Your new experience awaits. Try the new design now and help us make it even better

HYPOTHESIS AND THEORY article

Front. Comput. Sci., 18 July 2025

Sec. Human-Media Interaction

Volume 7 - 2025 | https://doi.org/10.3389/fcomp.2025.1524024

Ready or not? Psychologists’ perceptions of work readiness in the age of AI

  • Faculty of Medicine, Nursing and Health Sciences, School of Psychological Sciences, Clayton, VIC, Australia

The Fourth Industrial Revolution, driven by rapid AI advancements, presents significant challenges and opportunities for psychologists. As these dynamics evolve, it is crucial to prepare practitioners adequately, yet current accreditation frameworks seem insufficient. This qualitative study explored work readiness (WR) perceptions among fully registered psychologists in Australia within this shifting technological landscape. Semi-structured interviews with six participants, analysed using reflexive thematic analysis and a social constructionist approach, revealed three themes: (1) Human-Centric WR Skills Are Irreplaceable by AI, (2) AI Will Enhance Psychological Practice, and (3) Education Must Adapt to Emphasise AI-Integration and Human-Centric Qualities. The study identifies a critical gap between current accreditation, which often prioritises technical skills, and the humanistic capacities valued by psychologists. These findings advocate for a re-evaluation of psychology’s accreditation pathway, promoting a more holistic approach to preparing future psychologists for an AI-integrated profession. This research contributes foundational insights to the discourse on WR in psychology, emphasising the balance between technological proficiency and essential human skills in navigating AI integration. It also encourages dialogue between accreditation bodies and psychologists to reconcile differing WR perspectives and ensure the future preparedness of practitioners.

Introduction

In an era marked by rapid advancements in artificial intelligence (AI)–a technology designed to perform tasks traditionally requiring human intellect (Luxton, 2014)–both students and professionals must adapt to a fast-changing job market (World Economic Forum, 2019). Such shifts are certainly taking place in psychology, where technology is playing an increasingly central role (Innes and Morrison, 2021). In this context, ‘work readiness’ (WR) has gained attention, emphasising the importance of attitudes, skills, and attributes essential for workplace success (Caballero et al., 2011; Prikshat et al., 2019). However, a significant research gap exists in understanding WR within psychology, especially amid a shortage of trained psychologists and rising demand for mental health support in Australia (Neall et al., 2022). There is also a widening disconnect between academic training and professional demands as AI becomes more integrated into psychological practice (Hagstrom and Maranzan, 2019). Defining WR in contemporary psychological practice could help close this gap, ensuring that training keeps pace with the evolving demands of AI-driven change.

A theoretical model that reconciles the disconnect between accreditation frameworks and day-to-day clinical practice may help address this gap. Existing research has yet to conceptualise work readiness in psychology in a way that accommodates both the mechanistic priorities of training and regulation and the humanistic, relational nature of therapeutic work, particularly as AI integration accelerates. This study aims to address that theoretical gap by proposing a new framework for understanding how human and AI capabilities might be aligned in the future preparation of psychologists. In doing so, it repositions WR as a dynamic, relational construct rather than a static set of competencies.

WR has long been a strong predictor of job performance, career progression, and promotion (Casner-Lotto and Barrington, 2006), yet research into WR for psychology graduates is lacking. While efforts are underway to conceptualise and measure WR (Caballero et al., 2011; Coetzee, 2014), existing research has primarily concentrated on generic, transferable skills and attributes, rather than those which are discipline-specific. However, the literature suggests that discipline-specific, country-contextualised frameworks are necessary to account for the heterogeneous nature of WR (Prikshat et al., 2019). Given the unique demands of psychological practice, a psychology-specific conceptualisation and measure of WR are essential.

An adjacent concept is that of professional competency, which underpins the accreditation frameworks designed to prepare psychologists in Australia. This shift towards a competency paradigm is a global trend, led in Australia by the Psychology Board of Australia (PsyBA) and integrated into education by the Australian Psychology Accreditation Council (APAC; Gonsalvez et al., 2021). However, Gonsalvez et al. (2021) notes a key challenge in the competency-based approach is the concept of ‘competence’ itself, as the term conceives professional readiness as a static endpoint. Instead, Gonsalvez et al. argued that professional readiness is an ongoing, active process to maintain and enhance knowledge, skills, and attitudes amid rapidly evolving research and service delivery. Although Gonsalvez et al. proposed evolving the framework to make it more fluid, it is unclear how such fluidity can be incorporated into a conceptually-rigid framework that also requires accurate measurement of competency development.

It is important to note that the focus and scope of current WR conceptualisations differ substantially from those of the competency paradigm. ‘Competence’ is defined as one’s skill across multiple domains (e.g., assessment, diagnosis); and ‘competency’ refers to the particular skills within those domains (e.g., use of specific tests; Stevens et al., 2017). Specific personal attributes, for example, are not listed in PsyBA’s and Australian Health Practitioner Regulation Agency’s (AHPRA) Professional Competencies for Psychologists (2023), nor APAC’s Accreditation Standards for Psychology Programs (2019). These competency frameworks thus differ from existing conceptualisations of WR, in which professional skills are just one factor within a more holistic framework, which includes aspects like attitudes, personal development and soft skills (Caballero et al., 2011).

These aspects are centrally-important for practice, according to working psychologists. Salter and Rhodes (2018) found that nine of its 11 Australian clinical psychologist participants were dissatisfied with their training. They viewed it as overly focused on technical skills, neglecting vital elements of their development, such as authenticity, self-reflection, and the integration of their personal selves in therapy. Similarly, 20 Australian psychologists in Robinson et al. (2019) believed their training’s focus on CBT and technical skills marginalised the personal and relational aspects central to their practice and development. It thus appears that governing and accreditation bodies conceptualise professional WR differently to practicing psychologists.

Literature review

Having introduced perspectives on professional WR, it is now crucial to consider this concept alongside AI’s integration into psychological practice. Given that professional WR in psychology is under-researched, further considering AI’s impact makes this area of inquiry particularly novel. By examining emerging discourse surrounding the profession’s preparedness for AI, we can gain valuable insights into how these technological shifts are shaping, and potentially reshaping, understandings of WR.

Starting with technology-focused perspectives, it has been observed that AI innovation in psychology is outpacing both AI research and its integration into training (Hagstrom and Maranzan, 2019). Notably, many of the technical skills that underpin Australia’s accreditation frameworks are among the first candidates for AI automation (Innes and Morrison, 2021). However, educational and professional bodies have yet to fully acknowledge or address this emerging trend (Morrison and Innes, 2022). This raises important questions regarding how psychologist preparedness should be reconceptualised amid technological advancements.

Innes et al. (2022) suggest that the industry’s lack of response stems from the social perception that psychological practice is largely immune to AI, a perception shaped by an influential study by Frey and Osborne (2017). However, Innes and Morrison (2021) argue this study inaccurately portrays the psychologist’s role, focusing on skills like empathy and creativity. Whereas, they note, Australia’s accreditation bodies emphasise assessment, formulation, intervention, and evaluation. Innes and Morrison describe this approach emphasised by accreditation bodies as ‘mechanistic’, suggesting that it makes the psychologist’s role well suited for algorithmic automation. Prior literature also has also critiqued the rise of a mechanistic paradigm in psychology, noting it diminishes the human elements of therapy (Norcross, 2005; Salter and Rhodes, 2018). Thus, while evidence-based practice is important, over indexing on this approach may downplay crucial and irreplaceable human-centric WR qualities.

Another prevalent view is that AI will likely outperform humans in technical skills, as it exhibits reduced bias, fewer mistakes, greater efficiency, increased accessibility, encyclopaedic knowledge, and rapid learning abilities (Abrams, 2021; Bickman, 2020; Graham et al., 2019). These perspectives often explicitly or implicitly caution that, due to this superior technical performance, AI will inevitably make humans psychologists less relevant. Such views exhibit hallmarks of technological determinism, which posits that technology’s capabilities are the primary drivers of societal change (Țicău and Hadad, 2021). While there is consensus that strong evidence on the safety and effectiveness of these tools is required before they are fully embraced (Bickman, 2020; Hagstrom and Maranzan, 2019), once that emerges, Innes et al. (2022) suggests a fundamental ethical question arises: if best practice can be delivered by automation, then why should it not be? These deterministic stances resonate with the mechanistic social construction of psychological practice, both of which subordinate the role of humans in the unfolding landscape of AI’s integration.

Shifting to emerging socio-ecological perspectives on AI in psychology, which examine the interplay of social and environmental factors influencing the adoption of these tools (Kilanowski, 2017), it becomes clear that both clients and practitioners are increasingly incorporating technology into practice. The COVID-19 pandemic necessitated a surge in mental health services at a time when traditional in-person consultations were largely inaccessible (Australian Institute of Health and Welfare, 2022). This accelerated the uptake of digital tools, which has reduced dependency on face-to-face care (Morrison and Innes, 2022). Concurrently, practitioners are also participating in these trends. Limbic AI has become a popular practitioner tool for automating a range of administrative and technical tasks in the UK, including assessment, diagnoses, interventions, and risk identification (Limbic, 2024a). Overall, these trends signify evident shifts in technological perceptions, preferences, and practices in psychology.

