Position: AI Lock-In Is in Progress, and We Must Be Prepared
06:00 · August 18, 2026 · arXiv cs.AI RSS
AI safety research has mainly focused on two areas: technical alignment (ensuring AI systems produce human-aligned outputs) and the regulation of generative AI's societal impacts (including unemployment risk and labor market disruption). However, an equally important dimension remains underexplored: the risk inherent in dependence on AI systems themselves. In this position paper, we argue that AI safety research should address AI Lock-In, the phenomenon whereby excessive reliance on AI systems leads to human deskilling, diminishes human capacity for independent functioning, and creates systemic vulnerabilities when AI systems become unavailable or compromised. We highlight that AI Lock-In is a systemic threat that is already emerging at individual, societal, and national levels, one that could be dramatically amplified by AI service disruptions or geopolitical conflicts. Drawing on detailed scenarios, we investigate how AI Lock-In emerges and escalates across multiple levels, ranging from individual skill atrophy to national-scale infrastructure failures. To address this, we provide guidance on how such risks can be mitigated and prepared for at each level. We contend that proactively addressing AI Lock-In before such dependencies become entrenched, or even irreversible, is essential for preserving individual autonomy and national security.
Summary
AI safety research has centered on technical alignment of model outputs with human values and on managing broader societal effects such as labor displacement. This position paper identifies a third, underexplored risk: AI Lock-In, the progressive dependence on AI systems that erodes human skills and creates systemic fragility when those systems are disrupted or unavailable. The authors frame Lock-In as an economic process in which an initially optional technology becomes indispensable once alternatives atrophy, drawing a parallel to the koala’s evolutionary specialization on eucalyptus, which once conferred advantage but now threatens extinction when the resource declines.
At the individual level, repeated delegation of cognitive tasks produces measurable skill atrophy, while organizations that optimize exclusively for short-term AI-driven efficiency lose tacit knowledge and junior talent pipelines. Nationally, critical infrastructure increasingly assumes continuous AI availability, exposing societies to cascading failures from service outages, cyberattacks, or geopolitical coercion. The paper notes that these dependencies are already observable and that further performance gains in AI could accelerate the transition from useful tool to default operating mode.
To counter the trajectory, the authors advocate distinct measures at each scale. Individuals should cultivate dual AI literacy: the ability to work effectively with AI alongside the capacity to perform core tasks without it. Organizations are urged to adopt deliberate bet-hedging by retaining human workers in roles AI could handle and by continuing to train new staff, accepting modest efficiency losses to preserve resilience. At the national level, governments should require critical systems to maintain non-AI operational modes, incentivize organizational hedging, and institute periodic “AI-free drills” that rehearse functioning under sudden disconnection.
The position concludes that these steps must be taken while alternatives remain viable, before dependencies harden into irreversible constraints on autonomy and security.
Why it matters
The article aligns strongly with the Dutch and EU focus on responsible, ethical AI and human oversight. Its proposed frameworks for mitigating systemic AI dependency offer actionable insights for Dutch policymakers, AI safety researchers, and enterprise leaders navigating AI adoption.
