Neuroscientist Warns ChatGPT Could Harm Memory, Attention and Creativity

A cognitive neuroscientist has warned that widespread use of the AI chatbot ChatGPT could damage memory, attention and creativity, urging people — including those in his field — to avoid the tool.

Dr. Jared Cooney Horvath, a 42-year-old neuroscientist and author who created the metacognition program The Learning Blueprint, told the Daily Mail that he does not use ChatGPT and recommends others follow suit because, he said, the risks "outweigh the benefits." He said the chatbot encourages what he called "digital dependence," and warned that users may "no longer have the skill or knowledge" to complete tasks without AI assistance.

"Everything we know about how these tools work suggests that they're not going to be good in the long term," Cooney Horvath said, according to the interview. "People are using ChatGPT to avoid having to think and avoid doing everything that keeps our brains healthy." He added that "the key to all brain health is novelty and moderate stress. And when you use tools to avoid that, don't be surprised when things start to go haywire."

Cooney Horvath suggested that sustained reliance on generative AI could reduce opportunities for the kinds of mental effort that support memory consolidation, sustained attention and creative problem solving. He also said there are simple tests people can perform to assess whether they have been affected, though he did not specify those tests in the interview published by the Daily Mail.

The warning comes amid rapid growth in ChatGPT use: millions of people access the OpenAI tool daily for tasks such as drafting text, answering questions and generating ideas. Advocates say generative AI can increase productivity and serve as a valuable aid; critics have highlighted risks including misinformation, overreliance and potential impacts on learning and cognitive development.

Researchers at the MIT Media Lab have also explored the cognitive effects of using AI to perform intellectual tasks. A study released by the lab examined the consequences of using ChatGPT to write essays and reported negative effects on learning and skill acquisition when students outsource writing tasks to the tool. The researchers called for further study to understand the longer-term cognitive and educational implications as AI becomes more integrated into daily workflows.

Experts in neuroscience and education said the debate reflects broader questions about how technology reshapes human cognition. Some scientists caution that offloading cognitive tasks to external devices — from calculators to search engines — can change which skills are practiced and retained, but they also note the potential for tools to augment human capabilities when used deliberately. Cooney Horvath stressed that novelty and manageable cognitive challenge are central to brain health, and that routine avoidance of effortful thinking could reduce those inputs.

OpenAI and other developers have pointed to measures such as transparency, user education and tool design to mitigate overreliance, though the company did not respond to requests for comment in the Daily Mail piece. Educators and institutions are evaluating policies and pedagogical approaches to address the role of generative AI in classrooms and workplaces, balancing utility with concerns about skill erosion.

The discussion over AI's cognitive effects is ongoing, and researchers have called for systematic studies to quantify potential harms and benefits. Cooney Horvath's public stance contributes to a widening conversation among neuroscientists, educators and technologists about how to integrate AI tools without undermining cognitive development and independence. His comments, publicized in a national tabloid interview, add urgency to calls for research, clear guidance for users and education on when and how to use AI responsibly.