Some perspectives suggest that additional socio-ecological factors–such as increased demand, limited access, demographic vulnerabilities, and cost considerations–are driving the adoption of AI technology. Unmet demand appears to be one of the most widely cited concerns (Neall et al., 2022; Reupert et al., 2018). The Australian Healthcare Index (Australian Patients Association Healthengine, 2023) reports that two-thirds of clients wait over two months for care, with Gen Z and millennials often forgoing treatment due to cost, despite their higher susceptibility to mental health concerns. Other issues include service delivery and suicide rates in rural areas (Hirsch and Cukrowicz, 2014), and clients’ day-to-day time pressures (Hagstrom and Maranzan, 2019). The widespread consensus is that technology is increasingly being used to overcome these issues by providing low cost, readily available tools that can be accessed anywhere (Boucher et al., 2021; Emily, 2016; Schueller et al., 2019).

Shifting to emerging perspectives from educators, recent literature highlights growing concern about preparing psychologists for an AI future. Australian Psychology Learning and Teaching (2023), a consortium of Australian universities, discussed the repercussions of technology’s impact. Key themes included: how innovation is significantly outpacing the agility of educational methods; that innovation is also outpacing the speed of research; the difficultly in evaluating ethical implications of fast changing AI tools; the need to integrate technological competency within academic programs; and the need for a collective shift in sentiment–from AI resistance to acceptance.

Hagstrom and Maranzan (2019) mirrors the themes from AusPLAT (2023), suggesting that the delivery of psychological services is changing faster than accreditation frameworks and institutions. The paper advocates for AI awareness in students and professionals as a foundational step. Additionally, it advocates for educating psychologists on assessments of AI tool reliability and validity, as well as addressing legal and ethical considerations. It also stresses the need for ‘explainability’–the capacity for psychologists to explain decisions made by AI, which contributes to making AI tools trustworthy and transparent. The general perspective of Hagstrom and Maranzan (2019) is that the surge of AI is inevitable, and that educators and psychologists must prepare.

Finally, Morrison and Innes (2022) articulated their view on an approaching crossroads for the profession. For instance, they observe that human empathy is frequently cited as crucial for psychologists and perceived as beyond the reach of AI. However, they note that trends toward systematising the profession indicate that the perceived importance of empathy in future psychologists is overstated. This raises questions about which skills and attributes are prioritised by psychology’s socially accepted paradigms, and whether these emphases are aiding or hindering the WR of current and future practitioners.

This study aimed to address the noted gaps in the literature and explore the nascent topic of psychology professional WR in the context of AI advancement. The goal was to uncover preliminary insights, providing a foundation for future research. Accordingly, this study investigated the research question: How do psychologists perceive work readiness within the context of AI advances?

Method

Research design

Given that this study aimed to explore WR through subjective experiences, a qualitative design was utilised to enable an in-depth exploration of psychologists’ WR perspectives and to collect rich, detailed data (Braun and Clarke, 2021). In adopting a social constructionist epistemology, WR perspectives are understood as active constructions shaped by language, social interactions, and their contexts, including technological, socio-ecological, and educational dimensions (Burr and Dick, 2017).

Aligning with this interpretive flexibility, the orientation to data was primarily experiential, focusing on psychologists’ subjective experiences and perceptions, while incorporating aspects of a critical orientation, acknowledging the influential role of contextual influences (Terry et al., 2017). To accommodate this flexibility, the constructionist stance was married with the theoretically-flexible technique of thematic analysis to implement a constructionist thematic analysis, as outlined by Braun and Clarke (2021).

Participants

Six fully-registered psychologists currently practicing in Australia formed the participant cohort. The inclusion of only Australia-based registered and practicing psychologists ensured participants possessed current experiential knowledge and relevant WR perspectives. These criteria ensured relevant, information-rich cases, thus offering an adequate data corpus to comprehensively address the research question (Vasileiou et al., 2018).

The sample size of six was considered ideal for generating a rich, diverse dataset across transcripts while maintaining practical feasibility for in-depth qualitative analysis (Terry et al., 2017). Sufficient information power was achieved through the specific sample of psychologists, which aligned with the research question, and was further supported by the interview format and open-ended questions and probes (Malterud et al., 2016).

Psychologist interactions occurred exclusively online, utilising Zoom for interviews, and email and LinkedIn direct messaging for recruitment. Purposive and homogeneous sampling exclusively targeted fully-registered Australian psychologists (Willig, 2013).

Participant ages ranged from 26 to 69 (M = 41.17, SD = 15.14), with a gender distribution of 5 women (83.33%) and 1 man (16.67%). The sample was mostly Caucasian/Australian (4 of 6; 66.67%), with 2 participants practising in urban settings (33.33%) and 4 in regional settings (66.67%). Additional demographic information was collected through the intake survey, along with participants’ use of technology. This can be found in Tables 1, 2. Data was collected weekdays from 9:00 a.m. to 5:00 p.m. via the University’s secure Zoom platform. As an incentive, each participant received a $30 gift card.

Table 1
www.frontiersin.org

Table 1. Participant demographics.

Table 2
www.frontiersin.org

Table 2. Participant demographics and technology use.

Data collection and procedure

The study received ethical approval from the University’s Human Research Ethics Committee (Project ID: 40833). Invitations were distributed confidentially through the researcher’s professional network via LinkedIn. Interested participants completed a digital Explanatory Statement, informed consent, and intake survey.

Semi-structured interviews were conducted via the University’s Zoom platform, which offered automatic and secure transcription. The semi-structured format, with its flexibility and adaptability, cohered with the study’s exploratory nature, its social constructionist epistemology and its experiential orientation (Terry et al., 2017). Open-ended questions, prompts and probes enabled participants’ voices and experiences to come forward (Byrne, 2022; Karatsareas, 2022).

The interview comprised six main questions, with follow-up prompts to elicit further information. The six questions were: (1) How would you define work readiness for psychologists? (2) What attributes or skills are most important for work readiness? (3) Do you think AI is influencing what it means to be work ready? (4) How do you imagine AI might affect psychological practice in the near future? (5) Are current training programs preparing psychologists adequately? (6) What changes might support future work readiness? These were crafted in line with key themes identified in the introduction and literature. Interviews lasted for 27–52 min.

The questions were crafted in line with themes reviewed in the introduction. For example, one question explored whether their training adequately prepared them for current/future practice (Hagstrom and Maranzan, 2019). Interviews lasted for 27–52 min. The schedule was designed to ensure participant comfort and ethical adherence, with steps including: pre-interview warm-up, review of informed consent, main interview, and a post-interview debrief (Turner and Hagstrom-Schmidt, 2022).

Data analysis

The analysis used an inductive, data-led approach, which cohered with the exploratory nature of the study. Analysis employed a social constructionist lens, with themes created using reflexive thematic analysis (RTA). Themes were built in a bottom-up fashion from the transcripts. Quality analysis was ensured through deep, reflexive engagement by the researcher, following the six-step thematic analysis process outlined by Braun and Clarke (2021).

Transcript information was open-coded using NVivo software, which aligned with the inductive approach and reflected meanings as conveyed by participants (Braun and Clarke, 2013). Coding occurred at two levels, both semantic and latent. For example, the phrase ‘It’s the human element that cannot be replaced’ was initially coded as “irreplaceable human skill,” which was later grouped under the sub-theme “centrality of relational work.” These and other similar codes formed the basis of a thematic map, which clarified sub-themes, broader themes, and their conceptual relationships.

Three overarching themes were inductively and recursively developed: Human-Centric Work Readiness Skills Are Irreplaceable by AI, AI Will Enhance Psychological Practice, and Education Must Adapt to Emphasise AI-Integration and Human-Centric Qualities. Themes were refined to ensure internal coherence, theoretical clarity, and clear distinctions between them (Braun and Clarke, 2020).

Utilising a social constructionist lens involved a critical stance, or ‘hermeneutics of suspicion’ (Smith et al., 2012), to delve beyond surface data and interrogate the assumptions and meaning systems underlying participants’ expressed understandings. Given the researcher’s interest in humanistic approaches to psychology and AI ethics, reflexive journaling and supervision were used to remain critically aware of how these values might shape theme construction and interpretation (Braun and Clarke, 2021). Transparency regarding researcher positionality was essential to honour the trustworthiness of interpretations.

Saturation was not the analytic goal, consistent with the reflexive TA approach, which instead prioritises depth and meaning-making over frequency or completeness (Braun and Clarke, 2021). The iterative and recursive nature of theme development aimed to produce a rich, nuanced understanding of the dataset.

