AI Acts Human But Still Not Seen as Conscious — SkimNews
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- Ludwig-Maximilians-Universität München researchers led by Louis Longin published a study in Cognition showing participants rated even the most responsive AI as less conscious than the least responsive human.
- Participants in nearly 1,100 trials judged AI and humans similarly on 'awareness,' but drew a firm line at 'consciousness,' attributing significantly less of it to machines regardless of identical behavior.
- Louis Longin emphasized the study’s novel approach: directly comparing mental state attributions to AI and humans acting the same way, unlike prior research that asked general questions about AI minds.
- Ophelia Deroy suggested people use mental terms for AI not because they believe machines have inner experiences, but because language lacks better descriptors for complex machine behavior.
- Bahador Bahrami highlighted that public discourse often overstates AI’s mental capacities, while everyday judgment is more nuanced—dependent on whether terms like 'awareness' or 'consciousness' are used.
Why it matters: The public’s resistance to attributing consciousness to AI—even with human-like behavior—means companies and media that describe AI as having intentions or inner experience may be shaping perception beyond actual belief. This linguistic framing could mislead users about AI capabilities, affecting regulation, trust, and ethical design.
Ask SkimNews
