Psychology Education Needs an Ambitious Vision of AI Proficiency – Comment on Robinson et al. (2026)

The proposal to embed artificial intelligence (AI) proficiency into the psychology major is timely and of potentially significant importance to the field. By mapping AI proficiency onto the APA's Guidelines for the Undergraduate Psychology Major 3.0 (APA, 2023b), the Principles for Undergraduate Education in Psychology (APA, 2023a), and The Skillful Psychology Student (Naufel et al., 2018), Robinson et al. (2026) provide psychology education with diverse inspiration for curriculum development. The layered approach between foundational AI literacy and more advanced AI fluency acknowledges AI proficiency as a developmental continuum. In this comment, I argue that the proposed framework and ambitions could be extended further toward a vision that positions the role of psychology and its graduates not only as proficient users and critical evaluators of AI but also as active contributors to the human-centered design and governance of AI that contributes to human flourishing.

Are We Literate Yet?

For a start, I cannot resist the temptation to point out the recurring waves of urgency around integrating emerging technologies into disciplinary curricula. As media and technologies have evolved, so too have the literacies considered essential for participation in society. Earlier calls for television literacy were followed by demands for computer literacy, media literacy, and digital literacy (Bawden, 2001). More recently, AI literacy has emerged as the newest addition to this lineage.

Each wave has produced curricular change, I guess in a rather evolutionary and natural way. It is not the question of whether new technology will find its way into educational curricula but rather how quickly, how deeply embedded, and how well. Each new literacy has raised a recurring set of questions: What do we have to learn now? What existing content may have to be displaced or foundationally renewed to address the ever-expanding literacy requirements? Who possesses the expertise and resources, including the time, to develop and maintain new curricula given rapid technological change? How do we strike the right balance when embedding any form of literacy into education? If it is defined too narrowly, it may quickly become outdated. If it is defined too broadly, it risks becoming redundant with foundational skills such as critical thinking, methodological rigor, and ethical reasoning that disciplines like psychology already cultivate.

Calling out this recurring pattern of literacies is not meant to diminish the call for AI proficiency. It could indeed be true that generative AI in particular is more transformative than previous technologies because it does not merely mediate information but generates, interprets, and acts on it, becoming a new actor to be dealt with across use cases. My point is that this historical context should make us especially deliberate about what kind of AI proficiency we aim for and with what purpose. If each new technology produces a call for literacy focused on competent use, the resulting curricula risk becoming useless with the next update. A more sustainable educational strategy emphasizes the underlying principles that remain relevant regardless of which specific tools dominate the landscape in any given year. And I think psychology is in an especially promising position to help better understand how AI shapes our world if we can indeed help future psychologists reap the benefits of the intersection between AI proficiency and psychology proficiency.

AI Proficiency to Shape AI

The target article's framework moves beyond a “learn to use AI” message, particularly at the fluency level. The authors envision graduates who understand AI-enhanced workflows, who can critically assess AI outputs in the context of psychological science, and who recognize the ethical and practical complexities of AI deployment. Tables 3–7 in the target article, for instance, include competencies such as building decision support tools, ensuring compliance with ethical guidelines, and developing human–AI collaboration frameworks. All of this points toward a role for future psychologists that becomes more proactive and design-focused with AI and complements the traditional backward-looking evaluative stance that psychology is often more comfortable taking (Landers, 2026).

My observation is not that the paper neglects this new role of psychologists. Rather, this role still remains somewhat secondary to a more dominant theme: AI proficiency is primarily framed in terms of what psychologists can accomplish with AI, rather than what they can contribute to AI and its responsible implementation in society. For instance, in Table 1 of the paper (which highlights suggestions for integrating AI proficiency with APA 3.0 goals), commonly featured competencies concern using AI for information gathering, employing AI in research and analysis, leveraging AI for writing and creating teaching content, and applying AI tools in professional development. Here and in the main text, the focus seems to be more on AI as something that helps psychologists rather than the more ambitious perspective that psychologists are people who help make AI better. As a specific example, take Goal 3's treatment of psychology's ethical values in relation to AI. Psychology does not merely have values that are relevant to responsible AI; it has theoretical, argumentative, and, in consequence, also normative foundations that should inform how AI systems are designed and governed. Embedding this claim explicitly in psychology curricula would signal to students that the role of psychology is not restricted to adapting to technology out of necessity but that psychologists can engage in shaping technology in a way that is beneficial for humans.

