How experiential learning and AI readiness foster career adaptation in hospitality and tourism education

Amid profound digital transformation, the rapid advancement of artificial intelligence (AI) is reshaping employment structures in the hospitality and tourism (H&T) industry. Technologies such as chatbots, facial recognition, concierge robots, intelligent Customer Relationship Management (CRM) systems, and virtual or augmented reality (VR/AR) are increasingly integrated into management and service delivery (Huang and Rust, 2018; Samala et al., 2020). From a managerial standpoint, AI enhances customer experience, optimizes operational efficiency, and fosters workplace flexibility. However, from a workforce perspective, it creates new challenges related to skill requirements and career adaptability. As automated systems replace routine tasks, employees must reskill in digital competencies, strengthen their technological interaction abilities, and adopt a lifelong learning approach to sustain competitiveness (Tussyadiah, 2020).

From a labor market perspective, the H&T sector faces two concurrent challenges: a large share of graduates working in jobs unrelated to their field, and a persistent shortage of workers equipped with both professional and technological competencies as the industry transitions toward smart tourism. In Vietnam, a national survey by the MOTIVE Research Group (2022) of over 4,000 graduates from the class of 2020 found that only 41.4% were employed “fully within their major,” 42.2% “partly aligned,” and 16.4% “not aligned,” underscoring the gap between higher education and labor market demands. Globally, the OECD (2017) estimated that approximately one-fourth of the workforce across member countries experiences skill mismatches, which reduces productivity and organizational performance. Similarly, the European Labour Authority (2024) reported that hospitality occupations—particularly chefs and kitchen staff—are among the most understaffed, with shortages present in three-quarters of the surveyed countries. Complementing these findings, Minor et al. (2025) identified widening digital skill gaps in areas such as online platform management, digital marketing, and the application of emerging technologies (AI, AR/VR) as key barriers to advancing global smart tourism. In Vietnam, experts from RMIT University Vietnam (2025) likewise warned that deficiencies in vocational and technological skills hinder the post-COVID-19 recovery and long-term growth of the H&T sector. Consequently, attracting and retaining a highly skilled workforce, including well-trained graduates, is essential for sustaining competitiveness. Within this context, understanding the psychological mechanisms shaping H&T students’ career intentions in the AI era holds crucial implications for both academia and industry.

From a theoretical standpoint, job intention (JI) reflects an individual’s degree of commitment and orientation toward career choice, persistence, and development (Klassen et al., 2023; Lent et al., 1994). A synthesis of prior studies on the career intentions of H&T students reveals that, although research on this topic is expanding, it remains fragmented in its theoretical grounding and insufficiently addresses the mechanisms of career adaptability amid technological change. Existing studies can be grouped into three main streams: (1) behavioral–cognitive approaches, (2) experiential learning (EL) and adaptability perspectives, and (3) value- and emotion-based approaches. Yet, an in-depth review reveals significant theoretical gaps.

First, behavioral–cognitive research, often based on the theory of planned behavior, has effectively predicted career intentions but offers a limited understanding of adaptability in dynamic environments. For instance, Ngatuni and Ulomi (2021) and Park et al. (2017) confirmed that attitudes and social norms influence career intentions; however, their models emphasize static cognition and overlook how individuals regulate psychological resources to cope with change. Similarly, Gyepi-Garbrah et al. (2023), drawing on goal-setting and expectancy theories, identified self-efficacy and goal commitment as key predictors; however, they did not explain the adaptive mechanisms that sustain motivation. Moreover, Atef and Al-Balushi (2017) reported that 41% of Omani students viewed H&T as a temporary career, suggesting weak career attachment, without clarifying the personal factors shaping this persistence. These limitations underscore the need for a more integrative framework that explains “who adapts better” and “why differences exist,” particularly as AI-driven transformation continues to reshape occupational demands in the H&T sector.

Second, research on EL and employability consistently highlights the importance of practical experience in strengthening students’ career intentions, yet few studies conceptualize such experiences as adaptivity. Giousmpasoglou and Marinakou (2021) found that satisfaction with hotel internships positively affects career intention, while Kong (2021) and Yang et al. (2016) demonstrated that EL enhances professional competence and the application of theoretical knowledge. However, these studies largely viewed EL as a pedagogical method rather than a personal attribute reflecting proactive career engagement. In the era of digital transformation, Stylianou and Pericleous (2025) introduced digital readiness, including AI readiness (AIR), as essential technological competencies for H&T students; however, their work focused solely on assessing preparedness rather than its impact on career intentions. Integrating EL involvement and AIR as dual antecedents offers a more holistic representation of adaptivity resources—EL involvement reflecting adaptability to traditional work contexts and AIR representing adaptability to digital environments. This remains an underexplored area, as no prior studies have jointly examined these constructs within a predictive framework of H&T students’ career intentions.

