Societies are ageing across the OECD, due to both longer life expectancy and declining fertility rates. At the same time, digital technologies are developing fast, reshaping how people learn, work and interact with each other and society. This Spotlight presents trends in the use of digital technologies by older adults for essential everyday activities that can support independence, strengthen social connections and improve quality of life in older age. It presents evidence showing that age differences in digital technology use are gradually narrowing, although uptake of frontier digital technologies like generative artificial intelligence (GenAI) remains highly concentrated among younger users. At the same time, it highlights barriers to AI adoption, such as mistrust, and physical and cognitive limitations, that can deter use. This Spotlight explores the potential of AI to enhance accessibility to digital tools and support for older adults when engaging with essential online services. It also explores practical strategies for improving information and communication technology (ICT) skills, accessible service design and evidence-based policymaking for effective digital participation in the AI age.
OECD Digital Economy and Society Outlook 2026
Abstract
Across the OECD, societies are ageing as longer life expectancy combines with low fertility rates to reshape demographic structures (André, Gal and Schief, 2024[1]). At the same time, digital technologies are transforming how people learn, work, communicate and access essential services at a pace unprecedented in previous technological shifts, bringing new opportunities but also potential risks. Demographic ageing places pressure on labour markets, productivity and public finances. However, digital technologies can help mitigate some of these effects by boosting innovation, enhancing productivity and supporting economic growth in the context of a shrinking workforce.
The benefits of digital transformation are unevenly distributed with some older adults accessing or benefitting from digital opportunities differently. This lack of balance has implications not only for individual well-being and social participation but also for society’s ability to draw on the experience and capabilities of older adults (Mubarak and Suomi, 2022[2]). Digital technologies, including artificial intelligence (AI), hold considerable promise for supporting independence, strengthening social connections and improving quality of life in older age (OECD, 2024[3]). As societies become more digital, understanding how older adults engage with digital technologies – along with the barriers they may face – is essential for designing policies that ensure that governments seize the benefits of digital technologies, including AI, for people of all ages.
AI is increasingly embedded in economic and social activities, including essential services. This Spotlight explores trends in the access and use of essential online services among older adults (aged 55 to 74). These services include online public services, health information seeking, Internet banking and e-commerce, as well as generative artificial intelligence (GenAI) use. Alongside the substantial opportunities offered by AI to older adults, the Spotlight looks at the multifaceted barriers older adults may face to embrace technology and AI such as physical limitations, data-driven age-based biases and trust-eroding practices. The Spotlight concludes with a discussion of the role of policy in supporting effective and safe use by older adults of digital technologies in general and AI in particular and ensuring that emerging innovations benefit people as they age.
As digital technologies – and increasingly AI – have become essential for participating in modern society, it is increasingly important that both the tools themselves and the skills required to use them are widely accessible to prevent the risk of deepening the gap in technology access (Lynn et al., 2022[4]). Essential public and private services are moving online to improve accessibility and efficiency, requiring users of all ages to adapt to these changes. Without adequate support, however, some groups risk falling behind in this transformation, potentially exacerbating age-related disparities.
Recent data indicate that differences in the use of digital technologies across age groups are gradually narrowing (OECD, 2026[5]). Adults now entering their 60s and 70s experienced digital technologies during much of their working lives, leading to a higher baseline of digital familiarity than previous generations. As a result, current cohorts of older adults tend to have greater exposure to information and communication technologies (ICTs) and correspondingly stronger ICT skills than earlier generations. Yet this does not guarantee that disparities will naturally decline; rapid and continuous technological innovation introduces new tools, interfaces and competencies that require ongoing adaptation. This means that even digitally experienced older adults may encounter new barriers as the technological frontier evolves.
Accessing the Internet is becoming universal for people of all ages across the OECD
AI tools are increasingly prevalent on the Internet – from search engines to e-commerce platforms. Expanding access to and use of the Internet is essential for narrowing differences in digital technology usage across age groups. Over the past decade, the average share of people across the OECD who used the Internet within the previous three months rose from 81% in 2015 to 94% in 2025 (Figure 3.S.1). Although Internet use increased across all age groups, the rise has been particularly steep among older adults. In 2015, fewer than six out of ten (58%) older adults were online. By 2025, this figure had increased to nearly nine out of ten (87%), demonstrating that gaps related to basic access are narrowing rapidly across cohorts.
Percentage of individuals using the Internet in the last three months, by age group, 2015 (or closest) and 2025 (or latest)
Notes: For Canada and Egypt, data refer to 2022. For Israel and the United States, data refer to 2015 and 2023. For Mexico, data refer to 2015 and 2024. For Switzerland, data refer to 2014 and 2025.
Source: OECD calculations based on OECD (2026[5]), ICT Access and Usage Database, https://oe.cd/dx/ict-access-usage.
Internet use is also becoming more intensive. By 2025, 90% of individuals accessed the Internet daily or almost daily, up from 70% ten years earlier. This trend also includes older adults: daily Internet users increased by almost 80% over the decade, reaching 78% in 2025 (OECD, 2026[5]). Evidence from the United States complements this pattern, with smartphone ownership among adults aged 65 and over rising from 13% in 2012 to 61% in 2021 (Faverio, 2022[6]). Greater smartphone penetration provides older adults with increased opportunities to connect, both through traditional web browsing and via app-based services.
Uptake of online public services varies among older adults
Increased Internet use is expanding opportunities for accessing a wider range of essential services online that incorporate a range of digital technologies, notably AI (OECD, 2025[68]). Although overall use of online public services is increasing and age gaps are narrowing, fewer older adults use these services than people in general. In 2025, 68% of older adults used the Internet to interact with public services, compared with 81% across all age groups. While this represents a significant uplift for both groups, the gap widened rather than narrowed in half of the countries with data available, highlighting uneven progress (Figure 3.S.2). Usage rates are highest – and age gaps the narrowest – in countries such as Norway, Denmark and the Netherlands, compared to Türkiye, which continues to exhibit wide gaps of more than 30 percentage points (p.p.). Beyond age itself, these disparities may also reflect variation in the availability, usability and user friendliness of online public services.