Results

The data were analysed using Reflexive Thematic Analysis, following Braun and Clarke (2021) six-step process, and through a social constructionist lens. This analysis revealed three overarching themes that reflect psychologists’ perceptions of WR in the context of AI advancements:

1. Human-Centric WR Skill and Attributes are Irreplaceable By AI;

2. AI Will Enhance Psychological Practice Through Innovative Solutions;

3. Education Must Adapt to Emphasise Both AI-Integration and Human-Centric Qualities.

A visual representation of themes can be seen in Figure 1. Despite differing ages, levels of professional expertise, and therapeutic orientations, there was strong homogeneity in the data, reflecting consensus in the construction of meaning within the three themes. Beyond these core themes, participants highlighted additional key facets of WR requiring further focus in education to close the gap between training and professional practice. These are time management, legal and Medicare knowledge, developing openness and resilience, and the ability to tolerate uncertainty and manage one’s anxiety. Participants were highly engaged and enthusiastic throughout the interviews, offering rich responses. To maintain anonymity and provide insight into demographic information, they will be recognised by pseudonyms, age, positions, and therapeutic orientations.

Figure 1
Flowchart depicting how psychology professionals perceive WR in the context of AI advances. Three branches: 1. Human-centric WR skills are irreplaceable by AI. 2. AI enhances psychological practice with innovative solutions. 3. Education must integrate AI and human-centric qualities.

Figure 1. Visual representation of the three themes.

Theme 1: Human-centric WR skills and attributes are irreplaceable by AI

Human-centric skills and attributes were positioned as the most valuable and crucial aspects of WR by all psychologists. These dimensions encompass the interpersonal, intuitive, and reflective qualities essential for psychologists to build strong therapeutic relationships, attune to clients’ needs, and navigate practice complexities with empathy and flexibility. The centrality of these aspects was emphasised during two separate interview contexts: first, when asked to specify the most important aspects of WR; and again, when asked how WR may change as AI advances. This was interpreted as unanimous consensus that human-centric WR dimensions will not be fundamentally changed by AI and will thus remain centrally important into the future.

The ability to develop therapeutic relationships was specified by most as the foundational human-centric WR skill. As was stated by Lydia (32, clinical psychologist registrar, CBT/ACT-focused):

‘Progress in therapy is related to the therapeutic relationship. So I think that’s something that we absolutely need to keep at the forefront of our practice. What are we bringing to that? How are we reflecting on the dynamics between ourselves and the clients? That’s what we can hold on to. And that’s what we need more training in.’

As is clear in this language, therapeutic relationships were seen as paramount, a prerequisite for therapeutic success, and something we need to preserve as AI advances. This reflects a pattern of explicit and implicit meaning across the transcripts. Differing from the above explicit statement, some implied the importance of the therapeutic relationship by focusing on the subtler WR attributes that underpin it. Tahlia, (32, clinical psychologist registrar, Jungian/psychodynamic-focused), emphasised embodying kindness, consistency, and safety for clients. Eva, (44, psychology lecturer, psychologist, supervisor, member of an ‘AI applications in psychology’ interest group) stressed, with passionate vocal emphasis, ‘staying human’ for clients, despite the often ‘dehumanising’ nature of education and practice. Tahlia, Eva and Hallie, (69, psychologist, integrative approach) all described the capacity for human attunement–getting beyond the purely cognitive and connecting authentically with clients, facilitating the exploration of deep, emotional and often-unconscious material. These perspectives were interpreted as a passionate, commonly held belief in the importance of the therapeutic relationship and a deep respect for its subtle complexities–a pattern of meaning supported by the psychologists’ varying ages, professions, and therapeutic orientations.

Another crucial and irreplaceable human-centric skill was seen as one’s reflective capacity. As stated by Lydia: ‘My number one thing seems to be…reflective capacity, it informs everything we do…if you do not have the reflective capacity there, you are sort of done before you start.’ Here reflective practice was framed as a keystone skill for professional WR, which reflects a prominent pattern of explicit and implicit meaning expressed by psychologists. To share an implicit example, Adrian (44, clinical psychologist, PhD, practice director) stressed that developing self-knowledge and learning how to ‘manage your anxieties’ affected multiple areas of practice, which can be interpreted as a keystone framing of a personal dimension that implicates self-reflection. There was no suggestion that the importance of reflective capacity would change as AI advances, indicating a shared belief among psychologists that self-reflection will remain an enduring and essential aspect of WR.

Deepening into the suspicious critical approach, all of the aforementioned views were interpreted as connecting to another clear pattern, made evident by omission. That is, when asked what is centrally-important for WR, now and as AI advances, there was little emphasis on professional technical skills. This suggests that these skills are seen as peripheral–an important aspect of WR, but not of the same central importance as the human-centric skills and attributes.

One negative case is worth noting, however: Adrian diverged by placing a high degree of importance on aligning with an intervention framework that resonates with you personally as a psychologist, while making little mention of content connected to relational dynamics. While Adrian’s areas of focus in this regard were not representative of broad consensus, highlighting this negative case strengthens the analysis by acknowledging the diversity in participant experiences, thus providing a more nuanced representation of the data. Of note, Adrian was the only male in the study and had the highest level of education.

Adrian, however, did converge with broad consensus in his perspective regarding the need for flexibility and intuition. He emphatically stressed that manualised, inflexible approaches to the use of intervention frameworks can be detrimental:

‘[What’s important is] orienting to a process-oriented approach. Because I think there’s only so many diagnoses you can kind of work with. And you look at the data. A lot of the CBT interventions for specific diagnosis, they are based on very pure samples that do not really exist in real life. And if you were to take those approaches, as soon as it gets a bit more complicated, it starts to go off track’.

Adrian advocated for an intuitive approach that transcends the limitations of frameworks, allowing for a more nuanced exploration of clients, their unique circumstances, and contexts. Adrian’s perspective in this regard represents a marked pattern of explicit meaning across the transcripts. For example, Lydia’s stance was ‘very anti the rigid models,’ as was Carlotta’s (26, psychologist, CBT/counselling-focused). Hallie stated:

‘We have to tolerate sitting in the not knowing. And out of that not knowing, we ask. And then we receive. But if you are already full of your own answers, if your ego is so insecure that you have to fill it with all the knowledge that you can have, how can you?’

This commonly held view that rigidity can inhibit practice was interpreted and coded into the human-centric WR attributes of ‘intuition and flexibility’.

These–along with all the aforementioned skills and attributes–were thus constructed and interpreted as inherently human, crucial for WR, and unable to be replaced by AI as it advances. However, AI was seen as capable of integrating into other areas of psychological practice and supporting various WR dimensions.

Theme 2: AI will enhance psychological practice through innovative solutions

While the central place of human-centric skills and attributes was seen as unlikely to change, AI was perceived as capable of integrating with and enhancing other professional tasks and skills. Participant commentary regarding current and potential AI uses made this collective perspective explicitly clear. For example, Carlotta, Tahlia and Lydia are using ChatGPT to help write letters. Lydia suggested AI for conducting many assessments. Adrian imagined AI-automated interventions. Tahlia, advocating a more comprehensive approach, envisioned AI taking over as many administrative and technical tasks as possible. In analysing these comments individually, collectively, and within the broader context of participants’ overall expressions, a common motif was interpreted: it is the rote skills that are perceived as prime candidates for AI integration. Moreover, as illustrated by the examples above, it was again implied–this time by omission–that no one viewed AI as suitable for integration with the human, relational dimensions of WR.

Within this common proposition of AI for rote skills, there were, however, diverging views regarding where specifically AI should and should not integrate. In this regard, psychologists implicitly constructed a common categorical delineation between two feasible areas: administrative WR skills and technical WR skills. There was little disagreement that AI should support administration. As Eva put it: ‘hopefully [AI will] be utilised for a lot of the tasks that none of us appreciates much [like] recording your placement minutes…things that take time for no apparent reason.’ This captures the commonly held view which framed administrative aspects as unnecessarily laborious and ripe for AI disruption.

Beyond administration, while practitioners also envisioned AI integration with technical skills, there were nuanced perspectives nested within these constructions. Lydia for instance stated: ‘AI for cognitive assessments and diagnostic assessments, like where you are just working through diagnostic criteria or psychometrics.’ Similarly, Carlotta suggested AI could help with reports, where it feels like ‘you are reinventing the wheel’. These quotations reflect the dominant view that many technical aspects of the job demand inordinate time and complex skill sets, and that strategic AI use could create a more efficient reallocation of practitioner resources. However, concerns were also raised in this regard. Eva stated: ‘[It’s] a more ethical question…when we move into practice. Whose fault is it, if…there’s an accident [in which a client harms themself or another]?’ Similarly, Carlotta raised concerns with privacy, managing risk, and navigating human complexity. These views cannot be interpreted as an aversion toward AI because the same two practitioners envisioned its use for other technical (Carlotta) and administrative skills (Carlotta and Eva). Rather, the concerns are more appropriately interpreted as considerations that need to be addressed prior to AI integrating with technical WR skills.