Third, studies grounded in self-value and affective perspectives have emphasized the psychological dimension of JI. Atitsogbe et al. (2019) and Pitan and Müller (2019) demonstrated that perceived employability (PE)—a domain-specific form of self-efficacy—significantly influences job choice. Given the emotional intensity of the H&T sector (Hochschild, 2012), job self-congruence (SC) has been identified as a key driver of intrinsic motivation and commitment (Kristof-Brown et al., 2005; Sheldon et al., 2020). While PE reflects an external orientation toward labor-market demands, SC aligns with personal values. Moreover, Biraglia and Kadile (2017) and Wang et al. (2021) validated Job passion (JP) as a pre-employment affective factor crucial in emotionally demanding fields like H&T. Although prior research acknowledges cognitive and affective influences on JI, a unified mechanism linking these dimensions with behavioral outcomes remains absent—an integration vital to understanding how cognition factors (e.g. PE, SC) and affect factors (e.g. JP) jointly shape career intentions.

Building upon the identified gaps, this study integrates two theoretical foundations: career construction theory (CCT) and the cognition–affect–conation (C–A–C) model. CCT (Savickas, 2002) explains who can adapt through four adaptive dimensions—adaptivity, adaptability, adapting response, and adaptation result—while the C–A–C model (Lavidge and Steiner, 1961) elucidates how adaptation unfolds through the sequential linkage of cognition, affect, and behavior. Integrating these frameworks transforms CCT’s adaptability process into a concrete psychological mechanism: EL involvement and AIR represent adaptivity, reflecting readiness for career development; PE and job SC signify adaptability, or cognitive evaluation of vocational fit; JP captures the adapting response, an affective driver of action; and JI denotes the adaptation result, the behavioral outcome. This integration not only extends CCT by clarifying both who adapts and how adaptation occurs but also offers a comprehensive explanation of how career intention is formed among H&T students in the AI era. The findings are expected to advance CCT and provide practical insights for attracting and retaining talent in the H&T sector.

CCT, proposed by Savickas (2002), explains career development through four hierarchical processes: adaptivity (career readiness), adaptability (adaptive capacities), adapting responses, and adaptation results. In H&T education, students actively construct their careers rather than merely acquire knowledge. Attributes such as EL involvement and AIR represent adaptivity, reflecting preparedness and proactive engagement with professional and technological demands. Building upon this foundation, adaptability—manifested through PE and job SC—reflects students’ confidence and socio-psychological fit within the profession. These cognitive capacities nurture JP as an adaptive response, ultimately leading to JI, the adaptation outcome of the career construction process.

Complementing this view, the C–A–C model (Lavidge and Steiner, 1961) conceptualizes decision-making as a sequential process moving from cognition to affect and then to conation. Within this model, PE and SC constitute the cognitive stage, JP represents the affective response, and JI corresponds to the conative stage. IEL and AIR serve as stimuli that initiate this sequence by shaping cognition. Hence, the adaptivity elements of CCT simultaneously operate as the entry point in the C–A–C process, activating the cognition–affect–conation chain.

Integrating CCT and C–A–C offers a comprehensive explanatory mechanism: while CCT identifies who adapts through adaptivity and adaptability, C–A–C clarifies how adaptation unfolds via sequential psychological processes. This theoretical alignment translates the abstract career-construction logic of CCT into a concrete sequence—from readiness (EL involvement and AIR) through cognition and affect to behavioral intention—providing an interdisciplinary lens for understanding H&T students’ career intentions in technologically evolving contexts.

EL is an educational approach that has garnered significant academic attention, emphasizing the integration of practical experience and critical reflection as central mechanisms for constructing knowledge and fostering personal development (Kolb and Kolb, 2005). Kolb’s (1984) EL model outlines four sequential stages: concrete experience, reflective observation, abstract conceptualization, and active experimentation, with reflection functioning as the pivotal component. Rooted in active learning and experiential education, EL prioritizes learner participation and reflective engagement (Arcodia et al., 2021).

In higher education, particularly within H&T, EL is implemented through diverse modalities. Yang et al. (2016) identify three major categories of EL activities: school-based, professional-skill-related, and practice-based. School-based EL activities include guest lectures, case studies, simulations, role-plays, and project-based tasks, all of which foster critical thinking and problem-solving skills (Lin et al., 2017; Yang et al., 2016). Skill-based EL encompasses certification courses, professional competitions, and community service, aiming to strengthen students’ technical competencies (Yang et al., 2016). Practice-based EL involves internships, cooperative education programs, and field trips that enhance adaptability, communication skills, and industry exposure (Yan and Cheung, 2012; Yang et al., 2016). Despite their varied formats, these EL approaches share a common goal: providing students with meaningful practical experiences that equip them with essential disciplinary knowledge and professional capabilities.

