Dual Impact: How AI Is Reshaping Retirement Savings and Late Careers at the Same Time

This is Part One of a two-part series. Part One maps the exposure facing adults in the 50–70 bracket: how AI is shaping both their retirement savings and their working lives. Part Two will present concrete strategies for navigating the new AI economy.

What's in This Article

  • The windfall embedded in your retirement account

  • The costs that come with the boom

  • What the research says about older workers and AI

  • The harder side of the picture: job loss and the bridge employment problem

  • Who's carrying the weight, and who isn't

  • Questions worth asking before Part Two

Introduction

Somewhere in the fine print of the AI boom, there's a contradiction that hasn't gotten nearly enough attention. Many adults in the 50-to-70 bracket are financially tied to AI in two directions at once. On one side, AI-driven technology stocks have been responsible for the bulk of S&P 500 gains since late 2022, quietly boosting the retirement accounts of millions who never made a deliberate choice to bet on AI. On the other side, the same AI wave is raising electricity bills, accelerating late-career job loss, and complicating the math of retirement in ways that don't make the same headlines.

This isn't a reason to panic. It is a reason to understand the full picture rather than just the parts that feel good. A March 2025 paper from the Wharton Pension Research Council is among the first serious attempts to examine both sides of this exposure for older workers simultaneously. What the researchers found is worth knowing, because it changes how you think about the risks you're carrying right now, and the ones you may not realize are arriving.

The windfall embedded in your retirement account

If your retirement savings are in an S&P 500 index fund or a target-date fund, a meaningful share of that money has been riding the AI wave whether you knew it or not. The five largest AI-associated companies (Microsoft, Alphabet, Amazon, Apple, and Nvidia) now represent nearly 29% of the S&P 500 by market weight, according to Motley Fool analysis from early 2025. When the full group of top tech companies is counted, their share of the index rises above 40%, a concentration that exceeds the peak of the dot-com era.

The returns have been substantial. The index returned 26% in 2023 and nearly 25% in 2024. BlackRock's 2025 analysis found that AI stocks accounted for roughly 75% of S&P 500 total returns since ChatGPT launched in late 2022. For someone in their 50s or 60s holding a typical target-date fund, with equity allocations ranging from 40% to 70% depending on target retirement year, these gains have been a genuine tailwind.

Empower's January 2026 data shows average 401(k) balances for Americans in their 50s at approximately $629,000, and for those in their 60s at approximately $577,000. Total U.S. retirement assets reached $49.1 trillion in 2025, up more than 11% year-over-year, according to the Investment Company Institute.

That's the good news. The question worth holding alongside it: your portfolio is more concentrated in AI stocks than it may appear, and you almost certainly didn't choose that. Most people holding index funds are carrying that exposure passively, through fund structures that track an index where AI companies now dominate. If that concentration continues to pay off, it continues to benefit you. If it reverses, the impact runs deeper than the headlines might suggest.

The costs that come with the boom

The data centers running the AI infrastructure behind those stock gains require extraordinary amounts of electricity. U.S. data centers consumed 183 terawatt-hours in 2024, representing 4.4% of total national electricity use, and that figure is projected to more than double by 2030, according to S&P Global. The cost of building and powering that infrastructure doesn't stay with the tech companies. Much of it gets distributed across residential ratepayers.

A Carnegie Mellon University study found that data center growth could increase average U.S. residential electricity bills by 8% by 2030, with increases exceeding 25% in high-demand markets. National residential electricity prices rose 11.5% in 2025 alone, more than double the general inflation rate that year. In Virginia, where data centers already consume more than 25% of state electricity, Dominion Energy's approved rate increases will add more than $11 per month to average residential bills starting in January 2026. In Arizona, the state's major utilities are seeking rate increases near 14%, driven in significant part by data center demand from the Phoenix metro area's 125-plus data centers.

For someone on a fixed income, a permanent upward shift in utility costs isn't a minor adjustment. It changes the math of what retirement actually costs month to month. The states most affected by data center-driven rate increases (Arizona, Nevada, Texas, and Virginia) are also among the most popular retirement destinations in the country.

What the research says about older workers and AI

The Wharton Pension Research Council paper, authored by Carlo Pizzinelli and Marina Mendes Tavares, offers the most comprehensive picture to date of how AI is likely to affect workers in the 55-to-69 age range. Their findings are more nuanced than either the optimistic or alarming framings suggest, and they're worth understanding carefully.