Two additional and noteworthy interpretations merit inclusion here. First, this view that AI can integrate with technical skills implies that these skills are less central to the human’s role, thereby aligning with the earlier perspective that positions technical skills as peripheral. Second, to further clarify the narrative, psychologists did not indicate that technical skills can be simply taken over by AI. Rather, they suggest that AI should be used by humans as a supportive tool, which would still require the practitioner’s higher-order interpretative, intuitive, and integrative capacities to effectively leverage its potential.

These views, collectively interpreted with a suspicious stance, and alongside development of Theme 1, implied that by supporting administrative and technical tasks, AI could allow psychologists to focus on the core WR aspects they deemed central. Tahlia alone explicitly articulated this point:

‘It would be a great opportunity to use AI for that [technical and administrative] stuff and for us to really spend more time in our training around our own embodiment, our own knowing ourselves, and becoming better relators, more attuned people.’

Tahlia’s insight captures the full narrative constructed from the explicit and implicit meanings within this theme. By way of contrast, Hallie articulated that AI threatens the humanness of the profession, and that these tools could make the human’s role redundant. But this standalone view was not representative of broad consensus. Tahlia adds further nuance here: ‘If we are just focusing on techniques…I’d be concerned’. Her construction is again reflective of the broader meaning within this theme, which, alongside Theme 1, positions AI as better suited to the rote and the technical, and humans better suited to the relational, intuitive and interpretive.

These nuanced perspectives highlight the need for a dynamic, evolving understanding of WR which strikes a careful balance between gradual AI integration along with the preservation of human-centric dimensions. As the industry moves forward, the next critical step identified by psychologists was to ensure that education keeps pace, seamlessly integrating AI while solidifying the essential human-centric aspects of WR.

Theme 3: Education must adapt to emphasise both AI-integration and human-centric qualities

Psychologists put forward an argument that education must adapt and maintain pace with technological advances, ensuring future practitioners are equipped with the relevant skills and attributes for an AI-integrated landscape. Two clear motifs were nested within this theme: that education must become more AI-integrated and that it must also become more human-centric. While these views represent two often-polarised positions in the wider industry discourse, here they were interpreted as complementary perspectives.

Regarding the first motif, psychologists stressed that innovative technologies and AI must be integrated into psychology education. In expressing this, some reflected back on their education, like Lydia, who highlighted that a lack of technological literacy in her training created shortfalls in her readiness for work. Other psychologists looked ahead. Both Eva and Adrian emphasised the importance of AI literacy as crucial WR knowledge for future psychologists. Eva argued that students who can integrate AI into their practice will have significant advantages, and that educators ‘have a responsibility’ to prepare students for an AI-integrated profession. Adrian highlighted that ‘education needs to be about [AI] limitations’ and ethical considerations to ensure practitioners can use these tools effectively and responsibly. Implicit throughout these views and the wider commentary was a positivity and openness regarding AI integration. Moreover, these views were not passive; they were active and informed stances, stressing the critical need for responsible educational reforms to keep pace with technological progress. The weight of these constructions was supported by the varied and current experiences of the psychologists across educational and professional contexts.

The second key view put forward by psychologists was that education in the age of AI must become more human-centric. They believe a gap currently exists between what is taught and what is required in this regard, and that AI integration is unlikely to change that. Education, they believe, must become more practical and focused on developing relational, interpersonal abilities. For instance, Lydia remarked ‘It’s insane that we…do 4 years of an undergrad psychology degree and never once be in the room with another human being…to practise some of these core skills.’ Here Lydia shared a common view that experiential learning–working with, and learning how to attune to, real people–was one of the key drivers of WR, and required greater emphasis in the accreditation pathway.

Complementing this view was the common critique that psychology education overly emphasises theory, research, and manualised interventions. According to Tahlia, these aspects, which were central in her training–particularly the undergraduate level–have little use in her current role. Eva echoed this framing, suggesting that overemphasis of these dimensions also occurs at the postgraduate level. Its heavy focus on technical aspects over the ability to authentically connect with clients, she suggests, removes the ‘humanness’ from psychologists. Conversely, some practitioners did reflect positively on aspects of their post-graduate education, attributing this to having supervisors with psychodynamic (Lydia) or holistic (Carlotta) orientations. This was however an uncommon experience and minority view. Moreover, both Lydia and Carlotta equally criticised their postgraduate training’s focus on the manualised, technical aspects of CBT over the human-centric components of therapy they believe to be vital for WR, echoing Eva’s construction. Interpreting these views in the context of the larger patterns of meaning made clear that participants see a critical need for an educational balance that integrates technological proficiency while deeply valuing, preserving and embedding the human-centric WR skills and attributes.

The dual focus on AI-integration and human-centric dimensions in education was not seen as contradictory by the researcher, but as complementary and mutually reinforcing. Such meaning was drawn by interpreting this theme suspiciously and alongside Theme 2, in which psychologists did not position AI integration as making the profession more mechanistic or inhuman. Rather, they positioned AI as offering time-saving tools, strategies and insights, which would allow psychologists to increase their focus on the human-centric dimensions of practice. Following this reasoning, their emphasis on both AI integration and human dimensions in education makes sense, as it reflects their ‘division of labour’ vision (AI for the rote/technical, humans for the relational/intuitive/interpretive) and how to prepare practitioners for it.

Discussion

The findings highlighted three overarching themes covering the central significance of human-centric skills and attributes, potential areas for AI to enhance professional practice, and the need for education to adapt accordingly. These insights align with existing literature emphasising the enduring importance of human-centric skills and attributes (Flückiger et al., 2018; Norcross, 2005; Robinson et al., 2019; Salter and Rhodes, 2018; Stamoulos et al., 2016; Wampold, 2015) and extend these understandings by illustrating how AI is perceived to impact these skills, along with technical competencies. Furthermore, the findings underscore the critical need for educational programs to evolve to prepare future psychologists for an AI-enhanced profession.

A profession with an identity crisis

The practitioners in this study constructed psychology as a profession experiencing an identity crisis, which is negatively impacting the WR of professionals. As highlighted by the first theme, psychologists perceived the most crucial WR dimensions to be human-centric skills and attributes, a perception that will remain even as AI integration advances. This theme offers a humanistic construction of a psychologist’s role, resonating with prior literature examining psychologists’ perspectives (Robinson et al., 2019; Salter and Rhodes, 2018; Stamoulos et al., 2016). This view also aligns with meta-analyses that emphasise the centrality of the therapeutic relationship in client outcomes (Flückiger et al., 2018; Wampold, 2015). However, this perspective is at odds with the view embedded in the profession’s governing bodies and accreditation frameworks, which will be touched upon shortly. Nonetheless, this study concurs with and extends existing literature by conceptualising psychologists’ humanistic perspectives of their roles within the evolving concept of psychology professional WR, and against the backdrop of AI integration.

Within this humanistic understanding of their roles, psychologists viewed reflective practice, flexibility, intuition, and the capacity to form therapeutic relationships as essential skills and attributes. These findings echo and build upon key literature examining psychologists’ perspectives of what their professional role entails (see Robinson et al., 2019; Salter and Rhodes, 2018; Stamoulos et al., 2016). Salter and Rhodes (2018) and Robinson et al. (2019) both emphasised that humanistic dimensions such as personal development and reflective capacities are central to practise. This study supports and extends that view, further highlighting reflective practice as a keystone skill which fosters many other therapeutic capacities. A key distinction is that while previous studies focus on development processes, this study, guided by the WR framework (Caballero et al., 2011), focused on delineating discrete WR skills and attributes, such as relational abilities, reflective capacities, and intuition. While these areas are complementary–as the skills/attributes support the professional development process and vice-versa–the current study is the first to delineate specific human-centric dimensions comprising professional WR. Collectively, these studies underscore the importance of human-centric WR dimensions, while indicating that these dimensions involve important skills, attributes and processes for both current and future practice.

The perspectives of psychologists in the current study, however, diverge significantly from those embedded in the accreditation frameworks governing education and practice. As noted in the introduction, human-centric skills, attributes and developmental processes receive little to no focus in The PsyBa’s and AHPRA’s Professional Competencies for Psychologists (2023), nor in APAC’s Accreditation Standards for Psychology Programs (Australian Psychology Accreditation Council, 2019). Yet, these WR dimensions were the most highly valued by this study’s psychologists, as well as psychologists in the aforementioned prior studies, indicating a consistent and widespread view held by practitioners. Thus, regulatory bodies and practising professionals appear to hold two opposing perspectives regarding the role of a psychologist. Accordingly, psychologists also reported a gap between what they were taught, and what is now required of them in professional settings–which echoes other psychologists’ views (Robinson et al., 2019; Salter and Rhodes, 2018).

