Guest editorial: AI for a better future – advances, challenges and future research directions Free

Artificial intelligence (AI) is increasingly integrated into everyday digital environments and is transforming how individuals interact with technologies, organisations and service systems. AI-powered applications, such as conversational agents, recommendation systems, generative AI tools and service robots, are now widely integrated into consumer services, organisational processes and digital platforms. As these technologies continue to evolve, they are reshaping how value is created, delivered and experienced across digital ecosystems.

Research across the information systems (IS) and organisational behaviour disciplines has increasingly recognised the transformative potential of AI technologies. Previous studies have explored how organisations manage AI deployment, how intelligent systems influence decision-making processes and how digital technologies reshape socio-technical systems within organisations and digital markets (Berente et al., 2021; Rai et al., 2019). These lines of research highlight that AI should not be understood solely as a technological artefact but rather as a socio-technical phenomenon embedded in complex organisational and societal contexts.

In marketing and service contexts, AI technologies are increasingly influencing how organisations interact with consumers and deliver digital services. AI-powered systems, such as voice assistants, chatbots, recommendation algorithms and service robots, enable firms to automate interactions, personalise services and enhance decision-making processes. Emerging research demonstrates that AI technologies can enhance customer experiences, support service efficiency and augment human capabilities in both organisational and consumer environments (Ameen et al., 2021, 2025; Sabz et al., 2026). In addition, recent advances in generative AI are rapidly transforming industries by enabling organisations to generate content, analyse large volumes of information and support decision-making processes in new ways (Ameen et al., 2022; Feliciano-Cestero et al., 2023). For example, AI-driven predictive capabilities, such as Google DeepMind's AlphaGenome, reshape enquiries relating to healthcare innovation and personalised services, spanning medical prevention, diagnosis and treatment, to promote a better future (UK Parliament, 2026).

Despite these opportunities, the rapid proliferation of AI technologies also raises important challenges for organisations, consumers and society. Concerns related to algorithmic bias, misinformation, privacy risks and digital manipulation are becoming increasingly prominent as AI systems gain greater autonomy and influence over digital interactions. AI-generated content, for example, can blur the boundaries between human- and machine-generated information, raising new questions about transparency, accountability and trust. Furthermore, AI-based decision systems may inadvertently reinforce existing inequalities or introduce new forms of digital vulnerability, particularly for individuals with lower levels of digital literacy or limited access to technological resources.

These tensions highlight the need for deeper scholarly understanding of how AI technologies shape consumer behaviour, service experiences, organisational practices and societal outcomes. Addressing these challenges requires interdisciplinary perspectives that bridge marketing, service, organisational behaviour and IS research. In particular, scholars must examine how AI-enabled technologies influence human experiences, social relationships, decision-making processes and perceptions of fairness in digital environments.

Against this backdrop, this special issue on “AI for a better future” aims to advance scholarly understanding of the role of AI in shaping consumer experiences, organisational practices and societal outcomes. The special issue focuses on behavioural and interdisciplinary research examining how different forms of AI technology, including generative AI, conversational agents, service robots and intelligent systems, affect individuals, organisations and society. By bringing together diverse theoretical and methodological perspectives, the contributions aim to shed light on how AI can be designed, deployed and governed in ways that enhance well-being, support responsible innovation and contribute to a more inclusive and sustainable digital future.

The papers in this special issue examine how AI is reshaping digital interactions across consumer, service, workplace and societal settings. Collectively, the papers show that AI not only supports well-being, engagement and decision-making but also introduces new forms of vulnerability, anxiety, ethical tension and misinformation. The contributions can be grouped into four broad themes: (1) AI and consumer well-being in human-AI relationships; (2) AI in consumer and service experiences; (3) AI in workplace and service recovery settings and and (4) AI, responsibility and the digital society. Figure 1 summarises the organising framework that illustrates the themes derived from this special issue.

