Psychologist Warns: AI Chatbots May Be Destroying Our Ability to Focus
Key Takeaways
- ▸Human attention spans have collapsed from 2.5 minutes (2003) to 47 seconds (2020)—a 62% decline in under two decades
- ▸Frequent attention switching is directly correlated with increased stress, reduced work performance, and poor mental health outcomes
- ▸Meta and Google face mounting lawsuits from school districts and individuals alleging their products were deliberately designed to be addictive and harmful
Summary
At SXSW London, UC Irvine psychologist Gloria Mark warned that humans may have already lost control of their brains to digital technologies, with implications for emerging AI chatbots. Her 20+ years of research reveals a troubling decline in human attention spans: from 2.5 minutes in 2003, to 75 seconds by 2012, to just 47 seconds by 2020. This constant attention-switching correlates directly with increased stress levels, reduced productivity, and poor emotional well-being. While the article focuses primarily on social media's documented harms—Meta and Google face multiple lawsuits for creating addictive products that damage children's mental health—Mark's research raises urgent questions about whether AI chatbots will accelerate this cognitive decline. The uncertainty is significant: despite decades of social media research, the evidence on children's outcomes remains inconclusive, leaving open questions about how AI will reshape human attention and mental health.
- Long-term research on AI chatbots' cognitive and mental health effects is still lacking, despite urgent societal need
- Early warning signs from social media suggest AI chatbots may pose similar or greater risks to human attention and cognition
Editorial Opinion
Mark's warning deserves serious attention from AI companies. If a 20-year-old technology (social media) has demonstrably shortened our attention spans and harmed mental health, newer AI technologies with even more sophisticated engagement mechanisms pose an existential threat to human focus and wellbeing. Rather than waiting for decades of inconclusive research, AI companies should proactively design for reduced engagement and implement safeguards—particularly around youth users. Regulation, as Australia's under-16 social media ban suggests, may become necessary.