These findings, situated within key historical developments, can be interpreted as a manifestation of an ongoing schism within the field. Duncan and Reese (2012) offer useful context in this regard. As they explain, the push by psychology’s governing bodies toward evidence-based treatments, influenced by the medical model, aimed to align psychological practice with scientific standards. While this shift enhanced the profession’s credibility, it also sparked practitioner backlash regarding the overemphasis on prescriptive, manualised treatments. Hence the tension between empirical validation and the individualised, relational nature of practice. This discord can also be framed as mechanistic versus humanistic constructions of psychological practice, as explored in the introduction, and this appears to be negatively affecting WR. Interpreted within this historical context, psychologists see WR as requiring a paradigm shift from the mechanistic to the humanistic–a need becoming increasingly pronounced as AI reshapes the profession.

AI will transform psychology, allowing psychologists to be more human

According to the second theme, psychologists believe AI will transform and enhance professional practice, particularly by supporting administrative and technical domains. There is a strong preference for AI assistance with rote tasks in these areas, provided that their privacy and safety concerns (noted in the results section) are adequately addressed. Thus, while Theme 1 represents the important human-centric WR dimensions that psychologists believe will remain unchanged in the context of AI, Theme 2 encompasses aspects they anticipate evolving as AI integration progresses.

The view that AI will impact rote WR skills aligns with the technological perspectives reviewed in the introduction. As noted therein, Innes et al. (2022) view many technical skills as suitable for automation, suggesting that AI is better suited to perform these tasks. Psychologists in the current study concurred. However, their views differed significantly regarding the anticipated implications for professional WR. Invoking a view of psychological practice that aligns directly with the mechanistic paradigm embedded in governing frameworks, Innes et al. (2022) assert that AI poses a threat to the profession. They argue that AI will render psychologists redundant by outperforming them in rote technical tasks, which they see as central to the professional role. Conversely, constructing a more humanistic view of their jobs, psychologists in this study perceived technical tasks as important but peripheral WR skills. Consequently, they appear less threatened by, and more optimistic toward, AI. Psychologists view AI as a tool that will empower them, allowing a reallocation of time and effort toward the centrally important humanistic WR dimensions, provided their concerns regarding privacy, safety and risk management are addressed.

Situating these practitioner perspectives within their wider technological context supports their optimistic, humanistic views of AI integration, while also providing insight into how their concerns are being mitigated. For instance, the popularity of two AI psychology tools can be examined. Firstly, Limbic AI, covered in the introduction, has experienced rapid practitioner and client adoption, creates 40.8% increase in client recovery (Limbic, 2024a) and according to news reports, may soon launch in Australia (9 News Australia, 2024). Concerns regarding privacy and security have been effectively addressed through stringent protocols, and through backing by the National Health Service (Limbic, 2024b). Secondly, Heidi, an Australian-made AI tool, has recently risen to prominence in over 50 countries (Heidi Health, 2024a). It records sessions and automates patient data documentation, management plan creation, patient communication, referrals, and other processes. As with Limbic AI, Heidi includes robust privacy and security protocols (Heidi Health, 2024b). Both tools are designed in collaboration with psychologists and positioned as ways to alleviate pain points, including administrative/technical burdens, burnout, and the disruptive nature of note-taking (Heidi Health, 2024a; Limbic, 2024a). Reflecting the trajectory of technology trends, these tools offer a narrative that aligns closely with the views of the current study’s psychologists–that AI will change WR by supporting the technical/administrative tasks, thereby helping practitioners focus on their unique, human-centric contributions.

The considered design and development of Limbic AI and Heidi, along with a growing array of AI tools (Jain et al., 2024), offer a counter-narrative to the deterministic perspective on AI technology discussed in the introduction (Innes et al., 2022). Developed through collaborative efforts involving public health services and practitioners, these innovations are specifically targeting client and psychologist needs. They are addressing the reviewed socio-ecological pressures such as access, demographic vulnerability, cost, wait times, and time constraints (Jain et al., 2024; Neall et al., 2022), thereby exemplifying emerging, targeted solutions. It thus appears that technology is not independently shaping the profession according to its inherent capabilities, as per the deterministic view. Rather, these examples align more closely with the ‘Social Shaping of Technology’ framework, which views technological development as inherently influenced by social and economic factors (Țicău and Hadad, 2021). While not reviewed in the introduction, this framework offers an accurate representation of the psychologists’ views toward AI in this study, as well as the trajectory of AI innovations and socio-ecological trends. As such, psychologists’ views–situated within their emerging technological and socio-ecological context–suggest that the nature of administrative and technical WR skills is evolving, with increasing AI augmentation likely to occur.

Psychology education in the age of AI: a dual path

Psychologists envisioned a new approach to education, reflecting a division of labour in psychological practice between the rote tasks and the higher-order relational, intuitive and interpretive aspects of psychology–with AI assisting with the former and humans managing the latter. As touched on in Theme 1, this was seen as requiring a paradigm shift in how the profession, educators, and governing bodies understand psychological practice. Continuing to overly mechanise, manualise and dehumanise psychologists was seen as problematic, as this will draw humans into direct competition with emerging AI tools, while concurrently diminishing what is centrally important, and uniquely and irreplaceably offered by humans. These findings are novel and not comparable to prior perspectives in the literature. Of note, psychologists’ division of labour view did not involve AI subordinating the human’s role in technical competencies. Rather, AI was seen as empowering practitioners WR capabilities in these technical areas, saving time and effort, while still requiring their higher-order intuitive, interpretive and reasoning capacities.

To prepare students for this vision of future practice, psychologists believed psychology education must make two key changes, the first of which is to become AI-integrative. This aligns closely with the views of the educators covered in the introduction. AusPLAT (2023) and Hagstrom and Maranzan (2019) both called for the integration of technological competencies into education, which was echoed in this study. A common sentiment from Australian Psychology Learning and Teaching (2023), Hagstrom and Maranzan (2019) and the current study was the need for AI literacy, which is collectively seen as awareness of AI tools and capabilities, understanding their limitations, assessing risks, critically evaluating each tool before use, and upholding ethical and legal standards. Moreover, Hagstrom and Maranzan (2019) and the current study agree that AI-literate students and practitioners will have significant advantages. Collectively, this discourse conveys a prudently optimistic view of AI technologies in education and practice. The current study underscores the necessity of preparing students to embrace AI critically, ethically, and conscientiously, while offering insights into anticipated skill sets for a future-oriented approach to WR.

The second key change proposed by psychologists was for increased emphasis on human-centric WR dimensions within educational frameworks. Notably, this study is the first to advocate for this humanistic shift in education specifically in the context of AI. This interpretation emerged from two patterns of meaning. First, as indicated in the results, psychologists believed that humanistic dimensions receive less educational focus compared to technical dimensions. They view this imbalance as a hindrance to professional WR, necessitating adjustment–a perspective that aligns with prior literature (Robinson et al., 2019; Salter and Rhodes, 2018). Contrastingly, psychologists in this study extended these views by asserting that technical skills are ultimately peripheral to the central importance of human-centric dimensions, closely mirroring psychologists’ views in Stamoulos et al. (2016). Second, AI was anticipated to increasingly take over the manual aspects of technical competencies without altering the central importance of human-centric WR dimensions. Consequently, in the context of AI, psychologists foresee future practice requiring less manual effort for technical tasks, thereby freeing up time for the vital humanistic aspects, and call for educational adjustments accordingly.

To further clarify the psychologists’ views on AI integration, technical competencies and psychology education, it is essential to highlight two points. First, the psychologists were not criticising the use of frameworks, or particular frameworks, such as CBT. Rather, they criticised the heavy and manualised focus on the technical aspects within these frameworks. As explained by Farber (2014), therapeutic relationships, interpersonal skills, and reflective practice are fundamental aspects of significance across each of the five dominant orientations to therapy. The psychologists in this study echoed this view, calling for an increased focus on these humanistic elements common across each orientation–a view strengthened by the diversity of psychologists’ own therapeutic orientations. Second, a distinction must be drawn between the use of AI in education in general, and the targeted use of AI in professional psychology settings and education. As opposed to Lodge et al. (2023), which focuses on the transformative nature of AI in general educational contexts (e.g., assessments, learning, teaching), the psychologists in this study focused on specific AI uses to augment professional skills and client needs. This distinction is crucial when discussing AI integration in psychology education, as it delineates between enhancing professional WR in the context of AI versus evolving the approach to education in general.

Critical synthesis

The overall narrative the researcher developed across the themes was that psychologists see the heart of WR to be humanistic dimensions, supported by technical competencies. Their views are guided by direct experience practising with clients. Psychology’s governing bodies, however, hold a different view, placing higher value on the technical, mechanistic understanding of practice. Their view appears motivated by an attempt to legitimise and standardise the profession through an evidence-based model. Two key implications must be noted here. First, as initially explored by Innes et al. (2022), the accepted social paradigm of a discipline matters immensely. Extending that notion to the present study, it appears that the current favouring of a mechanistic over a humanistic representation of psychology is influencing education, practice and how the industry prepares for the future. Second, in the context of AI, this mechanistic paradigm appears problematic as it overlooks current technological and socio-ecological trends. Consequently, it draws psychologists into direct competition with emerging AI tools while diminishing their irreplaceable and centrally important humanistic contributions. These diverging mechanistic and humanistic perspectives each align with two distinct approaches to professional psychologist preparation. On the one hand, the competency-based accreditation frameworks–which have been developed largely in accordance with the mechanistic, evidence-based perspective–views static and measurable skills as the indicators of psychologist preparedness (Gonsalvez et al., 2021). On the other hand, WR aligns with the humanistic perspective, considering professional preparation to be a holistic endeavour that, beyond skills, can also include attributes, developmental processes, and personal characteristics (Caballero et al., 2011).

