The digital divide is the gap between those who have access to, and can effectively use, digital technologies and those who cannot. It spans physical access (devices, connectivity, compute), skills and literacy, and the capacity to translate access into benefit. The concept predates AI and was developed to describe inequalities in internet and computer access; it serves as the general framing from which AI-specific divides are derived.
From digital divide to AI divide
Commentators distinguish a newer "AI divide" as a specialization of the older concept: an unequal distribution of the access, benefits, and opportunities created by AI across regions, communities, and socioeconomic groups. UNESCO has framed AI literacy as central to a "new digital divide," arguing that unequal capacity to understand and use AI compounds existing inequalities (Source: https://www.unesco.org/ethics-ai/en/articles/ai-literacy-and-new-digital-divide-global-call-action). The International Labour Organization has similarly warned that, absent cooperative international action, the AI transition could widen the gap between high- and low-income countries (Source: https://www.ilo.org/resource/news/mind-gap-bridging-ai-divide-will-ensure-equitable-future-all). Academic work has begun to characterize an "AI divide" specifically as the gap between those who can and cannot generate sustainable value from AI tools, distinct from mere access (Source: https://www.tandfonline.com/doi/full/10.1080/0144929X.2025.2500451).
A related stratification appears within populations that already have access. An Anthropic study of Claude users, paired with an Epoch AI income-stratification analysis, found usage concentrated among higher-income cohorts — with a large majority of users earning above $100,000 and underuse among lower-income groups — a pattern a Brookings analysis characterized as a "second digital divide" in which differences in effective use, not just access, drive disparate benefit (Source: https://www.anthropic.com/research/work-study-2026; https://epochai.org/blog/claude-user-income-stratification; https://www.brookings.edu/articles/ai-adoption-digital-divide-2026).
Relationships
- related: AI Divides (Literacy / Occupational / Ethico-Philosophical), AI Fluency Divide — the AI-specific specializations that depend on this general concept.
- related: AI Dividends (Universal Basic Capital, Digital Dividend, Global Dividend), AI and Productivity — adjacent debates over who captures AI's gains.
Provenance note
Page created 2026-06-15 (gap-scan) from supporting web sources (UNESCO, ILO, peer-reviewed and Brookings analyses). It resolves a dangling [[concepts/digital-divide]] reference relied on by AI Divides (Literacy / Occupational / Ethico-Philosophical) (named there as the "general concept") and the sources/project-deal page.