and respond to events. Consciousness, however, often implies something deeper: subjective experience, feelings, and an inner mental life. The distinction suggests that people may accept that an AI system can recognize information and react intelligently without believing that it actually experiences anything. In this context, the language used to describe AI can significantly influence how people perceive its capabilities and mental states.
As AI systems become more convincing in conversation and behavior, people may increasingly describe them using words associated with human thought and emotion. Terms such as “understands,” “wants,” “feels,” or “decides” can make machines seem more human than their underlying capabilities necessarily justify. The study highlights the importance of distinguishing observable performance from claims about subjective experience. An AI system may recognize sadness, respond sympathetically, or explain a difficult problem without those behaviors proving that it experiences emotions or possesses consciousness. This distinction matters for technology companies, journalists, policymakers, researchers, and users who want to discuss AI accurately and responsibly.
Could AI Ever Become Conscious?
The study does not establish whether AI can or cannot become conscious. Instead, it examines how people judge the consciousness of humans and artificial systems when their behavior is held constant. Consciousness remains a difficult scientific and philosophical problem, and convincing outward behavior alone does not settle the question of whether a system has subjective experiences. Future research will need to investigate how people form these judgments, what evidence they consider meaningful, and how concepts of consciousness might change as AI technology advances. The findings also encourage a broader discussion about whether consciousness should be assessed through behavior, internal mechanisms, subjective reports, or a combination of different forms of evidence.
Conclusion
The research from Ludwig Maximilian University of Munich reveals a striking feature of human thinking: we can recognize human-like intelligence in artificial systems while still resisting the idea that those systems possess human-like consciousness. Although people often judge AI and humans similarly when considering awareness, they draw a much stronger distinction when asked about consciousness. As artificial intelligence continues to advance, understanding this difference will become increasingly important.
The central lesson is that acting intelligently, appearing aware, and being conscious are not necessarily the same thing. Separating these concepts can help us discuss AI more precisely without underestimating its capabilities or assuming that human-like behavior proves the existence of an inner mental life.