Several papers explore the emotional and psychological role of AI in people's lives. Ng et al. (2026) examine relationships with AI companions by integrating the triangular theory of love and attachment theory. Using survey data from 527 users of AI companion applications, this study investigates how intimacy, passion and commitment shape attachment to AI companions and how that attachment influences social well-being. The findings show that these three components of love significantly shape attachment and that both interactive engagement and emotional attachment positively affect social well-being. This paper also identifies a moderating role for sweet deception, showing that affectionate but deceptive communication can strengthen the links between attachment and social well-being.

Feng et al. (2026) study whether AI fitness instructors can help alleviate loneliness in digital workout environments. Based on a quasi-experimental design with 592 participants, this paper compares AI and human instructors and examines the roles of psychological closeness, co-presence and enjoyment. The findings show that human instructors produce stronger psychological closeness overall, but AI instructors can also help reduce loneliness, especially when users experience high co-presence and enjoyment. This paper therefore demonstrates that AI can support emotional well-being in digitally mediated fitness contexts, although its effectiveness depends on the quality of the interaction experience.

Bakr et al. (2026) extend this concern with well-being by examining wearable self-trackers and digital vulnerability. Drawing on practice theory and interviews with 30 Fitbit users, this study identifies three usage patterns – light, fluctuant and intensive use – and shows that reflexivity around physical activity identity and goals shapes vulnerability to harms. Rather than treating self-tracking as uniformly beneficial, this paper highlights how AI-enabled monitoring technologies can create differentiated forms of risk depending on users' identities, goals and critical capacity.

These papers show that AI can support social connection, health and self-management, but that such benefits are contingent, relational and sometimes accompanied by new forms of emotional or digital vulnerability.

The second theme examines how AI shapes consumer engagement, brand relationships and service evaluations. Ekinci et al. (2026) investigate human versus AI-generated metahuman influencers in online brand engagement. Across two experiments with real brands, this study compares human influencer endorsement, metahuman influencer endorsement and brand-only conditions. The findings show that human influencers significantly enhance online brand engagement relative to a brand-only condition, primarily through attachment transfer. Metahuman influencers can also drive online brand engagement, but the moderating effect of influencer-brand fit applies only to human influencers. This paper therefore provides a nuanced view of the effectiveness of virtual influencers.

Qiu et al. (2026) examine chatbot guidance in product assembly contexts. Through an online scenario experiment and a laboratory experiment, this paper investigates whether static and video chatbot guidance affect perceived meaningfulness, brand intimacy and service evaluation. The findings reveal a serial mediation effect: chatbot guidance reduces perceived meaningfulness, which in turn lowers brand intimacy and, subsequently, service evaluation. This negative effect is strongest for video chatbot guidance, although it becomes weaker when assembly tasks are more complex or when consumers have lower hands-on ability. This study therefore shows that AI support does not always enhance service experiences; in some contexts, it can reduce the sense of effort and accomplishment that consumers value.

Al Amin et al. (2026) provide a systematic review of anthropomorphic AI in customer journeys. Analysing 122 articles, this study identifies key AI traits and roles and proposes the interaction-activation-outcome framework to explain how anthropomorphic AI activates cognitive, affective and social responses that shape behavioural, hedonic, utilitarian and sustainable outcomes across customer journeys. This paper provides an integrative conceptual foundation for understanding how anthropomorphic AI influences consumer responses.

These papers highlight that while AI technologies can enrich customer journeys, improve online brand engagement and provide chatbot guidance, their impact depends on whether they enhance or diminish meaningful consumer involvement.

The third group of papers focuses on AI in organisational and service environments. Fang et al. (2026) investigate how AI shapes job anxiety in the workplace. Using survey data from 675 respondents and drawing on the transactional theory of stress and coping, this study examines how different AI features influence AI-related stress and job anxiety. The findings reveal a dual effect: AI explainability increases job anxiety, whereas algorithm transparency reduces it. This study therefore demonstrates that different design characteristics of AI systems can either intensify or alleviate employee anxiety.