Delivering online public services that are accessible and user-centred begins with understanding the needs of diverse user groups and involving them throughout service design (OECD, 2022[7]). Such an approach can strengthen trust in digital government services by ensuring they respond effectively to users’ varied expectations and capabilities (OECD, 2021[8]; OECD, 2025[9]; OECD, 2024[10]). This is especially important when transitioning from in-person to digital channels, as factors such as age, educational attainment and ICT skills strongly influence use of online public services (Garín-Muñoz et al., 2019[11]).
Percentage of individuals using the Internet to access public authorities’ services during the last 12 months, by age group, 2013 and 2025
Notes: No data for this indicator were collected for 2015, leaving 2013 as the closest year for comparison.
Source: OECD calculations based on OECD (2026[5]), ICT Access and Usage Database, https://oe.cd/dx/ict-access-usage.
Searches for health information online are increasing as digital health services expand
The age gap in online health information seekers is smaller than that observed for the use of online public services. Across the OECD, 52% of older adults searched for health information online in 2025, compared to 62% of the total population. Mirroring general trends in digitalisation, some countries also report the highest shares of older adults seeking health information online. However, people in other countries are catching up, including in Croatia, Bulgaria and Hungary with age gaps narrowing most notably in Korea (Figure 3.S.3).
Between 2015 and 2025, the share of individuals seeking health information online increased by 14 p.p. across all ages. Among older adults, the share rose by 20 p.p., indicating that age gaps are narrowing over time (Figure 3.S.3). This shift also coincides with the expansion of telemedicine, which grew significantly during the COVID-19 pandemic as digital technologies and more flexible telehealth legislation enabled broader patient outreach and offered benefits for both patients and healthcare providers (Saigí-Rubió et al., 2022[12]).
Yet even as gaps narrow for basic activities like online information seeking, the continual evolution of digital health tools means that new gaps may emerge. This is especially the case for more advanced applications such as telemedicine. In addition, AI-based services, such as symptom checkers and personalised recommendations, are increasingly deployed (Kopka, He and Feufel, 2026[13]). Controlling for other factors, a person over 50 years old may still be up to 30% less likely to use telemedicine (Haddad and Bhat, 2025[14]).
Percentage of individuals using the Internet for seeking health information over the last three months, by age group, 2015 (or closest) and 2025 (or latest)
Notes: For Canada and Egypt, data refer to 2022. For Colombia and Mexico, data refer to 2015 and 2024. For Costa Rica, data refer to 2017 and 2025. For Israel, data refer to 2023. For Switzerland, data refer to 2014 and 2025.
Source: OECD calculations based on OECD (2026[5]), ICT Access and Usage Database, https://oe.cd/dx/ict-access-usage.
Internet banking is rising rapidly across all age groups
AI tools, including AI chatbots, have become increasingly prevalent in the financial services sector, including for use in Internet banking (Graham et al., 2025[15]). The use of Internet banking is growing rapidly across the OECD. In 2025, 72% of individuals used online banking services, up from 50% in 2015 (Figure 3.S.4). Importantly, the age gaps in usage are narrowing. Among older adults, usage nearly doubled over the same period – from 35% in 2015 to 62% in 2025. This trend is particularly pronounced in Northern European countries, where Internet banking is fully mainstreamed and integrated into public and private payment systems.
The growing use of Internet banking is a key component of the broader shift towards increasingly digital payment ecosystems. As the costs of digital transactions decline, maintaining cash infrastructures becomes more expensive, prompting financial institutions to expand online service delivery (OECD, 2025[16]). At the same time, opportunities to withdraw cash and access in-person banking services are diminishing (IMF, 2025[17]). Ensuring that digital financial services remain accessible, reliable and secure for all users is therefore essential, as digital banking services increasingly also integrate AI-based functionalities.
Without appropriate safeguards and support, people who prefer cash payments or have lower levels of ICT skills – both more common among older adults – risk being unable to fully benefit from the digital transformation. These groups may face increasing difficulties in managing everyday financial tasks, particularly as financial decision making and customer interactions are increasingly mediated by AI-driven tools and automated systems (OECD, 2024[18]).
Percentage of individuals using the Internet for banking over the last three months, by age group, 2015 (or closest) and 2025 (or latest)
Notes: For Canada and Egypt, data refer to 2022. For Colombia, Japan and Mexico, data refer to 2015 and 2024. For Israel and the United States, data refer to 2015 and 2023. For Switzerland, data refer to 2014 and 2025.
Source: OECD calculations based on OECD (2026[5]), ICT Access and Usage Database, https://oe.cd/dx/ict-access-usage.
E-commerce remains primarily a one-way activity for many older adults
The popularity of purchasing products (goods and services) online – whether through e-commerce retailers or digital intermediation platforms (DIPs) – continues to rise across the OECD, with AI features increasingly present on e-commerce websites. Between 2015 and 2025, the share of individuals who made online purchases within the past year increased by 22 p.p., reaching 72% by the end of the period (Figure 3.S.5). Older adults are also increasingly engaging in online shopping, with 54% making at least one online purchase in 2025 – almost double the share recorded in 2015 (28%).
Part of the growth in online purchases can be attributed to behavioural shifts during the COVID-19 pandemic, when many consumers turned to online retail out of necessity. However, the persistence of age gaps in shopping behaviour may point to long-standing preferences for local, in-person purchasing among older adults. These preferences continue to influence behaviour even as overall e-commerce use expands (Cano-Leiva et al., 2024[19]; Chun, Kwon and Yang, 2024[20]).
Percentage of individuals making online purchases over the last 12 months, by age group, 2015 (or closest) and 2025 (or latest)
Notes: For Canada and Egypt, data refer to 2022. For Colombia, Japan and Mexico, data refer to 2015 and 2024. For Switzerland, data refer to 2014 and 2025. For the United States, data refer to 2015 and 2023.
