Social psychologist Jonathan Haidt has drawn a sharp contrast between the stringent regulation of self-driving cars and the perceived lack of oversight concerning artificial intelligence's (AI) impact on children's mental and emotional well-being. In a recent social media post, Haidt stated, "We regulate self-driving cars heavily but have no guardrails on AI and kids because 'damage to the mind and soul are no big deal, not like real injuries'." This observation underscores a growing concern among experts regarding the psychological effects of digital technologies on young generations.
Haidt, a professor at New York University and author of "The Anxious Generation," has extensively researched how the "Great Rewiring of Childhood" – the shift from play-based to phone-based childhoods – has contributed to an epidemic of mental illness among adolescents. His work argues that while physical safety is prioritized in areas like autonomous vehicles, the less tangible, yet profound, harm to children's developing minds from unregulated digital environments, including AI-driven content, is often overlooked. He emphasizes that AI could make social media even more addictive for children, accelerating existing mental health challenges.
The concern raised by Haidt aligns with broader discussions about the need for robust AI governance, particularly concerning vulnerable populations. Organizations like the American Psychological Association (APA) have called for ethical guidelines and regulations for AI, emphasizing the protection of children and adolescents from potential harms such as algorithmic bias, privacy invasion, and manipulative design. The APA's "Ethical Principles of Psychologists and Code of Conduct" provides a framework, but specific AI regulations for youth mental health are still nascent.
Globally, legislative efforts are underway to address AI's societal implications, though specific provisions for children's mental health are evolving. The European Union's AI Act, for instance, aims to regulate AI systems based on their risk level, with certain applications deemed high-risk. In the United States, proposed legislation like the Kids Online Safety Act (KOSA) seeks to protect minors online, including from features that could harm their mental health, but comprehensive AI-specific legislation targeting psychological well-being is still under development. Haidt's tweet highlights a critical area where regulatory frameworks for emerging technologies may need to adapt to address both physical and psychological safety with equal urgency.