Lan et al. (2026) examine AI agents in tricky complaint situations where firms must respond to users without genuinely resolving the issue. Across three experiments, this study identifies two intention-hiding strategies – evasive hiding and rationalised hiding – and tests how these interact with agent type. The findings show that users are more willing to forgive AI agents than human agents when evasive hiding is used, whereas human agents receive more favourable responses when rationalised hiding is used. These effects operate through perceived negative motives and perceived sincerity. This study further shows that different types of AI capabilities also influence these outcomes.

These studies show that AI adoption in organisations affects not only operational efficiency but also employee emotions and customer interpretations of organisational intent.

The final set of papers addresses broader ethical, governance and societal issues. Kirshner and Lawson (2026) examine how competitive pressure shapes responsible AI deployment. Across three experimental studies, they distinguish between horizontal unethical competition, where many competitors adopt similar unethical AI practices, and vertical unethical competition, where a competitor engages in severe unethical behaviour. The findings show that horizontal unethical competition increases the likelihood of launching unethical AI regardless of regulatory focus. By contrast, under vertical unethical competition, prevention-focused goals can counteract the pressure and promote more responsible decisions.

Finally, Agarwal et al. (2026) provide a meta-synthesis on generative AI and disinformation. Integrating qualitative research on fake news, misinformation, disinformation and deepfakes, this study develops an integrated framework grounded in social information processing theory. The findings show that generative AI acts as a double-edged sword: while it facilitates the creation and dissemination of disinformation, it simultaneously offers analytical capabilities that can help detect and mitigate false content. This study therefore highlights the need for governance mechanisms, technological detection tools and stronger media literacy.

Collectively, these papers broaden the focus of the special issue beyond individual-level interactions with AI and show that achieving a better AI-enabled future also depends on responsible governance, ethical decision-making and institutional responses to digital harms.

The papers in this special issue make four main contributions. First, they show that AI is increasingly involved in the emotionally significant aspects of life, from companionship and loneliness to self-tracking and well-being. As AI's empathetic capabilities increasingly simulate close human relationships, new forms of consumer attachment and well-being emerge. Second, they demonstrate that AI's effects on consumer and service experiences are not uniformly positive; they may strengthen engagement in some contexts while reducing meaningfulness or authenticity in others. Third, they highlight the organisational implications of AI adoption, showing that AI shapes employee experiences, customer engagement and perceptions of organisational intent. For example, the integration of AI changes employees' interactions with customers and the nature of their daily work practices, introducing job-related strain and heightened stress. Fourth, they extend the discussion to societal and governance challenges, including responsible AI deployment and the growing problem of AI-enabled disinformation. Our special issue moves the conversation beyond overly optimistic or overly pessimistic views of AI to show that AI's contribution to a better future depends on how technologies are designed, governed and integrated into human contexts.

Building on the contributions of this special issue, there is a growing need to deepen and broaden scholarly understanding of how AI can contribute to a better future for individuals, organisations and society. The papers included in this issue collectively demonstrate that AI technologies are increasingly embedded in consumer experiences, service interactions, organisational processes and digital ecosystems. While these technologies can enhance well-being, engagement and efficiency, they also introduce ethical, psychological and societal challenges.

Moreover, AI should not be examined solely through a technological lens. Instead, its implications emerge through complex interactions between technological capabilities, human behaviours, organisational practices and societal contexts. AI systems increasingly function as social and emotional actors, service providers, decision-support tools and content creators. As such, their influence extends beyond operational efficiency to shape human experiences and relationships.

As AI technologies continue to evolve rapidly, particularly with advances in generative AI, anthropomorphic agents and autonomous systems, future research must address both the opportunities and the risks associated with their integration into everyday life. Researchers should therefore move beyond narrow technological perspectives and examine AI as a socio-technical phenomenon embedded within broader service and digital ecosystems.

Based on the insights that emerge from the contributions in this special issue, we propose four key avenues for future research: (1) AI, vulnerability and inclusive digital experiences; (2) AI-driven customer experience management; (3) AI and the future of work and (4) AI and responsible digital society. These research directions offer promising opportunities to advance theory and inform practice at the intersection of marketing, service research and digital technologies (illustrated in Figure 2).