Source: OECD calculations based on OECD (2026[5]), ICT Access and Usage Database, https://oe.cd/dx/ict-access-usage.
The number of individuals selling products online via DIPs has also increased. However, participation in circular economy activities such as selling second-hand items continues to lag far behind online purchasing. In 2025, on average 25% of individuals across the OECD sold products online, compared with only 14% of older adults. Growth since 2015 has been modest for both groups, rising from 17% and 8%, respectively (OECD, 2026[5]).
Growth patterns suggest that e-commerce remains largely a one-way activity for many users, particularly for older adults. This pattern persists despite the expanded opportunities offered by DIPs and AI-driven platforms to sell personal production, refurbished goods or unwanted items to both local and international audiences. These opportunities are facilitated through personalised recommendations, dynamic pricing and automated listing tools.
As age gaps in online access and use of online essential services continue to diminish, important challenges remain and require careful attention. To enable all to benefit from digital transformation, it is crucial to understand the varied needs and concerns that may influence the engagement of older adults with digital technologies and services. Research shows that older adults face more challenges than younger people when adopting digital technologies. As a consequence, older adults risk being left behind in an increasingly digital society (Neves, Waycott and Malta, 2018[21]; Reneland-Forsman, 2018[22]; Rojo-Pérez, Fernández-Mayoralas and Rodríguez-Rodríguez, 2021[23]).
With AI becoming increasingly embedded in digital solutions, patterns of AI adoption must be better understood. A stronger evidence base could enable policymakers to help older adults maximise the opportunities AI brings. Such evidence would include identifying barriers to adoption and how they can be overcome.
Identifying and reducing barriers to adoption is challenging because different generations have lived through different social, cultural and historical contexts that shape their relationships with digital technologies (Lissitsa, Zychlinski and Kagan, 2022[24]; Chee, 2023[25]). There are also differences within age cohorts: within the 55–74-year-old age group, some are fully engaged in their careers; others are approaching retirement; and some are already retired. This leads to vastly different patterns of digital engagement and levels of reliance on technology. Moreover, digital technologies are rapidly transforming, introducing new forms of interaction that require continual adaptation and placing additional demands on users who may already face barriers to digital engagement.
Older adults lag their younger counterparts in AI adoption, presenting opportunities for older adults to reap more AI-related benefits
Over time, cohort replacement leads to a natural closing of age gaps in ICT use as established digital habits are carried into older age. However, age gaps will not disappear entirely as the digital technology frontier and its applications are constantly evolving. The latest leap in digital technology for personal use is GenAI. Across the OECD, 37% of people used GenAI tools in 2025 (Figure 3.S.6). Usage rates are heavily skewed towards younger people, with 16- to 24-year-olds almost five times as likely to use GenAI than those aged 55–74. As few as 1% of those 75 or older use the technology in OECD Member countries with available data.
Percentage of individuals using any GenAI tools in the last three months, by age group, 2025 (or latest)
Notes: The indicator on GenAI captures individual-level use of tools such as ChatGPT, Copilot, Gemini, LLaMA, Midjourney or DALL-E to create new content – including text, images, programming code or videos – over the past three months. For Colombia, data refer to 2024 with a reference period of 12 months. For Japan, data refer to 2024 and no reference period is specified. In addition, data relate to individuals aged 20 to 69.
Source: OECD calculations based on OECD (2026[5]), ICT Access and Usage Database, https://oe.cd/dx/ict-access-usage.
AI provides a range of opportunities for older adults
Despite low uptake, AI offers significant opportunities to enhance older adults’ access to services and support everyday activities. The many capabilities offered by AI illustrate why closing the age gap matters for the everyday lives of older adults. AI could significantly enhance older adults’ access to essential online services by reshaping interactions around their capabilities and preferences. Through adaptive interfaces that adjust text size, layout and complexity, AI can reduce demands on cognitive and fine motor skills that may deter older users. Proactive, context-aware guidance can further help by anticipating needs, offering step-by-step assistance and preventing errors before they occur. Natural language technologies, particularly conversational agents such as chatbots, can enable older adults to engage with online platforms intuitively and without the anxiety associated with navigating menus or remembering procedures. Coupled with features like voice control, personalised recommendations and multimodal feedback, AI-driven systems can simplify tasks that previously may have felt overwhelming. In so doing, they can support greater independence, confidence and participation in digital environments (Leite et al., 2024[26]).
AI also offers concrete opportunities to improve the physical health of older adults, in some cases compensating for age-related decline (Osei-Asiamah et al., 2025[28]). AI-enabled assistive technologies such as speech-to-text and text-to-speech tools, cognitive support systems, adaptive reminders and continuous health monitoring can help offset sensory loss, while reducing cognitive load and supporting independent living among older adults (Chae et al., 2025[28]; Gursoy et al., 2025[29]). In healthcare and long-term care contexts, AI-enabled telemonitoring and decision-making tools can complement human care. To that end, it can enable earlier detection of risk and reduce the physical burden associated with travel or frequent in-person visits, especially when paired with support for older adults to use and understand the technology. Thus, AI holds potential not only to accommodate physical and cognitive limitations but to actively mitigate their impact on older adults’ autonomy and well-being.
Beyond digital interactions, AI may play an important role in maintaining independence and reducing isolation. By extending access to transport and services, AI offers opportunities to support older adults in their physical environments and mobility, thus improving overall mental health. Assistive robotics, smart home systems and connected devices can help with everyday tasks, monitor health conditions and support independent living (Ghorbani et al., 2023[30]; Gamble, Chami and Nancoo, 2024[31]; Scott, 2026[32]). In the mobility domain, advances in autonomous vehicle technologies may further expand independence by enabling older adults to maintain access to transport even as driving ability declines. By reducing reliance on physical, cognitive or sensory capacities associated with driving, such systems could support continued participation in social and economic life, provided they are designed to be accessible, reliable and aligned with users’ needs (Zandieh and Acheampong, 2021[33]; Kim, 2024[34]).
