Study finds dark personality traits correlate with frequent AI use

Despite the overwhelming popularity of artificial intelligence (AI) in recent years, it appears that AI browsing only takes up about 1% of online activity, and a recent study has uncovered that it’s mostly people with dark personality traits who do it regularly.

Indeed, analysis of data retrieved from over 14 million website visits by 499 university students and 455 members of the general public has suggested that AI-related browsing is far less common than you might assume, taking up less than 1% of all website visits, according to a study published in Cyberpsychology, Behavior, and Social Networking.

Frequent AI users score high in Dark Triad

Additionally, individuals conducting this type of online activity more frequently than others have tended to score high on certain aversive or ‘dark’ personality traits associated with narcissism and psychopathy. They have also had more positive views toward AI overall.

Digging deeper, the researchers have also discovered that ‘prolific users,’ or those whose AI browsing accounted for more than 4% of their total website visits, scored much higher on measures of Machiavellianism, narcissism, and psychopathy compared to their peers and the general public.

According to study author Emily McKinley, a PhD candidate at the University of California, Davis, who spoke to PsyPost: 

“People who use AI more tend to score higher on aversive personality traits, particularly Machiavellianism, narcissism, and psychopathy, though these patterns were stronger among students.”

Another interesting conclusion from the study pertained to activities before and after AI use, where many would precede their AI utilization with browsing the internet and telecom sites like search engines and login pages, and followed their AI deployment with visits to websites related to education, computers, or professional tasks. 

This suggested that AI tools were a common part of a workflow, particularly in academic or work contexts, and were therefore seen more as productivity aids than a source of entertainment.

Meanwhile, scientists at the Massachusetts Institute of Technology (MIT) have warned about the risk of so-called ‘cognitive debt,’ which could arise as the result of a growing reliance on large language models (LLMs) like ChatGPT, Gemini, and others, to write content, potentially causing harm to our brains and thinking skills.