AI is increasingly integrated into everyday consumer services, from conversational agents and recommender systems to health technologies and digital companions. While these technologies offer important benefits, such as convenience, personalisation and enhanced service accessibility, they may also introduce new forms of consumer vulnerability (Bentley et al., 2024). Individuals may experience information asymmetries, algorithmic manipulation, emotional dependence on AI systems or reduced ability to critically evaluate automated recommendations.

The marketing and service literatures have traditionally conceptualised vulnerability as arising from individual characteristics or situational factors. However, AI-enabled environments introduce new structural forms of vulnerability related to algorithmic opacity, data exploitation and unequal digital capabilities. Consumers with lower digital literacy, limited access to technological resources or greater reliance on AI-enabled services may be particularly exposed to such risks.

As AI technologies become more autonomous and emotionally intelligent, understanding how these systems influence consumer autonomy, identity, emotions and well-being becomes increasingly important. Future research should therefore examine how organisations can design AI-enabled services that enhance accessibility and inclusion while safeguarding vulnerable individuals from potential harms. Future research could address the following questions:

1. How do AI-enabled services reshape different forms of consumer vulnerability in digital environments?

2. Which consumer groups are most exposed to risks related to algorithmic decision-making, emotional AI interactions or automated recommendations?

3. How does algorithmic transparency influence vulnerable consumers' trust and decision-making?

4. What design principles can help organisations develop inclusive AI-enabled services that support consumer autonomy and well-being?

5. How can firms balance personalisation and automation while protecting vulnerable consumers?

AI has led to various new opportunities to deliver customised experiences and reshape value delivery for stakeholders. AI-mediated touchpoints enhance efficiency, personalisation and the ongoing transformation of customer journeys across a wide range of industries. Organisations can now capitalise on data analytics and real-time insights to better understand customer preferences, anticipate needs and deliver tailored interactions. These opportunities highlight AI's ability not only to improve efficiency but also to deepen immersive, relational and emotional connections between customers and organisations. Empathetic capabilities (Huang and Rust, 2024) in customer journeys contribute to new forms of experiential value, fostering stronger emotional bonds. Service experiences also involve third-party actors who adopt different roles in service encounters, such as bystanders or endorsers (Abboud et al., 2021). Therefore, consideration should be given to how AI can effectively orchestrate seamless touchpoints for multi-actor goal alignment.

As AI also becomes increasingly agentic, novel forms of customer journeys and strategies are being reinvented (Bornet et al., 2025), which call for innovative forms of experience design. However, power imbalances increase the likelihood that consumers experience negative emotions or disengagement (Abboud et al., 2023). In the AI-mediated context, where customers may perceive a lack of autonomy and limited ability to make decisions due to a lack of control or agency, firms may need effective and adaptive AI-agent deployment to shift control back to customers (Jha et al., 2026).

The increased integration of AI into service environments thus entails notable challenges. For example, the deployment of service robots in frontline settings introduces complexities that affect how customers interact, influencing overall service dynamics and reactions (Phillips et al., 2023). Similarly, engaging with AI agents may negatively affect customers' sense of meaningfulness, thereby limiting positive brand evaluations. As a result, new forms of human–machine engagement are emerging (Azer and Alexander, 2025), highlighting the need for increased explainability and transparency in value delivery. Such challenges, coupled with customers' potential over-delegation to AI, disengagement or emotional detachment, potentially undermine value outcomes.

Given these opportunities and risks, several important areas warrant further exploration. These include the role of organisations in designing, managing, and delivering effective AI-driven customer experiences, the effective application of AI capabilities across different key sectors to promote greater access to services and insights into new patterns of customer engagement and disengagement with AI. Future work could focus on the following questions:

1. How do empathetic capabilities of AI agents (e.g. companion robots) impact customers' emotional engagement during service experiences and lead to long-term relationships?