Overcoming barriers to AI use is needed to improve quality of life for some older adults
Despite these opportunities, several barriers may limit the adoption and effective use of AI among older adults. As AI becomes increasingly embedded in everyday life, barriers include negative perceptions about AI, low ICT skills, and physical and cognitive limitations. The growing use of voice-first interfaces, AI chatbots and AI companions requires users to adapt to new modes of interaction. These can be challenging for older adults with limited ICT skills or reduced confidence in their own abilities to use emerging technologies (Sacar et al., 2024[35]). At the same time, the personalised and persuasive nature of AI systems raises concerns about trust and vulnerability, particularly in contexts shaped by loneliness or cognitive decline. These dynamics highlight how skill gaps, scepticism, lack of trust, and physical and cognitive limitations interact to shape older adults’ engagement with AI. Some of these factors may be mitigated through the design of AI-enabled solutions.
Older adults tend to have more negative perceptions about AI than younger people
Perceptions and attitudes towards AI constitute an important barrier as older adults tend to perceive AI more sceptically than younger people, reflecting lower exposure, knowledge gaps and more limited perceived usefulness. In 2025, 43% of older adults perceived no need for it, almost double the rate of 16–24-year-olds, at 25%. Furthermore, 12%, lacked the necessary skills and 9% were unaware of the technology (OECD, 2026[5]). These perceptions are mirrored in broader attitudes towards the impact of AI on society. Results from the OECD Truth Quest Survey in 2024 (Figure 3.S.7) show that those aged 55–64, and 65 and over, are significantly more likely than young adults to disagree that AI will have a positive impact on their lives. The net disagreement rates of -10.3 and -11.2 p.p., respectively, highlight an age gradient in perceived benefits from AI.
Percentage of adults who feel AI will have a positive impact on their life by age, 2024
Source: OECD (2026[69]), “Share of adults who feel AI will have a positive impact on their life”, OECD Going Digital Toolkit, https://goingdigital.oecd.org/indicator/82 based on OECD (2024[37]), The OECD Truth Quest Survey: Methodology and findings, https://doi.org/10.1787/92a94c0f-en.
The perception of older adults about AI’s usability, privacy and security affect adoption and can act as barriers. Concerns are particularly prominent in high-stakes contexts such as healthcare and banking. Here, AI is perceived as lacking empathy, being insufficiently explainable or posing risks to personal data, which contribute to resistance even when potential benefits are acknowledged (Nawaz et al., 2025[38]; Rony et al., 2025[39]). At the same time, negative perceptions of AI among older adults are also shaped by overall anxiety and low self-confidence. Technology anxiety manifesting as fear, hesitation or apprehension remains a widespread psychological barrier among older adults and is closely associated with lower perceived ease of use and usefulness of AI-enabled technologies (Kim, Freddolino and Greenhow, 2023[40]). A lack of awareness and insufficient access to tailored learning opportunities further reinforce scepticism. This leaves some older adults unconvinced of AI’s value in everyday life or uncertain about their ability to use it safely and effectively (Shandilya and Fan, 2022[41]; Hutton, 2024[42]).
Trust in digital and AI-enabled systems also shapes adoption decisions and links to how technology is perceived. For instance, older adults with chronic illnesses are often less likely to use digital health solutions, primarily because they place greater trust in face-to-face medical care (Győrffy et al., 2023[43]). The disembodied nature of digital interactions can undermine the sense of reassurance and personal connection many patients associate with in-person care. Evidence from healthcare settings shows that older adults frequently express concerns that AI-supported decision making lacks human judgement; may not fully account for individual circumstances; and offers limited opportunities to question or contextualise recommendations (Rony et al., 2025[39]). Trust is further weakened by worries about data privacy and security, particularly when sensitive health information is processed by opaque AI systems (Wong et al., 2025[44]). For those already managing complex health conditions, this perceived loss of personal attention can deter engagement with otherwise beneficial digital health services (Veras et al., 2025[45]).
Older adults tend to have lower ICT skills than younger people
Limited ICT skills remain a key barrier. While many stand to benefit from the rapid rollout of AI-enabled features, a significant number of older adults lack the necessary basic ICT skills to take advantage of them (Vercruyssen et al., 2023[46]). The effective use of AI builds on a broad set of foundational and generic ICT skills, including the ability to navigate digital interfaces, manage information, communicate online and assess digital risks. Thus, AI does not eliminate the need for ICT skills but may raise the threshold required for meaningful and capable use. At the same time, not all interactions with AI rely on traditional interfaces. Emerging interaction modes – such as voice-based systems and assistive technologies – may reduce some of these skills requirements, as previously discussed.
Although access to the Internet has expanded across age groups in OECD Member countries, pronounced skill gaps persist. Older adults are disproportionately represented among those with low or only task-specific digital proficiency. These gaps reflect cumulative differences in digital exposure over the course of life and are compounded by lower participation in adult learning at older ages. As AI-enabled functionalities become increasingly embedded in online essential services, from online public services to health and e-commerce, insufficient ICT and learning skills risk translating into new barriers for older adults. This, in turn, risks shifting the age gaps from technology access to effective technology use (OECD, 2024[47]).
Physical and cognitive limitations can reduce AI uptake for older adults
For some older adults, physical and cognitive limitations may further constrain engagement with AI technologies. Ageing is sometimes framed through a lens of deficit, historically associated with dependence, decline and withdrawal from economic life (Boudiny, 2013[48]). While not true for all, older adults often have physical and cognitive limitations that come naturally with ageing. As such, they experience a decline in mobility, cognition, hearing and vision that can make it difficult to navigate evolving digital interfaces (Hong et al., 2013[49]). Further, reduced fine motor control and dexterity can, for example, make it difficult to use touchscreens. Meanwhile, declining vision can impair the ability to read on-screen text and interpret visual content. This cognitive decline constrains the ability to learn new skills, retain information across interfaces and troubleshoot technical issues independently.