The researchers classified occupations into three categories:

  • High exposure + high complementarity: AI augments the worker (e.g., judges, senior lawyers)

  • High exposure + low complementarity: AI is more likely to substitute the worker (e.g., legal assistants)

  • Low exposure: limited AI impact either way

Their key finding for older workers: the current generation of people aged 55 and older are disproportionately concentrated in sectors with the highest AI substitution risks such as banking and insurance. On the other hand, many older workers are already in the kinds of roles where AI is more likely to make them more productive than to push them out. That's genuinely good news, if those workers can acquire the skills to use these tools effectively.

There's an important caveat, however. The researchers also found that older workers have significantly lower occupational fluidity than younger ones. They switch jobs less often, change industries less frequently, and take longer to find new work after job loss. That lower fluidity isn't a character flaw; it reflects longer tenure, deeper specialization, and accumulated expertise in a given field. But it also means that when disruption does arrive, older workers have fewer runways for adjustment.

The harder side of the picture: job loss and the bridge employment problem

Not every older worker is in a high-complementarity role. Banking, insurance, and administrative services, where many adults in the 50-to-64 range have built long careers, face among the highest AI substitution risks of any sector. These aren't jobs at the margins of the economy. They are precisely the jobs older workers have historically counted on as stable, well-paying, and accessible through late career.

The Bureau of Labor Statistics data from 2024 shows that the reemployment rate for displaced workers aged 55-64 is 55.3%, compared to 74.5% for workers aged 25-54. That's nearly a 20 percentage point gap. Workers over 50 take an average of 19 months to find new work after displacement, about 50% longer than workers in their 20s and 30s, according to AARP research. After involuntary separation, household income for workers over 50 falls an average of 42%, per Urban Institute data, and the effects persist well into retirement.

A 2024 study published by Springer Nature found that a one standard deviation increase in a job's automatability raises the probability of unplanned early retirement by 1.4 to 1.7 percentage points. AI isn't creating this dynamic from scratch; the pressures on older workers have been building for years. But it is accelerating the timeline, particularly in the sectors where they're concentrated.

This acceleration threatens what researchers call "bridge employment" — the lower-intensity, part-time, or transitional work that many people in their late 50s and early 60s rely on to bridge the gap between a primary career and full retirement. ProPublica research found that more than half of workers aged 50-54 are involuntarily pushed out of longtime jobs before they plan to retire. Among workers aged 55-64 who have retired, more than half did so involuntarily.

Who's carrying the weight, and who isn't

The picture above assumes meaningful retirement savings. A significant share of adults in the 50-to-70 bracket don't have them. A 2024 AARP survey found that 1 in 5 Americans aged 50 and older have no retirement savings at all. Approximately 24 million older adults depend on Social Security for at least half of their income. Nearly half of Americans aged 60 and older have household incomes below the Elder Index, the minimum threshold for covering basic needs in their region, according to the National Council on Aging.

The Wharton researchers explicitly acknowledge this dimension. They note that the equity wealth channel of AI's economic effects "would be particularly relevant for older workers, who on average own larger amounts of equity assets." The operative phrase is "on average." For those with substantial savings, the AI boom has been a genuine financial benefit. For those without savings, it is invisible as an asset and very visible as a cost: on the utility bill, in the job market, in the accelerating pace of change they have to navigate with fewer tools.

This isn't a reason to feel despairing about either group. It is a reason to be honest about the fact that the AI boom is not delivering the same experience to everyone in this age bracket, or anywhere close to it. Brookings Institution research found that for every doubling of regional AI innovation, the labor share of income declines by 0.5 to 1.6 percentage points. Productivity gains from AI accrue disproportionately to capital owners. Those who own capital benefit; those who rely primarily on labor income bear more of the adjustment costs.

Questions worth asking before Part Two

Part Two of this series will take up the question of what to do with all of this. Before that, it's worth sitting with a few honest questions.

How much of your retirement portfolio is actually tied to AI stocks? If you hold a target-date fund or an S&P 500 index fund, the answer is probably more than you think. Knowing your equity allocation and the top holdings in your fund takes a few minutes and gives you a more accurate picture of what you're riding.

What's your plan if late-career disruption arrives? The research on older workers' reemployment rates is clear-eyed: the consequences of unexpected job loss late in a career are serious and tend to be long-lasting. A financial buffer, a realistic sense of what bridge work might look like, or a reduced-spending adjustment period are all worth thinking through before rather than during a disruption.

Are your utility costs built into your retirement budget? If you live in or plan to retire in Arizona, Nevada, Texas, or Virginia, the projected rate increases through 2030 are publicly available from your state utility commission. Building realistic utility projections into your retirement math now is easier than adjusting later.

Do the AI jobs and skills landscape apply to your work? The Wharton research offers a genuinely useful framework: if your occupation involves judgment, client relationships, independent decision-making, or professional accountability, AI is more likely to augment your work than replace it. If your role involves more routine, structured tasks, it's worth understanding the landscape more specifically.

