Korean survey finds AI skill gap deepens inequality

The Korean survey is a reminder that the race to AGI isn’t just about model quality and compute; it’s also about who can actually use these systems. If nearly four out of five citizens already believe AI skills will widen inequality, that’s a political and cultural fact AI companies have to design around. As more workplaces bake tools like Gemini, Claude and Copilot into everyday workflows, workers who lack the confidence or training to use them risk becoming structurally less competitive—even if the underlying models are available to everyone.([v.daum.net](https://v.daum.net/v/20260529005317453?utm_source=openai))

From a strategic standpoint, this “AI divide” can affect how quickly AGI‑class tools can be safely deployed. If only a narrow slice of the population can wield powerful agents effectively, pressure grows to automate whole job families outright rather than augment them, which can provoke backlash and regulatory friction. Conversely, societies that invest early in broad‑based AI literacy—primary education through adult retraining—are likely to absorb more of the productivity gains and push further into high‑value human–AI collaboration.

For labs and platforms, surveys like this are a warning: usability, guardrails and educational tooling aren’t nice‑to‑haves; they’re central to adoption and to the social license needed to deploy more capable systems over time.