While AI offers immense opportunities, mitigating risks in AI-enabled essential services is needed
The growing use of AI also introduces new risks. Recent technological advances in GenAI have led to new and expanded forms of online scams targeting older adults (Herrera, Van Sickle and Podhradsky, 2024[50]). AI is also allowing scammers to extend their reach beyond traditional online interfaces where awareness may generally be higher and expand their attacks through unfamiliar means. Most notably, this includes voice cloning and deepfakes, amplifying existing fraud risks through greater personalisation, realism and scale (Napal, 2024[51]).
Older adults – who on average have lower ICT skills, higher emotional vulnerability and cognitive limitations – may find attempts at fraud through voice cloning and personalised scams harder to resist. This, in turn, may erode their trust in digital environments, particularly where AI enables more convincing and scalable forms of manipulation (Sima et al., 2026[54]). Loneliness, isolation and fear of abandonment play an important part, making older adults more likely to continue engaging with individuals that may harbour ill intent (Napal, 2024[51]). In 2022, 31% of individuals aged 65 and over across the OECD lived alone (OECD, 2025[53]). This context is often associated with fewer social interactions and reduced access to informal help and support, exacerbating risks in the face of potentially fraudulent behaviours.
Beyond individual exposure, risks also emerge from the design and deployment of AI systems in essential services. AI-enabled decision-making systems are increasingly embedded in essential services that many older adults rely on – including healthcare, long-term care, pensions, social benefits and finance (OECD, 2019[54]; OECD, 2025[55]). Lower digital engagement among older adults means their experiences and data may be underrepresented in the datasets used to train AI models, increasing the potential for misclassifications (Shumailov et al., 2024[56]; OECD, 2025[57]; Ramadan et al., 2026[58]).
Taken together, these risks illustrate how AI can amplify challenges related to data quality, decision-making assumptions and user vulnerabilities when deployed at scale. The risks, in turn, highlight the importance of designing online services that function reliably across user groups of all ages. Evaluating AI systems to detect unexpected outcomes; ensuring datasets reflect a broad range of experiences; and testing tools under realistic conditions can help older adults benefit from digital technologies without encountering unnecessary barriers.
Beyond skills development, policy can help support more effective digital technology use among older adults. Older adults can face multifaceted barriers when using digital technologies – including AI enabled services and tools. Consequently, effective interventions require co-ordinated efforts involving the private sector, local governments and civil society to scale impact and respond directly to the needs of older adults. This Spotlight points to four interrelated areas where targeted policy can help address persistent barriers to digital technology use and AI adoption among older adults, while supporting digital participation for all: strengthened skill use; greater awareness of opportunities associated with AI; more accessible online services; and a stronger evidence base.
Strengthen ICT skills beyond basic use
Several governments have developed programmes to enhance ICT skills, including AI, and to support lifelong learning among older adults. These programmes aim to equip older adults with AI solutions that help them navigate digital and non-digital environments more effectively, including online essential services, app use and frontier tools such as GenAI. Programmes often address common risks, such as privacy and cybercrime, alongside practical ICT competencies. For greater impact, programmes should be tailored to specific user needs, be accessible across different contexts and build confidence, as well as competence in interacting with AI-driven systems. They should strengthen both practical ICT skills and resilience in an increasingly complex digital environment.
Examples in the OECD include:
-
Austria: the programme for AI Literacy and Skills is a national initiative aligned with Austria’s existing AI strategy. It explicitly targets groups with lower digital literacy, including older adults. The programme aims to strengthen understanding of AI concepts and practical interactions with AI-enabled tools, bridging gaps between basic ICT skills and more advanced AI-enabled applications (OECD, 2025[59]).
-
Korea: the Ministry of Education’s “Hangeul Sunshine Bus” delivers mobile media and information literacy education in venues such as senior centres, auditoriums and community centres, supporting older adults with low ICT skills or mobility challenges. Curricula extend beyond basic ICT skills to include app-based ordering, Internet banking and creating social media content (Ministry News, 2024[60]; MSIT, 2025[61]).
-
Ireland: Age-Friendly AI is a national AI literacy initiative targeting older adults to ensure they can understand, question and engage with AI-enabled technologies through a combination of public dialogue, co-creation workshops and community-based training. Covering topics such as automation, data use, privacy, bias and everyday AI applications, the programme delivers AI-focussed ICT skills training in libraries and community organisations, enabling large-scale reach (Age-Friendly Ireland, 2025[62]).
-
Israel: digital literacy initiatives under Israel’s National Digital Programme support older adults with limited digital skills through community-based training delivered across 169 local authorities. Programmes cover practical topics such as digital government services, online banking, e-commerce and cyber safety. They increasingly include AI-related applications for everyday use, information access and interaction with digital services. Training is primarily delivered through blended learning models combining in-person support with digital platforms, such as Campus IL (Mendels and Weiner, 2025[63]).
Raise awareness of the opportunities of AI
Awareness-raising initiatives and community-based support can empower older adults to benefit more fully from digital technologies and to enhance their social connections. Such campaigns can inform older adults about the potential benefits of digital technologies, including AI-enabled services in healthcare, finance and public services, as well as support initiatives. Outreach can target individuals experiencing social isolation or living in underserved areas through diverse channels, community events and local networks. Public libraries can also play a role as trusted local hubs supporting access to online services and the development of ICT skills, helping older adults engage with digital technologies more confidently (Labo Société Numérique, 2024[64]). Hands-on demonstrations and interactive sessions can help demystify complex technologies, including AI systems, and make them easier to understand (Matisola, 2024[65]). Peer-led workshops and community ambassadors can also provide practical guidance, leveraging familiar networks to build confidence and familiarity with AI-enabled tools.
Design accessible online services while maintaining offline options
Online services can be made more accessible by incorporating design principles that consider vision, cognition and physical limitations when deploying AI-enabled interfaces such as chatbots, voice systems and automated decision-making tools (Trischler and Westman Trischler, 2022[66]). Overly complex interfaces, technical language or excessive information can create barriers for older adults, making smart devices and online services harder to use. AI systems can provide powerful solutions to these challenges. Involving older adults in the design and testing process could enable service providers to better understand their diverse needs and everyday practices and to assess how AI systems perform under real-world conditions (Fischer, Peine and Östlund, 2020[67]; OECD, 2024[3]). The participation of older adults in prototyping and testing could help ensure services remain intuitive and aligned with practical user needs.