Three Domains Where Psychology Graduates Can Shape AI

I think the perspective on AI proficiency for psychologists can be more ambitious and explicitly integrate a design-oriented vision. Building on what Landers (2026) has articulated as psychology-in-the-loop – calling for more integrated engagement of psychological theory and method in the design, deployment, and evaluation of technical systems – I see at least three domains where psychology graduates could make unique contributions and become multipliers of psychological expertise and interests: evaluating AI systems, designing human-AI interaction, and contributing to AI governance, policy, and their operational implementation. Notably, these contributions will be substantially more meaningful when psychologists realize that they, much like any other discipline, are not in a position to solve the challenges of AI alone (Landers, 2026). An omission in the target paper's suggestions, to me, is thus highlighting that a central aspect of AI proficiency is the ability to readily engage with other disciplines on the topic of AI. Knowing the complexities of interdisciplinary work, knowing who to talk to and how to talk to them to understand and produce meaningful joint work, provides psychology graduates with a skill that is crucial for real-world impact in AI. For the case of AI, the psychological perspective becomes especially meaningful when integrated with the expertise of people trained in computer science, ethics and philosophy, and law (which, admittedly, is strongly colored by my own experiences).

Evaluating AI systems. Evaluation is core to a psychologist's methodological toolkit. For instance, experimental design, psychometrics, behavioral measurement, this is what is needed to rigorously evaluate whether AI systems actually improve the human outcomes they claim to enhance. This evaluation is needed across the entire AI lifecycle: before deployment, at runtime, and during auditing. I think it is no stretch to advance psychology curricula to enable psychology graduates to evaluate whether an AI-driven mental health chatbot produces therapeutic benefit or causes harm, to assess whether an AI hiring tool introduces or reduces bias in selection decisions, or to determine whether AI-based learning platforms are helping or hurting student motivation. These are urgent needs in every sector where AI meets humans and where psychologists not only bring the evaluation expertise but also the domain knowledge to craft meaningful evaluation pipelines that address real-world complexities. To become even more meaningful, the evaluation perspective of AI-proficient psychologists needs to be complemented with the technical expertise of computer scientists to automate and scale evaluation efforts.

Designing human-AI interaction. Psychological theory offers sustainable principles for optimizing the design of how humans and AI systems interact across a wide range of contexts. Theories from cognitive psychology inform when and how AI should present information to support rather than overwhelm human decision-making, whether in clinical settings, educational platforms, or consumer-facing applications. Research on judgment and decision-making provides guidance for designing human-AI interaction that fosters appropriate, rather than blind, reliance among patients, students, and citizens alike. Theories from developmental psychology can raise critical questions about how children and adolescents interact with AI systems and what safeguards their cognitive and emotional development requires. Clinical psychology contributes knowledge about when AI-driven mental health interventions help and when they risk harm, particularly for vulnerable populations. Theories grounded in social psychology on justice and intergroup dynamics shape how AI-supported decisions should be communicated to those affected, whether in hiring, criminal justice, or healthcare triage. Theories from work and organizational psychology explain how to design good AI-enhanced workplaces that contribute to human well-being. These principles are grounded in stable features of human behavior and will not become obsolete when the next generation of language models arrives (Bigman et al., 2026). To become even more meaningful, collaborating with individuals with in-depth training in moral theories could provide theoretically and normatively grounded design principles for human-AI interaction.