This study contributes a preliminary theoretical framework–the division of labour–which proposes that psychologists retain responsibility for relational, intuitive, and interpretive tasks, while AI supports administrative and technical processes. From this framework, several testable propositions emerge: for example, it is proposed that psychologists with greater AI literacy may feel more work-ready, particularly in more heavily AI-augmented settings; that AI integration may free up time for deeper relational work; and that outcomes may improve when human and machine input are intentionally and ethically divided. These propositions offer pathways for future hypothesis-driven or mixed-methods studies to assess the framework’s utility in practice.

The proposed division of labour framework aligns conceptually with several broader theoretical models. First, it builds on the Social Shaping of Technology perspective, previously introduced, by illustrating how psychologists actively shape AI adoption to preserve and elevate their humanistic functions (Țicău and Hadad, 2021). Second, the model draws from Person–Environment Fit Theory, which posits that optimal performance and satisfaction occur when individuals’ competencies align with evolving workplace demands (Edwards, 1991). It also echoes Role Theory, which helps explain how psychologists construct, adapt, and negotiate their professional identity amid shifting responsibilities introduced by AI integration (Biddle, 1986). Collectively, these theoretical lenses help situate the division of labour model as a socially informed, context-sensitive framework for understanding and enhancing WR in the age of AI.

Future research is encouraged to empirically test these propositions across diverse contexts and populations. Mixed-methods or longitudinal studies could explore whether psychologists trained in AI literacy report greater confidence and work readiness in digitally supported environments. Experimental studies might investigate whether AI use, such as automated note-taking or triage, improves client outcomes or reduces practitioner burnout. Qualitative research could also examine the perspectives of psychologists with more manualised or tech-forward orientations to broaden the representativeness of future findings.

Strengths and limitations

The current study is the first to extend the emerging line of WR research into the field of psychology professionals and the context of AI. One of the study’s primary strengths was its inclusion of a diverse range of psychologists, encompassing variation in ages, positions, locations, therapeutic orientations, experience levels, educational backgrounds, and historical contexts of training. This diversity facilitated a comprehensive understanding of WR, while supporting the homogeneity of the findings, suggesting that views were shaped by common experiences of practice, rather than individual therapist differences.

The study’s qualitative methodology is another significant strength. It allowed for in-depth exploration of psychologists’ personal views and experiences, offering a rich and nuanced understanding of the topic (Braun and Clarke, 2021). Within this methodology, the use of reflexive thematic analysis provided a systematic and theoretically flexible method for analysing psychologists’ perspectives, while allowing the researcher’s subjectivity to be used as an interpretive tool (Braun and Clarke, 2013). This approach was particularly valuable for navigating this nascent research topic, while grounding its findings in relevant perspectives and real-world experiences.

Another strength was the social constructionist epistemology, which was vital for interpreting and understanding psychologists’ perspectives based on their language, experiences and contexts (Burr and Dick, 2017). The result is a well-rounded understanding of perspectives, which was cognisant of wider environmental conditions. Importantly, this epistemology–in conjunction with reflexive thematic analysis–facilitated an adequately detailed exploration of the topic, thereby supporting the study’s quality and trustworthiness (Lewin et al., 2018).

The current study also has several transferability limitations that should be considered when interpreting the findings. Firstly, while there was a spread of therapeutic orientations, the practitioners aligned philosophically in their humanistic stance. This may reflect a broader trend; for instance, prior Canadian research found psychologists leaned away from manualised methods like CBT and more toward humanistic, psychodynamic approaches (Stamoulos et al., 2016). However, it may also reflect a recruitment or self-selection bias. The researcher acknowledges that this orientation may not be representative of all Australian psychologists, particularly those aligned with more cognitive, behavioural, or tech-forward approaches. While the study sought to include diverse perspectives, future research should purposively sample psychologists from a wider range of philosophical and practice orientations to explore whether perceptions of WR differ accordingly.

Additionally, the study may be subject to self-selection bias, as participants were volunteers. This means that those with particularly strong opinions or interests may have been more likely to take part (Robinson, 2014). Thus, the analysis and interpretations may be biased and not be fully representative of all Australian psychologists. Another limitation relates to sample demographics. While the sample provided rich experiential insights, it lacked gender diversity, with the majority of participants identifying as female. This limits the generalisability of the findings, and future studies should seek a more balanced demographic representation. The pace of technological advancement presents another limitation, as the speed of AI innovation creates difficulties in conducting AI-related research that remains relevant (Dwivedi et al., 2021). As such, the perspectives, insights and tools featured here may become quickly outdated, necessitating continuous scholarly review to maintain the applicability and relevance of the current findings.

Future directions

Given the findings and limitations, three key directions for future research are recommended. First, expanding the sample to focus on or include psychologists with a positivist philosophical stance could provide additional insights into WR in the context of AI (Teo and Teo, 2018). Additionally, researchers could investigate whether most Australian practitioners, like those in the current study and Stamoulos et al. (2016), philosophically align in valuing humanistic over positivist approaches. These steps could help determine the broader applicability of these findings and guide development of suitable accreditation frameworks. Second, to overcome the analytic challenges posed by self-selection bias, future research could employ differing recruitment strategies. For example, random sampling from various professional psychology organisations may help overcome this bias, thereby supporting the findings’ generalisability (Rueda et al., 2022). Third, longitudinal studies could examine how perceptions of WR and AI evolve over time (Malau-Aduli et al., 2022). As AI technology continues to advance, it is essential to track these changes and understand their long-term implications for psychological practice. Such studies could provide ongoing insights into how educational and training programs can continue to adapt to support WR.

Implications

These findings are preliminary and offer initial insights into WR for psychologists, laying the groundwork for further studies, ultimately working toward a psychology-specific, Australia-contextualised, future-oriented conceptualisation of WR. However, three preliminary implications are apparent and may inspire further research and discourse. First, this study echoes calls for a re-evaluation of the mechanistic social representation of psychology (Norcross, 2005). Further discussion exploring a shift toward a more humanistic representation is warranted, potentially informing the evolution of accreditation frameworks, and aligning the understanding of psychology professional WR between psychologists and industry bodies (Duncan and Reese, 2012). As AI integration progresses, the urgency of this shift becomes increasingly apparent.

Second, as found in prior literature (Robinson et al., 2019; Salter and Rhodes, 2018), this study highlights the limitations of the current accreditation frameworks in adequately preparing psychologists for practice. Conceptually, the framework lacks the flexibility to incorporate human-centric attributes and developmental processes (Gonsalvez et al., 2021) which psychologists have consistently perceived as crucial throughout the literature (Robinson et al., 2019; Salter and Rhodes, 2018). Consideration must be given to whether supplementation with, or a move toward, a more holistic WR framework is necessary.

Third, education must adapt to address the shortfall in WR reported by psychologists in this and previous studies (Robinson et al., 2019; Salter and Rhodes, 2018). Primarily, it is crucial to address the overemphasis on theoretical and technical skills at the expense of practical and humanistic capacities. Both psychologists and the reviewed technological advancements suggest that AI will reduce the labour required for technical tasks such as assessments, freeing up time for higher-order abilities that typically result from these tasks, as per the Social Shaping of Technology framework (Țicău and Hadad, 2021). Educational programs must monitor the evolution of AI tools and adapt accordingly. The role of humans in technical competencies may need reimagining as AI integration develops. For example, AI literacy should be more clearly defined and embedded in training. AI literacy could be operationalised as the capacity to critically evaluate AI tools based on evidence of validity, ethical safeguards, transparency, and bias detection (Hagstrom and Maranzan, 2019). Practical training could include case studies using evidenced and emerging tools, with students asked to assess their limitations, data handling, and impact on therapeutic rapport.

Finally, this study echoes Hagstrom and Maranzan (2019) by underscoring the importance of integrating AI literacy and ethical considerations into training programs. As AI tools become more prevalent, WR will involve practitioners’ ability to use these tools effectively and responsibly.

Conclusion

In conclusion, this study illuminates the intricate dynamics of WR among psychologists in the context of AI. The findings underscore the paramount importance of human-centric skills–such as reflective practice, intuition, and therapeutic relationship-building–which remain central despite increasing AI integration. AI is anticipated to transform administrative and technical tasks, thereby enabling psychologists to focus more deeply on these core human dimensions.

Furthermore, the study highlights a significant divergence between practitioners’ views and those of governing bodies, suggesting an urgent need for accreditation frameworks to evolve. These frameworks must balance the integration of AI with a greater emphasis on developing humanistic capabilities to adequately prepare future psychologists.

