A pediatrician's commentary published Monday in STAT News urges a fundamental shift in how artificial intelligence is evaluated for children. Dr. Dua Hassan argues that current AI applications for kids rely too heavily on engagement metrics rather than rigorous health outcomes.
Dr. Hassan, writing in her professional capacity, stresses that pediatricians lack the evidence needed to make informed recommendations. She advocates for randomized controlled trials that specifically measure real developmental outcomes, not just how long children interact with AI tools.
The piece highlights a growing gap between the rapid deployment of AI in children's products and the slow pace of clinical validation. No specific AI products or companies are named, but the critique targets the broader ecosystem of educational and entertainment AI.
For pediatricians, the lack of definitive data creates a dilemma: they cannot confidently recommend or warn against AI tools without evidence of harm or benefit. The call to action is directed at regulators, researchers, and technology developers to prioritize child-specific clinical trials.
The op-ed represents a single clinician's perspective and does not reflect official medical society positions or peer-reviewed research beyond the author's own expertise.