Maintaining offline alternatives is important, as some older adults may prefer or require in-person services even when online options are available. Lower digital engagement of older adults is often linked to functional constraints, health conditions or reduced confidence rather than a lack of interest, making exclusive reliance on online or AI-mediated services potentially exclusionary (Kim, Freddolino and Greenhow, 2023[40]). Maintaining offline or hybrid service options can therefore act as an important safeguard as AI-enabled automation expands across essential services, ensuring continuity of access for those with limited ICT skills or health constraints. Providing parallel in-person services can also reduce pressure on older adults to adopt technologies. This, in turn, allows ICT skills and confidence to develop at a pace aligned with individual needs and circumstances, while preserving access in increasingly automated services (Shandilya and Fan, 2022[41]).
Expand the evidence base
As people age and societies become progressively older, high-quality data on digital technology use among older adults are essential to track trends over time and identify potential barriers. Such data are especially critical for those aged 75 and older, for whom evidence remains limited. Comparable indicators across countries can support cross-national learning and provide insight into effective approaches for different contexts. Expanding the evidence base also enables evaluation of the impact of media and information literacy programmes; helps anticipate emerging challenges; and informs the design of strategies to support sustained engagement with digital technologies. More disaggregated data on usage patterns, skills and service preferences can guide development of interventions that better align with older adults’ capabilities and everyday practices.
References
[62] Age-Friendly Ireland (2025), “Age-friendly AI: Ireland’s: National artificial intelligence literacy initiative”, https://agefriendlyireland.ie/initiative/age-friendly-ai-irelands-national-artificial-intelligence-literacy-initiative (accessed on 11 March 2026).
[1] André, C., P. Gal and M. Schief (2024), “Enhancing productivity and growth in an ageing society: Key mechanisms and policy options”, OECD Economics Department Working Papers, No. 1807, OECD Publishing, Paris, https://doi.org/10.1787/605b0787-en.
[48] Boudiny, K. (2013), “‘Active ageing’: From empty rhetoric to effective policy tool”, Ageing & Society, Vol. 33/6, pp. 1077-1098, https://doi.org/10.1017/S0144686X1200030X.
[19] Cano-Leiva, J. et al. (2024), “How has COVID-19 changed individuals’ e-commerce and shopping mobility habits? Evidence from Madrid region”, Transportation Research Part A: Policy and Practice, Vol. 190, p. 104295, https://doi.org/10.1016/j.tra.2024.104295.
[28] Chae, K. et al. (2025), “Stakeholders’ perceived benefits and concerns regarding artificial intelligence in the care of older adults”, Journal of the American Geriatrics Society, Vol. 74/2, pp. 355-364, https://doi.org/10.1111/jgs.70228.
[25] Chee, S. (2023), “Age-related digital disparities, functional limitations, and social isolation: Unraveling the grey digital divide between baby boomers and the silent generation in senior living facilities”, Aging & Mental Health, Vol. 28/4, pp. 621-632, https://doi.org/10.1080/13607863.2023.2233454.
[20] Chun, H., E. Kwon and D. Yang (2024), “The rise of e-commerce and generational consumption inequality: Evidence from COVID-19 in South Korea”, Regional Science and Urban Economics, Vol. 104, p. 103971, https://doi.org/10.1016/j.regsciurbeco.2023.103971.
[6] Faverio, M. (2022), “Share of those 65 and older who are tech users has grown in the past decade”, https://pewrsr.ch/3HZd2ao (accessed on 21 October 2025).
[67] Fischer, B., A. Peine and B. Östlund (2020), “The importance of user involvement: A systematic review of involving older users in technology design”, The Gerontologist, Vol. 60/7, pp. e513-e523, https://doi.org/10.1093/geront/gnz163.
[31] Gamble, E., P. Chami and T. Nancoo (2024), “Artificial intelligence-powered technologies for independent living among older adults: A review”, Age and Ageing, Vol. 53/Supplement_4, https://doi.org/10.1093/ageing/afae178.177.
[11] Garín-Muñoz, T. et al. (2019), “Models for individual adoption of eCommerce, eBanking and eGovernment in Spain”, Telecommunications Policy, Vol. 43/1, pp. 100-111, https://doi.org/10.1016/j.telpol.2018.01.002.
[30] Ghorbani, F. et al. (2023), “A decision-aware ambient assisted living system with IoT embedded device for In-home monitoring of older adults”, Sensors, Vol. 23/5, p. 2673, https://doi.org/10.3390/s23052673.
[15] Graham, G. et al. (2025), “Chatbots in customer service within banking and finance: Do chatbots herald the start of an AI revolution in the corporate world?”, Computers in Human Behavior, Vol. 165, p. 108570, https://doi.org/10.1016/j.chb.2025.108570.
[43] Győrffy, Z. et al. (2023), “Older adults in the digital health era: Insights on the digital health related knowledge, habits and attitudes of the 65 year and older population”, BMC Geriatr, Vol. 23/1, p. 779, https://doi.org/10.1186/s12877-023-04437-5.
[14] Haddad, A. and C. Bhat (2025), “Telemedicine adoption before, during, and after COVID-19: The role of socioeconomic and built environment variables”, Transportation Research Part A: Policy and Practice, Vol. 192, p. 104351, https://doi.org/10.1016/j.tra.2024.104351.
[50] Herrera, L., L. Van Sickle and A. Podhradsky (2024), “Bridging the protection gap: Innovative approaches to shield older adults from AI-enhanced scams”, 2024 Cyber Awareness and Research Symposium, pp. 1-9, https://doi.org/10.1109/cars61786.2024.10778759.