To help reconcile these divergent perspectives, this study proposes a preliminary theoretical framework–the division of labour model–which positions AI as a complementary tool supporting technical tasks, while psychologists retain responsibility for interpretive and relational work. This model offers a practical and theory-informed vision for hybrid psychological practice. Finally, the study lays a foundation for future hypothesis-driven and mixed-methods research to empirically assess how AI literacy, ethical division of tasks, and evolving workplace demands shape professional WR in psychology.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by the Monash University Human Research Ethics Committee. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

CM: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review & editing. AS: Methodology, Project administration, Supervision, Writing – review & editing.

Funding

The author(s) declare that no financial support was received for the research and/or publication of this article.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The authors declare that no Gen AI was used in the creation of this manuscript.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

9 News Australia. (2024). AI Psychologist Chatbot Could Be Part of Australia’s Official Health System. Available online at: https://www.youtube.com/watch?v=x37nJhKn6b8 (Accessed April 2, 2024).

Google Scholar

Abrams, Z. (2021). The promise and challenges of AI. APA Monit. Psychol. 52. Available at: https://www.apa.org/monitor/2021/11/cover-artificial-intelligence

Google Scholar

Australian Institute of Health and Welfare (2022). Mental health: prevalence and impact. Available online at: https://www.aihw.gov.au/reports/mental-health/mental-health#references

Google Scholar

Australian Patients Association Healthengine. (2023). Australian healthcare index. Available online at: https://australianhealthcareindex.com.au/

Google Scholar

Australian Psychology Accreditation Council. (2019). Accreditation standards for psychology programs (v1.2). Available online at: https://apac.au/wp-content/uploads/2021/03/APAC-Accreditation-Standards_v1.2.pdf

Google Scholar

Australian Psychology Learning and Teaching. (2023). First Roundtable for 2023. Available online at: https://www.ausplat.com/2023/03/12/new-post/ (Accessed March 12, 2024).

Google Scholar

Bickman, L. (2020). Improving mental health services: a 50-year journey from randomized experiments to artificial intelligence and precision mental health. Adm. Policy Ment. Health Ment. Health Serv. Res. 47, 795–843. doi: 10.1007/s10488-020-01065-8

PubMed Abstract | Crossref Full Text | Google Scholar

Biddle, B. J. (1986). Recent Developments in Role Theory. Annu. Rev. Sociol. 12, 67–92. doi: 10.1146/annurev.so.12.080186.000435

Crossref Full Text | Google Scholar

Boucher, E. M., Harake, N. R., Ward, H. E., Stoeckl, S. E., Vargas, J., Minkel, J., et al. (2021). Artificially intelligent chatbots in digital mental health interventions: a review. Expert Rev. Med. Devices 18, 37–49. doi: 10.1080/17434440.2021.2013200

Crossref Full Text | Google Scholar

Braun, V., and Clarke, V. (2013). Successful qualitative research: A practical guide for beginners. London: Sage Publications.

Google Scholar

Braun, V., and Clarke, V. (2020). One size fits all? What counts as quality practice in (reflexive) thematic analysis? Qual. Res. Psychol. 18, 328–352. doi: 10.1080/14780887.2020.1769238

Crossref Full Text | Google Scholar

Braun, V., and Clarke, V. (2021). Thematic analysis: A practical guide. London: Sage Publications.

Google Scholar

Burr, V., and Dick, P. (2017). “Social constructionism” in The Palgrave handbook of critical social psychology. ed. B. Gough (London: Palgrave Macmillan), 59–80.

Google Scholar

Byrne, D. (2022). A worked example of Braun and Clarke’s approach to reflexive thematic analysis. Qual. Quant. 56, 1391–1412. doi: 10.1007/s11135-021-01182-y

Crossref Full Text | Google Scholar

Caballero, C. L., Walker, A., and Fuller-Tyszkiewicz, M. (2011). The work readiness scale (WRS): developing a measure to assess work readiness in college graduates. J. Teach. Learn. Grad. Employ. 2, 41–54. doi: 10.21153/jtlge2011vol2no1art552

Crossref Full Text | Google Scholar

Casner-Lotto, J, and Barrington, L (2006). Are they really ready to work? Employers’ perspectives on the basic knowledge and applied skills of new entrants to the 21st-century US workforce. Partnership for 21st Century Skills. Available online at: https://eric.ed.gov/?id=ED519465.

Google Scholar

Coetzee, M. (2014). Measuring student graduateness: reliability and construct validity of the graduate skills and attributes scale. High. Educ. Res. Dev. 33, 887–902. doi: 10.1080/07294360.2014.890572

Crossref Full Text | Google Scholar

Duncan, B. L., and Reese, R. J. (2012). Empirically supported treatments, evidence-based treatments, and evidence-based practice in J. C. Norcross (Ed.), Handbook of psychology, vol. 8. 2nd ed. Wiley. pp. 497–520.

Google Scholar

Dwivedi, Y. K., Hughes, D. L., Ismagilova, E., Aarts, G., Coombs, C., Crick, T., et al. (2021). Artificial intelligence (AI): multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice, and policy. Int. J. Inf. Manag. 57:101994. doi: 10.1016/j.ijinfomgt.2019.08.002

Crossref Full Text | Google Scholar

Edwards, J. R. (1991). Person-job fit: A conceptual integration, literature review, and methodological critique. John Wiley \u0026amp; Sons.

Google Scholar

Emily, A. (2016). Mental health: there’s an app for that. Nature 532:20. doi: 10.1038/532020a

Crossref Full Text | Google Scholar

Farber, E. W. (2014). “Theoretical orientation in the education and training of psychologists” in The Oxford handbook of education and training in professional psychology. eds. N. J. Kaslow and W. B. Johnson (New York: Oxford University Press), 67–86.

Google Scholar

Flückiger, C., Del Re, A. C., Wampold, B. E., and Horvath, A. O. (2018). The alliance in adult psychotherapy: a meta-analytic synthesis. Psychotherapy 55, 316–340. doi: 10.1037/pst0000172

PubMed Abstract | Crossref Full Text | Google Scholar

Frey, C. B., and Osborne, M. A. (2017). The future of employment: how susceptible are jobs to computerisation?. Technol. Forecast. Soc. Change. 114, 254–280. doi: 10.1016/j.techfore.2016.08.019

Crossref Full Text | Google Scholar

Gonsalvez, C. J., Shafranske, E. P., McLeod, H. J., and Falender, C. A. (2021). Competency-based standards and guidelines for psychology practice in Australia: opportunities and risks. Clin. Psychol. 25, 244–259. doi: 10.1080/13284207.2020.1829943

Crossref Full Text | Google Scholar

Graham, S., Depp, C., Lee, E. E., Nebeker, C., Tu, X., Kim, H. C., et al. (2019). Artificial intelligence for mental health and mental illnesses: an overview. Curr. Psychiatry Rep. 21, 1–18. doi: 10.1007/s11920-019-1094-0

Crossref Full Text | Google Scholar

Hagstrom, S. L., and Maranzan, K. A. (2019). Bridging the gap between technological advance and professional psychology training: a way forward. Can. Psychol. Psychol. Can. 60, 281–289. doi: 10.1037/cap0000186

Crossref Full Text | Google Scholar

Heidi Health. (2024a). Heidi health. Available online at: https://www.heidihealth.com/au (Accessed June 2, 2024).

Google Scholar

Heidi Health. (2024b). Privacy policy. Available online at: https://www.heidihealth.com/au/legal/privacy-policy (Accessed June 2, 2024).

Google Scholar

Hirsch, J. K., and Cukrowicz, K. C. (2014). Suicide in rural areas: an updated review of the literature. J. Rural Ment. Health 38, 65–78. doi: 10.1037/rmh0000018

Crossref Full Text | Google Scholar

Innes, C. J., Innes, J. M., and Morrison, B. W. (2022). Social representation of the profession of psychology and the application of artificial intelligence: European Union regulatory authority and the application of psychology as a paradigm for the future. Aust. N. Z. J. Eur. Stud. 14. doi: 10.30722/anzjes.vol14.iss1.15852

Crossref Full Text | Google Scholar

Innes, J. M., and Morrison, B. W. (2021). “Artificial intelligence and psychology” in The Routledge social science handbook of AI (London: Routledge), 30–57.

Google Scholar

Jain, A., Srivastava, G., Singh, S., and Dubey, V. (2024). “Application of artificial intelligence (AI) technologies in employing chatbots to access mental health” in A. Khang, V. Abdullayev, O. Hrybiuk, (eds.). Computer vision and AI-integrated IoT Technologies in the Medical Ecosystem (Boca Raton, FL, USA: CRC Press), 311–333.

Google Scholar

Karatsareas, P. (2022). “Semi-structured interviews” in Research methods in language attitudes. eds. R. Kircher and L. Zipp (Cambridge, United Kingdom: Cambridge University Press), 99–113.

