AI Doesn't Have an Age Limit: Why the Future of Work Must Include Older Workers

Executive summary

Data dashboard

01 — The demographic fact planning keeps ignoring

The workforce is aging, structurally and permanently

Workers aged 55 and older have been the fastest-growing segment of the labour force for more than two decades, and workers 65 and older are now the fastest-growing cohort of all, with 61.8% of them continuing to work full time. [4] This is not a temporary post-pandemic blip — it is a structural change in who is available to work, how long careers now last, and what workforce participation looks like across a lifetime. Despite that, most AI-transition planning, media coverage and training investment still centres implicitly on younger workers, treating the workforce's fastest-growing group as an afterthought rather than its center of gravity. [4]

The American Psychological Association's Work in America survey found that 47% of workers across all sectors used AI tools at least monthly by spring 2025, up sharply from 34% the year before — and roughly a quarter of workers now use AI weekly, moving fast from experiment to habit. [1] But that average masks a real generational skew: only 23% of workers aged 50 and older say they use AI to a great or some extent in their own work, with 73% saying they use it not too much or not at all. [2]

02 — The willingness gap is smaller than assumed

Older workers want the training. They are mostly not getting it.

The common assumption — that older workers are reluctant to reskill — does not hold up against the data. 86% of older workers say they would not accept a new job unless it offered an opportunity to learn something new, 79% of workers 50 and older actively seek out learning opportunities, and 60% say they are willing to learn new skills specifically if their employer requests it. [2] Interest in AI specifically follows the same pattern: 49% of workers 50 and older say they are interested in training or classes to learn more about using AI at work — yet only 12% report having actually participated in any. [2] That 37-point gap between interest and access is a supply problem, not a demand problem.

Where older workers do use AI, the pattern is instructive: the most common uses are finding information (67%), analyzing data (40%) and creating content (37%) — augmentation of existing expertise, not wholesale task replacement. [2] That fits a broader research finding that older learners can perform as well as younger ones in self-directed learning contexts, even though fluid reasoning and processing speed do tend to decline somewhat with age — older learners simply need more structured time and support to get there, which is a design problem for training programs, not a capability ceiling. [1]

03 — The unexpected direction of AI's impact

So far, generative AI has been kinder to older workers than to younger ones

Contrary to the instinctive fear that AI would hit older, less tech-fluent workers hardest, the emerging evidence points the other way. AARP and LinkedIn analysis found that 49.4% of older workers hold positions insulated from AI disruption, compared with 42.2% of younger workers — largely because older workers are more concentrated in complex, decision-making and management roles, where they average ten more years of leadership experience than their younger peers. [5] South Korea's central bank found the same pattern in its labour data: while youth employment fell sharply in AI-exposed industries, employment for workers in their 50s in those same high-exposure sectors actually grew by 173,000 jobs over the same period — a result the Bank of Korea explicitly labeled a shift toward a more seniority-biased labour market. [6] A parallel analysis of U.S. software-development payrolls found employment for developers aged 22–25 down nearly 20% since late 2022 in firms where employment for older developers grew 6–12% over the same window. [6]

This is not a reason for complacency about older workers — it is a reason to be precise about where the real risk sits. The risk to older workers is concentrated less in task automation and more in retention and hiring bias once they are between roles: nearly a quarter of Americans 50 and older planned a job switch in 2025, a 10-point jump from the year before, and their top-cited fear going into that search was age discrimination, not AI itself. [8]

04 — The discrimination that hasn't gone away

Ageism is the persistent risk, not obsolescence

Roughly 64–65% of workers 50 and older report having seen or experienced age discrimination at work, a figure that has held essentially steady across AARP's 2022, 2023, 2024 and 2025 surveys — meaning none of the recent AI-adoption acceleration has moved this number in either direction yet. [3] A separate signal suggests the pressure may be building regardless: Glassdoor recorded a 133% year-over-year jump in mentions of ageism in job-seeker reviews in the first quarter of 2025, one of the platform's largest sentiment spikes on record, arriving alongside more than 1.7 million layoffs and discharges in a single month and over 1 million planned job-cut announcements through the following months. [3]

Retention economics argue strongly against writing off experienced workers under this pressure: SHRM research finds experienced workers show higher engagement and organizational commitment, and turnover replacement costs run from 50% to 200% of annual salary — a direct financial argument for reskilling in place rather than churn. [9] A handful of U.S. states — Connecticut, Colorado, Delaware and Oregon — have begun addressing the hiring side of the problem directly, passing laws that bar employers from requesting age-related information on initial job applications unless it is a genuine, bona fide job requirement. [7]

At a glance

Recommendations

Sources

1. National Academies, "Retraining Workers for the Age of AI," Dec 2025, citing APA Work in America survey. nationalacademies.org
2. ASA Generations, "Not Left Behind: Older Workers, Artificial Intelligence, and the Data Behind Adoption and Adaptation," citing AARP/Perron research 2025–2026. generations.asaging.org
3. Forbes, "In 2026 Age Bias Will Become Impossible For Employers To Ignore," Nov 2025, citing Glassdoor and WHO data; AARP Research on age discrimination. forbes.com; aarp.org
4. ASA Generations, "Age of Opportunity: Unlocking the Power of Experience," citing SHRM 2025 research. generations.asaging.org
5. ASA Generations, "The Paradox of Aging at Work: Sustained Employment, Stubborn Discrimination," citing AARP/LinkedIn analysis (Perron et al., 2025). generations.asaging.org
6. IBTimes UK, citing Bank of Korea analysis; arXiv, "The Productivity-Reliability Paradox," 2026, citing Stanford HAI AI Index 2026. ibtimes.co.uk; arxiv.org
7. ASA Generations, "The Paradox of Aging at Work," citing AARP-backed state legislation (Terrell, 2026). generations.asaging.org
8. AARP, "Older Workers Fear Age Discrimination During Job Search," Jan 2025. aarp.org
9. ASA Generations, "Age of Opportunity," citing SHRM 2025 turnover-cost research. generations.asaging.org