2. How can the implications of increasingly autonomous AI capabilities be managed effectively by organisations when tailoring customer journeys to individual customers?

3. How can customers derive a sense of achievement and self-efficacy when interacting with AI agents and delegating actions and decisions?

4. How can AI effectively orchestrate interactions among multiple actors (e.g. customers, employees and third parties) in complex service ecosystems?

5. How can organisations achieve the right balance between technological innovation and customer-centred service design?

AI is transforming organisations by enhancing employee performance, improving organisational effectiveness and supporting problem-solving in complex situations (Raisch and Fomina, 2025). By automating routine tasks, augmenting decision-making and providing data-driven insights, AI enables employees to work more efficiently and focus on higher-value activities, thereby boosting overall productivity (Al Naqbi et al., 2024). At the organisational level, AI improves responsiveness and consistency in handling customer issues in service recovery scenarios (Huang and Rust, 2021).

However, the impact of AI extends beyond operational efficiency to shape employees' experiences with and perceptions of organisations. For example, the gratitude displayed by AI robots in human–AI collaboration can strengthen employees' organisation-based self-esteem, which in turn encourages organisational citizenship behaviours (Chen et al., 2025). Furthermore, employees' use of generative AI's creative capabilities enhances their own creativity because they are more likely to see difficult tasks as opportunities rather than obstacles (Zhao et al., 2025). However, a phenomenon in which service workers blindly rely on AI to complete their tasks, termed “AI complacency” (Le and Kunz, 2026), is a growing issue that poses a threat to effective human–AI collaboration.

In parallel, emerging research highlights improved applications of AI in service recovery, such as incorporating humour to humanise chatbots that might otherwise be perceived as cold or socially inept when replacing human representatives (Shin et al., 2023). Likewise, the way AI agents communicate during service failures can significantly shape customers' evaluations of the firm, enhancing recovery effectiveness. There are several possible research directions:

1. How do different AI applications (e.g. machine learning, deep learning, NLP and predictive vs generative AI) impact employees' perceived efficacy, and to what extent do these effects vary depending on task type and the level of human-AI collaboration?

2. What AI features (e.g. accuracy, consistency, adaptability and controllability) can be best leveraged to enhance employee well-being?

3. What are novel forms of human-AI collaboration in organisational settings, and how do they influence employees' overall work experience?

4. What are the tensions, the unintended consequences and the dark side of human-AI collaboration, particularly in service contexts?

5. How does reliance on AI-enabled service recovery impact employees' perceived autonomy, role identity and job satisfaction, and what are the downstream effects on their interactions with customers?

6. What AI designs (e.g. speed, transparency, emotional expression, apology or explanation framing and employee override capability) enhance or undermine the effectiveness of service recovery when replacing human service representatives?

AI technologies have become a foundational infrastructure of contemporary society. From hiring decisions to access to healthcare, financial services and public spaces, they effectively mediate participation in social, economic and civic life. As their influence expands, broader societal discussions have highlighted the potential negative implications of AI adoption, including increased social inequality and ethical concerns regarding polarisation, fairness, privacy and sharing of misinformation (Wirtz et al., 2023). For example, research has shown how hiring applications and credit scoring systems powered by AI may perpetuate existing human biases, thus reinforcing racial and gender stereotypes and producing inequitable outcomes (Hunkenschroer and Luetge, 2022). Similarly, AI facial recognition technologies often produce higher error rates for racialised and gender-diverse groups, reinforcing patterns of surveillance and marginalisation (Panarese et al., 2025). When deployed in public and in third spaces, AI systems risk amplifying exclusion and deepening social divides by tailoring interactions unequally or visibly favouring certain users (Paluch et al., 2026).

These issues are not simple technical glitches but reflect deeper structural dynamics. Such systems do not merely mirror societal biases; they formalise them within algorithmic processes, apply them consistently and frequently magnify their impact (Bonezzi and Ostinelli, 2021). Due to their reliance on vast quantities of both public and private data, AI technologies pose substantial risks to data security and privacy, including vulnerabilities to data breaches, manipulation and identity theft, as well as enabling practices such as highly personalised advertising, pervasive online tracking and ubiquitous surveillance (Dorotic et al., 2024).