Google Scholar

Kilanowski, J. F. (2017). Breadth of the socio-ecological model. J. Agromedicine 22, 295–297. doi: 10.1080/1059924X.2017.1358971

PubMed Abstract | Crossref Full Text | Google Scholar

Lewin, S., Booth, A., Glenton, C., Munthe-Kaas, H., Rashidian, A., Wainwright, M., et al. (2018). Applying GRADE-CERQual to qualitative evidence synthesis findings: introduction to the series. Implement. Sci. 13, 1–10. doi: 10.1186/s13012-017-0688-3

Crossref Full Text | Google Scholar

Limbic, AI. (2024a). Limbic AI. Available online at: https://limbic.ai/ (Accessed May 15, 2024).

Google Scholar

Limbic, AI. (2024b). Care privacy policy. Available online at: https://policies.limbic.ai/care-pp#:~:text=We%20ensure%20that%20your%20personal,be%20used%20to%2C%20identify%20you (Accessed June 2, 2024).

Google Scholar

Lodge, J. M., Yang, S., Furze, L., and Dawson, P. (2023). It’s not like a calculator, so what is the relationship between learners and generative artificial intelligence? Learn.: Res. Pract. 9, 117–124. doi: 10.1080/23735082.2023.2261106

PubMed Abstract | Crossref Full Text | Google Scholar

Luxton, D. D. (2014). Artificial intelligence in psychological practice: current and future applications and implications. Prof. Psychol. Res. Pract. 45:332. doi: 10.1037/a0034559

Crossref Full Text | Google Scholar

Malau-Aduli, B. S., Jones, K., Alele, F., Adu, M. D., Drovandi, A., Knott, G., et al. (2022). Readiness to enter the workforce: perceptions of health professions students at a regional Australian university. BMC Med. Educ. 22:89. doi: 10.1186/s12909-022-03120-4

PubMed Abstract | Crossref Full Text | Google Scholar

Malterud, K., Siersma, V. D., and Guassora, A. D. (2016). Sample size in qualitative interview studies: guided by information power. Qual. Health Res. 26, 1753–1760. doi: 10.1177/1049732315617444

PubMed Abstract | Crossref Full Text | Google Scholar

Morrison, B. W., and Innes, J. M. (2022). Judgement day for psychologists? How technology has arrived to save the day. World Finan. Rev. Available at: https://worldfinancialreview.com/judgement-day-for-psychologists-how-technology-has-arrived-to-save-the-day/

Google Scholar

Neall, A. M., Cooney, O., and Oppert, M. L. (2022). Enhancing employment outcomes for psychology graduates: developing a taxonomy of work-integrated learning. Int. J. Work-Integr. Learn. 23, 509–525. Available at: https://eric.ed.gov/?id=EJ1373505

Google Scholar

Norcross, J. C. (2005). The psychotherapist’s own psychotherapy: educating and developing psychologists. Am. Psychol. 60, 840–850. doi: 10.1037/0003-066X.60.8.840

PubMed Abstract | Crossref Full Text | Google Scholar

Prikshat, V., Kumar, S., and Nankervis, A. (2019). Work-readiness integrated competence model: conceptualisation and scale development. Educ. Train. 61, 568–589. doi: 10.1108/ET-05-2018-0114

Crossref Full Text | Google Scholar

Psychology Board of Australia and Australian Health Practitioner Regulation Agency. (2023). Professional competencies for psychologists. Available online at: https://www.ahpra.gov.au/documents/default.aspx?record=WD23%2F32537&dbid=AP&chksum=nrm1%2FMn4vf6%2Bfx7poWRgew%3D%3D.

Google Scholar

Reupert, A., Davis, M., Stewart, S., and Bridgman, H. (2018). A new education pathway for postgraduate psychology students: challenges and opportunities. Aust. J. Adult Learn. 58, 225–245. doi: 10.3316/ielapa.862363933341813

Crossref Full Text | Google Scholar

Robinson, N. L., Schweitzer, R. D., and O’Connor, E. L. (2019). Early reflections on becoming a therapist: development of reflective practice in clinical training programmes in an Australian context. Couns. Psychother. Res. 19, 388–398. doi: 10.1002/capr.12255

Crossref Full Text | Google Scholar

Robinson, O. C. (2014). Sampling in interview-based qualitative research: a theoretical and practical guide. Qual. Res. Psychol. 11, 25–41. doi: 10.1080/14780887.2013.801543

Crossref Full Text | Google Scholar

Rueda, M. D. M., Martínez-Puertas, S., and Castro-Martín, L. (2022). Methods to counter self-selection bias in estimations of the distribution function and quantiles. Mathematics 10:4726. doi: 10.3390/math10244726

Crossref Full Text | Google Scholar

Salter, M., and Rhodes, P. (2018). On becoming a therapist: a narrative inquiry of personal–professional development and the training of clinical psychologists. Aust. Psychol. 53, 486–492. doi: 10.1111/ap.12344

Crossref Full Text | Google Scholar

Schueller, S. M., Hunter, J. F., Figueroa, C., and Aguilera, A. (2019). Use of digital mental health for marginalized and underserved populations. Curr. Treat. Options Psychiatry 6, 243–255. doi: 10.1007/s40501-019-00181-z

Crossref Full Text | Google Scholar

Smith, J. A., Flowers, P., and Larkin, M. (2012). Interpretative Phenomenological Analysis: Theory, Method and Research. SAGE Publications.

Google Scholar

Stamoulos, C., Trepanier, L., Bourkas, S., Bradley, S., Stelmaszczyk, K., Schwartzman, D., et al. (2016). Psychologists’ perceptions of the importance of common factors in psychotherapy for successful treatment outcomes. J. Psychother. Integr. 26:300. doi: 10.1037/a0040426

Crossref Full Text | Google Scholar

Stevens, B., Hyde, J., Knight, R., Shires, A., and Alexander, R. (2017). Competency-based training and assessment in Australian postgraduate clinical psychology education. Clin. Psychol. 21, 174–185. doi: 10.1111/cp.12061

Crossref Full Text | Google Scholar

Teo, T., and Teo, T. (2018). The consequences of positivism in psychology. Outline of theoretical psychology: Critical investigations, 103–132. doi: 10.1057/978-1-137-59651-2_5

Crossref Full Text | Google Scholar

Terry, G., Hayfield, N., Braun, V., and Clarke, V. (2017). “Thematic analysis” in The SAGE handbook of qualitative research in psychology. eds. C. Willig and W. S. Rogers (Newbury Park, CA: Sage Publications), 17–37.

Google Scholar

Țicău, I. R., and Hadad, S. (2021). Technological determinism vs. social shaping of technology: the influence of activity trackers on users’ attitudes. Manag. Dyn. Knowl. Econ. 9, 147–163. doi: 10.2478/mdke-2021-0011

Crossref Full Text | Google Scholar

Turner, D. W. III, and Hagstrom-Schmidt, N. (2022). Qualitative interview design: a practical guide for novice investigators. Qual. Rep. 15, 754–760. doi: 10.46743/2160-3715/2010.1178

PubMed Abstract | Crossref Full Text | Google Scholar

Vasileiou, K., Barnett, J., Thorpe, S., and Young, T. (2018). Characterising and justifying sample size sufficiency in interview-based studies: systematic analysis of qualitative health research over a 15-year period. BMC Med. Res. Methodol. 18:148. doi: 10.1186/s12874-018-0594-7

PubMed Abstract | Crossref Full Text | Google Scholar

Wampold, B. E. (2015). How important are the common factors in psychotherapy? An update. World Psychiatry 14, 270–277. doi: 10.1002/wps.20238

PubMed Abstract | Crossref Full Text | Google Scholar

Willig, C. (2013). Introducing qualitative research in psychology: Adventures in theory and method. 3rd Edn. Maidenhead, UK: Open University Press, McGraw-Hill Education.

Google Scholar

World Economic Forum (2019). Strategies for the new economy: skills as the currency of the labour market. Available online at: http://www3.weforum.org/docs/WEF_2019_Strategies_for_the_New_Economy_Skills.pdf

Google Scholar

Keywords: work readiness, AI, artificial intelligence, psychology, psychologist, skills, attributes, Australia

Citation: McDonald C and Schweinsberg A (2025) Ready or not? Psychologists’ perceptions of work readiness in the age of AI. Front. Comput. Sci. 7:1524024. doi: 10.3389/fcomp.2025.1524024

Received: 10 April 2025; Accepted: 04 July 2025;
Published: 18 July 2025.

Edited by:

Athanasios Drigas, National Centre of Scientific Research Demokritos, Greece

Reviewed by:

Netty Merdiaty, Universitas Bhayangkara Jakarta Raya, Indonesia
Li Ding, Institut Lyfe (EX-Institut Paul Bocuse), France
Aikaterini Doulou, National Centre of Scientific Research Demokritos, Greece

Copyright © 2025 McDonald and Schweinsberg. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Curtis McDonald, Y3VydC5tY2RvbmFsZDdAZ21haWwuY29t

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.