[49] Hong, S. et al. (2013), “How old are you really? Cognitive age in technology acceptance”, Decision Support Systems, Vol. 56, pp. 122-130, https://doi.org/10.1016/j.dss.2013.05.008.
[42] Hutton, A. (2024), “Acceptance of artificial intelligence among older adults”, Wichita State University, https://hdl.handle.net/10057/29166.
[17] IMF (2025), Financial Access Survey (dataset), https://data.imf.org/en/datasets/IMF.STA:FAS (accessed on 20 October 2025).
[40] Kim, H., P. Freddolino and C. Greenhow (2023), “Older adults’ technology anxiety as a barrier to digital inclusion: A scoping review”, Educational Gerontology, Vol. 49/12, pp. 1021-1038, https://doi.org/10.1080/03601277.2023.2202080.
[34] Kim, P. (2024), “Adoption of autonomous vehicles among the aging population”, Issues In Information Systems, Vol. 25/2, pp. 24-35, https://doi.org/10.48009/2_iis_2024_103.
[13] Kopka, M., L. He and M. Feufel (2026), “Evaluating the accuracy of ChatGPT model versions for giving care-seeking advice”, Communications Medicine, Vol. 6/1, https://doi.org/10.1038/s43856-026-01466-0.
[64] Labo Société Numérique (2024), “Libraries on the front line of digital mediation”, Labo Société Numérique, https://labo.societenumerique.gouv.fr/en/articles/folder-libraries-first-on-line-for-digital-mediation.
[26] Leite, H. et al. (2024), “‘It’s not the boogeyman’: How voice assistant technology is bridging the digital divide for older people”, Technovation, Vol. 136, p. 103080, https://doi.org/10.1016/j.technovation.2024.103080.
[24] Lissitsa, S., E. Zychlinski and M. Kagan (2022), “The silent generation vs baby boomers: Socio-demographic and psychological predictors of the “gray” digital inequalities”, Computers in Human Behavior, Vol. 128, https://doi.org/10.1016/j.chb.2021.107098.
[4] Lynn, T. et al. (2022), “Digital Public Services”, in Digital Towns, Springer International Publishing, Cham, https://doi.org/10.1007/978-3-030-91247-5_3.
[65] Matisola, M. (2024), “Interactive videos as effective tools for media literacy education in communication and media courses”, Electronics, Vol. 13/23, p. 4738, https://doi.org/10.3390/electronics13234738.
[63] Mendels, J. and A. Weiner (2025), Quantifying Inclusion and Online Safety in a Multicultural Society, Israel Internet Association, Petah Tikva, https://en.isoc.org.il/wp-content/uploads/2025/01/ISOC-IL-Quantifying-online-Inclusion-and-Safety-Arab-Israelis-2025.pdf.
[60] Ministry News (2024), “복잡한 키오스크 사용법, ‘한글햇살버스’서 알려드려요 (Learn how to use complicated kiosks on the ‘Hangeul Sunshine Bus’)”, https://www.korea.kr/news/policyNewsView.do?newsId=148926978 (accessed on 21 August 2025).
[61] MSIT (2025), “MSIT’s work plan for 2025”, https://www.msit.go.kr/eng/bbs/view.do?sCode=eng&mId=4&mPid=2&bbsSeqNo=42&nttSeqNo=1065 (accessed on 14 September 2025).
[2] Mubarak, F. and R. Suomi (2022), “Elderly forgotten? Digital exclusion in the information age and the rising grey digital divide”, Inquiry, Vol. 59, https://doi.org/10.1177/00469580221096272.
[51] Napal, D. (2024), “Imitation is the sincerest form of fraudulent activity: Artificial intelligence in financial scams against older adults”, Bifocal, Vol. 45/5, https://www.americanbar.org/groups/law_aging/publications/bifocal/vol45/vol45issue6/artificialintelligenceandfinancialscams.
[38] Nawaz, M. et al. (2025), “Evaluation of barriers to artificial intelligence adoption: Grey multi-criteria decision-making”, Grey Systems: Theory and Application, Vol. 15/4, pp. 732-754, https://doi.org/10.1108/gs-12-2024-0147.
[21] Neves, B., J. Waycott and S. Malta (2018), “Old and afraid of new communication technologies? Reconceptualising and contesting the ‘age-based digital divide’”, Journal of Sociology, Vol. 54/2, https://doi.org/10.1177/1440783318766119.
[69] OECD (2026), “Share of adults who feel AI will have a positive impact on their life”, OECD Going Digital Toolkit, https://goingdigital.oecd.org/indicator/82 (accessed on 27 March 2026).
[5] OECD (2026), ICT Access and Usage Database, https://oe.cd/dx/ict-access-usage (accessed on 27 March 2026).
[55] OECD (2025), “AI and the future of social protection in OECD countries”, OECD Artificial Intelligence Papers, No. 42, OECD Publishing, Paris, https://doi.org/10.1787/7b245f7e-en.
[57] OECD (2025), “Enhancing Access to and Sharing of Data in the Age of Artificial Intelligence”, OECD Policy Briefs, No. 13, OECD Publishing, Paris, https://doi.org/10.1787/23a70dca-en.
[68] OECD (2025), Governing with Artificial Intelligence: The State of Play and Way Forward in Core Government Functions, OECD Publishing, Paris, https://doi.org/10.1787/795de142-en.
[9] OECD (2025), Government at a Glance 2025, OECD Publishing, Paris, https://doi.org/10.1787/0efd0bcd-en.
[59] OECD (2025), “OECD.AI Policy Observatory: Austria’s programme for AI literacy and skills”, https://oecd.ai/en/dashboards/policy-initiatives/programme-for-ai-literacy-and-skills (accessed on 16 March 2026).
[16] OECD (2025), “Safeguarding consumers’ access to cash in the digital economy: Policy considerations and approaches”, OECD Business and Finance Policy Papers, No. 81, OECD Publishing, Paris, https://doi.org/10.1787/189970b4-en.
