EY Ripples, in partnership with Microsoft, OpenAI, OATS from AARP, and Kite Insights, surveyed 2,515 people aged 60 to 85 across 16 countries between October and November 2025. The resulting report, Understanding Older Generations’ Adoption of AI, is among the first large-scale studies to document how this age group actually engages with AI tools, as distinct from surveys that treat older adults as a footnote within broader population samples. The findings confirm that the AI access gap is structural, not attitudinal.
Usage
AI use among respondents aged 60 to 85 is divided but not absent. About two in five have never used AI or tried it only once or twice. Around one in five use it often or very often. Among Baby Boomers specifically, 24% describe themselves as quite or very familiar with AI, 38% are actively learning about it through online resources, educational videos, or social media, and only 15% expressed no interest in learning at all.
When older adults do use AI, reported satisfaction is high. Of respondents who had tried AI tools, 79% cited learning as their primary use case. Among those using AI at work, 84% reported positive experiences. Creative use returned a positive rating of 80%.
The gap is not principally explained by resistance or disinterest. The majority of non-users are people who have not yet found a relevant access point, not people who have ruled AI out.
Demographic gaps
Employment status is the strongest predictor of AI adoption in this age group. Respondents still in work use AI at roughly three times the rate of retired respondents. This is significant because most AI literacy programmes, including the EU’s Article 4 obligations under the AI Act and the US LIFT AI Act currently in Congress, are designed around employment contexts: workplace training, employer-provided tools, and workforce upskilling. Retired adults fall outside the primary target of nearly every public policy instrument in this space.
A gender gap is also present. Thirty-one percent of women in the survey report never having used AI, compared to 20% of men. The report does not fully decompose the drivers of this gap, but the pattern is consistent with broader findings on digital access and confidence among older women.
Barriers
The most cited barrier to AI adoption is data privacy: 41% of respondents worry that AI will misuse or take their personal data. This is not technophobia; it is a calibrated concern about a documented risk. The next two barriers are navigational: 34% do not know which tools to use, and 23% do not know where to start. Seventeen percent find tools too difficult to use, and 19% report a lack of support when using AI tools. Cost is a concern for 11%; device access and physical limitations each affect fewer than 5%.
Nineteen percent report no barriers at all.
The barrier profile matters for what it rules out. Difficulty is not the primary obstacle. Cost and device access are marginal factors. The dominant barriers are trust and orientation, both of which are addressable through programme design rather than technology improvement.
On accuracy, 80% of respondents agree that not everything produced by AI has been checked to make sure it is true. EY treats this as evidence of baseline AI awareness. It also suggests that older adults are already applying a degree of critical scrutiny to AI outputs, which is precisely the disposition that AI literacy programmes aim to develop.
What the pattern shows
The EY survey makes visible a population that existing AI access infrastructure largely ignores. Older adults who are not in employment have no institutional access point to AI tools or AI training. Libraries, adult education centres, and community health services are the channels EY recommends, but none of these currently operate at the scale or with the funding that would match the size of the gap.
The report frames this as an “AI just transition,” borrowing language from climate policy to argue that older generations should not be left behind as AI becomes embedded in civic and economic life. The framing is useful because it signals that the issue is not remedial support for a struggling group but equitable distribution of access to a general-purpose technology that is reshaping employment, healthcare, public services, and daily information.
The three largest barriers, privacy concern, not knowing which tools to use, and not knowing where to start, are exactly what structured AI literacy provision addresses. The 85% of non-users who retain some interest in AI represent a large and reachable population. The question the report leaves open is who is responsible for reaching them, and on what timeline.
Source: EY Ripples, Understanding Older Generations’ Adoption of AI (March 2026)
Last updated: 2026-04-21