Part Two will address each of these questions directly, with concrete strategies for navigating the new AI economy at whatever stage of career or retirement you're in.

Sources

AARP. (2024). New AARP survey: 1 in 5 Americans ages 50+ have no retirement savings. https://www.aarp.org/press/releases/2024-4-24-new-aarp-survey-1-in-5-americans-ages-50-have-no-retirement-savings.html

AARP Public Policy Institute. (2025). How AI is impacting the future of work among adults age 50-plus. https://www.aarp.org/pri/topics/work-finances-retirement/employers-workforce/workforce-trends-older-adults-artificial-intelligence/

Arizona Clean Energy. (2025). Arizona utilities push 14% rate hikes as data centers drive soaring energy demand. https://www.azce.org/news/arizona-utilities-push-14-rate-hikes-as-data-centers-drive-soaring-energy-demand

BlackRock. (2025). AI revolution in retirement. https://www.blackrock.com/us/financial-professionals/retirement/insights/ai-revolution-in-retirement

Brookings Institution. (2024). AI's impact on income inequality in the US. https://www.brookings.edu/articles/ais-impact-on-income-inequality-in-the-us/

Bureau of Labor Statistics. (2024). 65.7 percent of long-tenured displaced workers were reemployed in January 2024. https://www.bls.gov/opub/ted/2024/65-7-percent-of-long-tenured-displaced-workers-were-reemployed-in-january-2024.htm

Carnegie Mellon University. (2024). Data center growth could increase electricity bills 8% nationally. https://www.cmu.edu/work-that-matters/energy-innovation/data-center-growth-could-increase-electricity-bills

Empower. (2025). Average 401(k) balance by age. https://www.empower.com/the-currency/life/average-401k-balance-age

Fidelity. (2025). Q4 2025 retirement analysis. https://about.fidelity.com/data-and-insights/q4-2025-retirement-analysis

Investment Company Institute. (2025). Quarterly retirement market data, Q4 2025. https://www.ici.org/statistical-report/ret_25_q4

Kiplinger. (2025). The average 401(k) balance by age. https://www.kiplinger.com/retirement/401ks/the-average-401k-balance-by-age

Motley Fool. (2025). These 5 AI stocks make up 28.8% of the entire S&P 500 index. https://www.fool.com/investing/2025/01/03/these-5-ai-stocks-make-288-entire-sp-500-index/

National Council on Aging. (2024). Addressing the nation's retirement crisis: The 80 percent. https://www.ncoa.org/article/addressing-the-nations-retirement-crisis-the-80-percent-financially-struggling/

Pew Research Center. (2020). Trends in U.S. income and wealth inequality. https://www.pewresearch.org/social-trends/2020/01/09/trends-in-income-and-wealth-inequality/

Pew Research Center. (2025). What we know about energy use at U.S. data centers amid the AI boom. https://www.pewresearch.org/short-reads/2025/10/24/what-we-know-about-energy-use-at-us-data-centers-amid-the-ai-boom/

Pizzinelli, C., & Mendes Tavares, M. (2025). AI and the future of work in an aging economy. Wharton Pension Research Council. https://pensionresearchcouncil.wharton.upenn.edu/wp-content/uploads/2025/08/AI-and-the-Future-of-Work-in-an-Aging-Economy.pdf

ProPublica. (2023). If you're over 50, chances are the decision to leave a job won't be yours. https://www.propublica.org/article/older-workers-united-states-pushed-out-of-work-forced-retirement

S&P Global. (2025). Data center grid-power demand to rise 22% in 2025, nearly triple by 2030. https://www.spglobal.com/energy/en/news-research/latest-news/electric-power/101425-data-center-grid-power-demand-to-rise-22-in-2025-nearly-triple-by-2030

Springer Nature. (2024). Skill-specific automatability of aging workers. https://academic.oup.com/workar/article/10/3/267/7111438

Urban Institute. (2024). How secure is employment at older ages? https://www.urban.org/sites/default/files/publication/99570/how_secure_is_employment_at_older_ages_2.pdf

Virginia State Corporation Commission. (2025). SCC issues order on DEV biennial review 2025. https://www.scc.virginia.gov/about-the-scc/newsreleases/release/scc-issues-order-on-dev-biennial-review-2025/scc-rules-in-dev-biennial-review-case.html

Note: This article was written by a human with the assistance of Claude Sonnet 4.6.
It has been independently fact-checked and the sources cited have been independently verified.

© 2026 neoClarity LLC.
Licensed under CC BY 4.0. You may republish it with attribution to neoClarity (https://neoclarity.ai)