Contributing to AI governance, policy, and their operational implementation. AI regulation is being shaped worldwide, and these regulatory efforts require input from disciplines that understand human behavior, cognition, and well-being. The EU AI Act, for instance, is concerned with preventing risks to safety, health, and human rights posed by AI systems. Understanding these risks and mitigating them, for instance, by understanding when AI chatbots lead to risky dependence behavior or negative effects on mental health, understanding how to deal with AI-based emotion recognition technologies, or understanding how to design meaningful human oversight and control of AI, are just some of the main topics within current AI governance and safety discussions that need input from psychology. Psychology graduates who understand both the science and the policy landscape are positioned to contribute to regulatory bodies, standards organizations, and institutional ethics boards. To become even more meaningful, working together with legal scholars can help in drafting regulation and standards and can help operationalize abstract legal concepts into concrete guidelines for practice.

Ambition Versus Feasibility

I want to acknowledge a tension in my own argument. I mentioned that there will be practical challenges of integrating AI proficiency into already crowded curricula, and the recurring question of who possesses the expertise and resources to teach this material is very real. If I now argue for an even more ambitious scope, this tension only intensifies. If AI proficiency is conceived as an additional skill added on top of existing courses, it likely competes for scarce curricular space. If AI is framed as a cross-cutting theme that shapes psychology and that can be shaped by psychology, it may substantially change how we teach and what we teach. This not only requires “adding stuff on AI” but rethinking what to focus on in existing curricula and how to evaluate and assess (e.g., in exams) whether students possess the aimed-for competencies.

However, I also believe that in order to get to work, it can simply mean highlighting that the theories and methods we have in psychology provide a very solid foundation; what changes is how explicitly students are taught to see the(ir) relevance to AI. While the question of what to teach remains difficult, I do think psychology is in a relatively comfortable position because the principles of human cognition, motivation, and social behavior that inform responsible AI design change more slowly than AI tools themselves (Bigman et al., 2026). A course built around the implications of work design for implementing AI at work, of cognitive load for how to design interfaces, or of social cognition to explain how integrating AI into groups affects relations between group members will remain relevant even for the next update of your favorite chatbot.

What Is at Stake

If programs taught students primarily to use AI tools competently, they would produce graduates whose skills overlap substantially with those of graduates from any discipline who have received similar training. Moreover, if the focus were on the tools, with increasing capabilities and autonomy of these tools, they would become the ones who do the job, and we would train psychologists who have no clear role to play in the society of tomorrow. If, however, programs take the proficiency idea of the target article ambitiously and train students to integrate core psychological expertise with the design, evaluation, and governance of AI, psychology can contribute meaningfully to developing good AI for the future.

This distinction also matters in light of the APA's core mission hinted at in the target article, which probably aligns with the mission of psychology in many countries around the globe: advancing education, upholding ethics, and enhancing public impact (APA, 2023c). The user-proficiency dimension of the target article serves the first pillar well and engages the second. But public impact requires that psychological knowledge shapes the technologies that affect people's lives. This is more likely to happen when graduates see themselves not only as literate users of AI but as professionals whose expertise is needed in the rooms where AI systems are designed and governed. Adjacent fields, particularly human–computer interaction from a background of computer science, are already staking claims in the human-centered AI space, possibly drawing on psychological science without psychologists at the table. AI proficiency education could help ensure that the next generation of psychology graduates will sit at this table.

Conclusion

Simply put, my argument relating to the target article is that it could reach further. The AI fluency level already hints at a more active engagement with AI design and governance. What I am suggesting is that this dimension be made more central and more explicit. As AI reshapes every domain in which psychologists work, the most important message we can send our students may be that psychological science belongs wherever AI systems affecting humans are being built and regulated (Landers, 2026). A framework that makes this message its backbone would not only better prepare graduates for the emerging labor market but would also strengthen psychology's standing as a discipline that does not merely observe and adapt to technological change but actively shapes it for human flourishing.