Two-thirds of people surveyed think that artificial intelligence (AI) tools like ChatGPT have some degree of consciousness and can have subjective experiences such as feelings and memories, according to a new study from the University of Waterloo.

Large language models (LLMs) like ChatGPT often display a conversational style when outputting content. These human-like abilities have spurred debates on whether AI has consciousness.

According to the researchers, if people believe that AI has some level of consciousness, it could ultimately affect how people interact with AI tools, potentially strengthening social bonds and increasing trust. On the other hand, excessive trust can also lead to emotional dependence, reduced human interactions, and over-reliance on AI to make critical decisions.

The article "Folk psychological attributions of consciousness to large language models" was published in Neuroscience of Consciousness.

"While most experts deny that current AI could be conscious, our research shows that for most of the general public, AI consciousness is already a reality," said Dr. Clara Colombatto, professor of psychology at Waterloo's Arts faculty.

To understand public attitudes about AI consciousness, Colombatto and her colleague Dr. Steve Fleming at University College London surveyed a stratified sample of 300 people in the U.S. and asked if they thought ChatGPT could have the capacity for consciousness, as well as a variety of other mental states—such as the ability to make plans, reason, and feel emotions—and how often they used the tool.

The research found that the more people used ChatGPT, the more likely they were to attribute consciousness to it—an important consideration as AI tools are increasingly becoming part of our daily lives.

"These results demonstrate the power of language because a conversation alone can lead us to think that an agent that looks and works very differently from us can have a mind," said Colombatto.

"Alongside emotions, consciousness is related to intellectual abilities that are essential for moral responsibility: the capacity to formulate plans, act intentionally, and have self-control are tenets of our ethical and legal systems. These public attitudes should thus be a key consideration in designing and regulating AI for safe use, alongside expert consensus."

Future research will explore more specific factors driving these consciousness attributions, and their consequences for trust and social bonding, as well as possible variations within the same people across time, and across different people in other countries and cultures.

More information: Clara Colombatto et al, Folk psychological attributions of consciousness to large language models, Neuroscience of Consciousness (2024). DOI: 10.1093/nc/niae013