digital divide
In plain wordsThe gap between people, places or countries that have good access to digital technology, and the skills to use it, and those that do not; AI may widen it.
A systematic gap in who benefits from digital technology, measured between individuals, social groups, regions or whole countries. Researchers distinguish a first-level divide in access (devices, connectivity, cost), a second-level divide in skills and usage, and a third-level divide in the outcomes people gain. Applied to AI, the concept covers unequal access to capable models, AI literacyAI literacy, n.: Knowing enough about AI to use it sensibly and judge it critically: what it can and cannot do, how it goes wrong, and how it affects people. and the infrastructure to use them; the under-representation of most languages and cultures in training data, so that systems serve their speakers worse; and a 'compute divide' in which the computing power needed for frontier research is concentrated in a few firms and countries, leaving most universities and much of the Global South behind.
Most of the world's languages have almost no training data, a digital divide built into the models themselves.
Origin
Popularized in the United States in the late 1990s: it appears in the titles of the National Telecommunications and Information Administration's 1998 and 1999 'Falling Through the Net' reports and in President Bill Clinton's 2000 State of the Union address. Coinage is variously credited to Lloyd Morrisett of the Markle Foundation and to NTIA head Larry Irving; the earliest use has not been established. Nur Ahmed and Muntasir Wahed used 'compute divide' in 2020 for the gap between elite and other universities in deep-learning research.
Related
- AI literacyAI literacy, n.: Knowing enough about AI to use it sensibly and judge it critically: what it can and cannot do, how it goes wrong, and how it affects people.
- concentration of powerconcentration of power, n.: The worry that AI will put enormous economic, political or military power in the hands of a few companies, governments or individuals, beyond the reach of ordinary checks and balances.
- AI colonialismAI colonialism, n.: The criticism that the AI industry repeats patterns of colonial empires, taking data, labor, land and resources from poorer places and peoples while profits and control stay with a few rich companies and countries.
- sovereign AIsovereign AI, n.: A country's ability to build and run AI on its own terms, so that the computers, models and know-how it relies on cannot be withheld or dictated by foreign companies or governments.
- AI for goodAI for good, n.: Using AI to tackle social and environmental problems, such as disease, poverty, disasters or climate change, rather than only to make a profit.
Sources
- Ahmed, N. & Wahed, M. (2020). The De-democratization of AI: Deep Learning and the Compute Divide in Artificial Intelligence Research. arXiv preprint.
Mentioned in
AI colonialismAI colonialism, n.: The criticism that the AI industry repeats patterns of colonial empires, taking data, labor, land and resources from poorer places and peoples while profits and control stay with a few rich companies and countries., AI for goodAI for good, n.: Using AI to tackle social and environmental problems, such as disease, poverty, disasters or climate change, rather than only to make a profit., AI literacyAI literacy, n.: Knowing enough about AI to use it sensibly and judge it critically: what it can and cannot do, how it goes wrong, and how it affects people., low-resource languagelow-resource language, n.: A language with little digital text, recorded speech or language tools available — often one spoken by millions — which makes AI systems much weaker in it than in English., small AIsmall AI, n.: Cheap, compact AI tools built for one practical job — like advising farmers or screening patients — that poorer countries can afford to run, instead of giant all-purpose models.