AI also offers unprecedented opportunities for the creation and the distribution of more sophisticated forms of misinformation. For instance, deepfakes – digitally manipulated synthetic media – can be used to deceive consumers in the marketplace (Li and Wan, 2023), while generative AI tools, such as ChatGPT, may generate false, inaccurate or misleading content (Larsen et al., 2025). Researchers have also warned against the increased use of AI-enabled chatbots as tools to help people manage their mental health, emphasising how such tools can increase the risk of harm to users and result in addictive behaviours (De Freitas and Cohen, 2024; Marriott and Pitardi, 2024).

Nevertheless, AI also offers considerable potential to enhance the efficiency and effectiveness of public services, support administrative systems and governance and contribute to social good, for example, by aiding efforts to combat climate change (Cowls et al., 2023). Thus, AI integration in the public sphere brings a tension between market-driven priorities and societal and public values. The development of AI systems is often guided by commercial imperatives, such as efficiency, scalability and profit, which can clash with democratic ideals like fairness, transparency and inclusion. These conflicts are particularly evident in automated decision-making systems, where optimisation processes may yield outcomes that reproduce or even amplify social disparities.

Future research should explore this tension and examine how AI technologies can be designed, developed and deployed in ways that allow consumers and citizens to benefit from the technology while preserving societal and public values. Future studies could address the following questions:

1. How can AI technologies be designed to reduce and overcome discrimination instead of reinforcing it?

2. Under what circumstances does adaptive AI enhance inclusivity, and when might it instead fragment and divide users' interaction?

3. Which elements influence how customers perceive issues, such as privacy, ethics and fairness, when engaging with AI technologies?

4. How should AI-enabled well-being applications be regulated to minimise potential risks and protect users?

5. Which ethical principles should inform the responsible design and deployment of AI technologies in sensitive sectors, such as healthcare, elder care and education?

6. How can harmful or unethical uses of AI technology, such as surveillance, behavioural manipulation or coercive nudging, be effectively prevented?

The contributions of this special issue highlight the need for interdisciplinary research to drive impact and collaboration between consumers, organisations, AI developers and policymakers to shape a better AI-driven future. Although AI innovations lead to a myriad of positive benefits, tensions and challenges also arise with AI adoption and usage. These have been explored using various methodological and contextual lenses, from looking at emotionally close human–AI relationships to macro-societal implications. Recent research progress highlights how AI can contribute to a better future for consumers by fostering intimate relationships through synthetic interactions, while such psychological dynamics also give rise to vulnerability from day-to-day digital experiences. Mirroring consumer contexts, AI reshapes how work is conducted and alters employees' job roles, motivation and well-being. Such duality is also present in organisations, where AI-driven experiences enhance productivity but may also give rise to consumer detachment or perceptions of illegitimacy due to AI-generated misinformation. Taken together, these issues raise important social questions about how effective policymaking and regulatory frameworks can be holistically designed to manage these dilemmas.

Moving forward, we call for further research that acknowledges the dual nature of AI and promotes responsible AI management to mitigate dilemmas that arise from AI integration or automation. There is a growing need for institutions to adopt corporate digital responsibility strategies and deploy ethical AI practices to maximise welfare, minimise AI-related societal tensions and reduce risks such as erosion of trust, algorithmic biases and disinformation. In addition, the imperative to build a responsible digital society requires new ways of collaborating, as well as more robust AI systems, frameworks and policies that promote inclusive AI literacy and upskilling. In light of persistent digital vulnerabilities, technologies should be designed, managed and adopted to alleviate potential harms and promote equitable, healthier and more sustainable communities in the age of AI.

We are very grateful to Professor Christy Cheung, the Editor-in-Chief of Internet Research, as well as the editorial team for their outstanding support for this special issue. We thank all contributing reviewers who provided thoughtful and constructive feedback for the manuscripts.