AI Lock-In: A New Frontier for AI Safety Research

A new position paper on arXiv (2608.14565) argues that AI safety research must address 'AI Lock-In,' a phenomenon where excessive reliance on AI systems leads to human deskilling, diminished independent functioning, and systemic vulnerabilities when AI systems fail or are compromised. The paper, authored by researchers, contends that while technical alignment and regulation of generative AI's societal impacts have been primary focuses, the risk of dependence on AI systems remains underexplored. It highlights that AI Lock-In is a systemic threat already emerging at individual, societal, and national levels, potentially amplified by AI service disruptions or geopolitical conflicts. The paper uses detailed scenarios to investigate how AI Lock-In emerges, aiming to prepare for its consequences. The authors call for AI safety research to include this dimension, emphasizing preparedness for a future where AI dependence is pervasive.