[53] OECD (2025), Social Connections and Loneliness in OECD Countries, OECD Publishing, Paris, https://doi.org/10.1787/6df2d6a0-en.
[3] OECD (2024), OECD Digital Economy Outlook 2024 (Volume 1): Embracing the Technology Frontier, OECD Publishing, Paris, https://doi.org/10.1787/a1689dc5-en.
[47] OECD (2024), OECD Digital Economy Outlook 2024 (Volume 2): Strengthening Connectivity, Innovation and Trust, OECD Publishing, Paris, https://doi.org/10.1787/3adf705b-en.
[10] OECD (2024), Recommendation on Human-Centred Administrative Public Services, OECD/LEGAL/0503, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0503 (accessed on 27 October 2025).
[18] OECD (2024), “Regulatory approaches to Artificial Intelligence in finance”, OECD Artificial Intelligence Papers, No. 24, OECD Publishing, Paris, https://doi.org/10.1787/f1498c02-en.
[37] OECD (2024), “The OECD Truth Quest Survey: Methodology and findings”, OECD Digital Economy Papers, No. 369, OECD Publishing, Paris, https://doi.org/10.1787/92a94c0f-en.
[7] OECD (2022), “OECD Good Practice Principles for Public Service Design and Delivery in the Digital Age”, OECD Public Governance Policy Papers, No. 23, OECD Publishing, Paris, https://doi.org/10.1787/2ade500b-en.
[8] OECD (2021), The E-Leaders Handbook on the Governance of Digital Government, OECD Digital Government Studies, OECD Publishing, Paris, https://doi.org/10.1787/ac7f2531-en.
[54] OECD (2019), Artificial Intelligence in Society, OECD Publishing, Paris, https://doi.org/10.1787/eedfee77-en.
[27] Osei-Asiamah, J. et al. (2025), “AI-driven adaptive user interfaces for enhancing interaction and accessibility for elderly users based on cognitive and physical abilities”, 2025 International Conference on Sustainability, Innovation & Technology (ICSIT), pp. 1-6, https://doi.org/10.1109/icsit65336.2025.11295099.
[58] Ramadan, R. et al. (2026), “The double-edged sword: A critical review of foundational medical datasets for AI benchmarks, biases, and the future of equitable healthcare”, Neurocomputing, Vol. 675, p. 132919, https://doi.org/10.1016/j.neucom.2026.132919.
[22] Reneland-Forsman, L. (2018), “‘Borrowed access’ – The struggle of older persons for digital participation”, International Journal of Lifelong Education, Vol. 37/3, pp. 333–344, https://doi.org/10.1080/02601370.2018.1473516.
[23] Rojo-Pérez, F., G. Fernández-Mayoralas and V. Rodríguez-Rodríguez (2021), Active Ageing and Quality of Life: A Systematized Literature Review, Springer, Cham, https://doi.org/10.1007/978-3-030-58031-5_4.
[39] Rony, M. et al. (2025), “Attitudes of older patients toward artificial intelligence in decision-making in healthcare”, Journal of Medicine, Surgery, and Public Health, Vol. 6, p. 100193, https://doi.org/10.1016/j.glmedi.2025.100193.
[35] Sacar, S. et al. (2024), “Designing age-inclusive interfaces: Emerging mobile, conversational, and generative AI to support interactions across the life span”, ACM Conversational User Interfaces 2024, pp. 1-5, https://doi.org/10.1145/3640794.3669998.
[12] Saigí-Rubió, F. et al. (2022), “The current status of telemedicine technology use across the World Health Organization European Region: An overview of systematic reviews”, Journal of Medical Internet Research, Vol. 24/10, p. e40877, https://doi.org/10.2196/40877.
[32] Scott, S. (2026), “Tech support”, Stanford Magazine, Vol. Spring, https://stanfordmag.org/contents/tech-support.
[41] Shandilya, E. and M. Fan (2022), “Understanding older adults’ perceptions and challenges in using AI-enabled everyday technologies”, Proceedings of the Tenth International Symposium of Chinese CHI, pp. 105-116, https://doi.org/10.1145/3565698.3565774.
[56] Shumailov, I. et al. (2024), “AI models collapse when trained on recursively generated data”, Nature, Vol. 631/8022, pp. 755-759, https://doi.org/10.1038/s41586-024-07566-y.
[52] Sima, H. et al. (2026), “Fight fire with fire: How does AI-powered technology empower the elderly anti-AI fraud through a socio-technical systems theory lens?”, Journal of Consumer Behaviour, Vol. 25/3, pp. 998-1017, https://doi.org/10.1002/cb.70106.
[66] Trischler, J. and J. Westman Trischler (2022), “Design for experience – a public service design approach in the age of digitalization”, Public Management Review, Vol. 24/8, pp. 1251-1270, https://doi.org/10.1080/14719037.2021.1899272.
[45] Veras, M. et al. (2025), “Ethics and equity challenges in telerehabilitation for older adults: Rapid review”, JMIR Aging, Vol. 8, pp. e69660-e69660, https://doi.org/10.2196/69660.
[46] Vercruyssen, A. et al. (2023), “How “basic” is basic digital literacy for older adults? Insights from digital skills instructors”, Frontiers in Education, Vol. 8, https://doi.org/10.3389/feduc.2023.1231701.
[29] Wang, M. (ed.) (2025), “Factors influencing elderly adoption of artificial intelligence robots for aging-in-place: Motivators, barriers, and emotional impact”, Work, Aging and Retirement, https://doi.org/10.1093/workar/waaf016.
[44] Wong, A. et al. (2025), “Exploring older adults’ perspectives and acceptance of AI-driven health technologies: Qualitative study”, JMIR Aging, Vol. 8, pp. e66778-e66778, https://doi.org/10.2196/66778.
[33] Zandieh, R. and R. Acheampong (2021), “Mobility and healthy ageing in the city: Exploring opportunities and challenges of autonomous vehicles for older adults’ outdoor mobility”, Cities, Vol. 112, p. 103135, https://doi.org/10.1016/j.cities.2021.103135.