The concept of “involvement” refers to the level of engagement required to participate in an activity or service (Good, 1990). In education, it reflects the physical and psychological energy students invest in learning (Astin, 2014). This study adopts Good’s (1990) behavioral-oriented perspective, encompassing: (1) participation—active engagement in practical experiences like internships, field trips and simulations within tourism and hospitality; (2) persistence—sustained effort and resilience in addressing challenges during EL; and (3) attention—focus on details and experiences related to EL activities (Vo et al., 2025). In the H&T context, higher involvement in EL activities strengthens learning outcomes, professional competencies, and personal growth (Astin, 2014).

In the technological era, AI has emerged as a transformative innovation across global industries (Samala et al., 2020). In H&T, while AI adoption enhances service efficiency, its complexity may hinder acceptance, as employees often perceive it as challenging to understand and operate (Chung, 2014). AIR—an extension of technology readiness—offers more profound insight into these challenges by reflecting individuals’ preparedness to integrate AI into their work (Richey et al., 2007; Yu et al., 2023).

AIR refers to “the preparedness of an organization or individual to adopt and utilize AI technologies effectively” (Li et al., 2023, p 2), encompassing one’s inclination and capability to apply AI for professional and personal goals (Parasuraman, 2000). In this study, it is conceptualized as students’ preparedness to engage with AI-based operational and managerial systems, including knowledge, skills, ethical awareness, and vision for AI’s future role (Karaca et al., 2021; Wang et al., 2023). In H&T education, a variety of interventions can be used to enhance students’ AIR. These include the use of interactive AI systems (e.g. chatbots, ChatGPT, Gemini); AI-integrated project-based learning (e.g. Tableau with AI-enhanced forecasting modules, ChatGPT for scenario design, customer analytics tools); critical AI evaluation tasks (e.g. comparing AI-generated forecasts with actual hotel data, assessing the accuracy of AI responses to complex service scenarios); human–AI collaborative environments (e.g. AI-generated itinerary suggestions refined by students, AI-assisted booking processes subsequently reviewed and adjusted by learners); data-literacy tools (e.g. Excel with AI add-ins, PowerBI dashboards augmented by AI capabilities); and ethical and responsible AI training (e.g. evaluating the credibility of AI-generated information, discussing privacy concerns arising from AI-based tourist data collection) (Almatrafi et al., 2024; Long and Magerko, 2020; Tzirides et al., 2024). As a key determinant of adaptability and human–AI collaboration, AIR fundamentally shapes career choices and commitment in AI-driven workplaces (Bankins et al., 2024).

According to the C-A-C model, the job decision-making process begins with students’ cognitive appraisal of prospective job options. In this study, SC and PE are conceptualized as key cognitive processes involved in this appraisal. Introduced by Sirgy (1982), SC refers to the perceived alignment between an individual’s self-concept and the image of an external object. It comprises actual SC (alignment with current self-identity) and ideal SC (alignment with aspirational self-image). In H&T, SC explains why students choose careers that align with both their personal identity and desired future self (adapted from Tran et al., 2022).

In recent years, employability has attracted significant scholarly attention amid declining job security, increasing labor market competition, and rising educational demands (Zakkariya and Nimmi, 2021). PE refers to an individual’s perception of their likelihood of securing new, equivalent, or improved employment opportunities (Berntson, 2008; Zakkariya and Nimmi, 2021). It reflects general and professional competencies, work attitudes, and career-planning confidence, all of which are critical for career exploration and adaptability in the dynamic H&T labor market (Chen et al., 2021).

In the affective stage, students develop emotional responses toward potential job options. JP, which reflects students’ emotional attachment and enthusiasm toward a job, embodies these responses. Derived from Vallerand et al.’s (2003) model of passion, JP is conceptualized as a strong, enduring emotional inclination toward a job that students find meaningful and enjoyable (Zigarmi et al., 2009). In this study, JP is characterized by autonomous, voluntary engagement in meaningful work and is formed before starting a job (Biraglia and Kadile, 2017; Vallerand et al., 2003).

The conative stage represents the translation of cognitive and emotional evaluations into behavioral intentions. In the context of job decision-making, this stage reflects students’ willingness and determination to pursue a particular career path based on the cognitive and affective evaluations they have made. JI is a key indicator of students’ commitment to their career goals, influencing both their immediate job choices and long-term professional engagement (Klassen et al., 2023; Lent et al., 1994).

AIR reflects an individual’s preparedness to effectively apply AI in achieving professional objectives (Li et al., 2023). In contrast, PE refers to an individual’s belief in their ability to obtain and sustain employment that matches their qualifications (Rothwell et al., 2008). By fostering AI-related competence and vision, AIR enhances confidence and adaptability in AI-integrated workplaces, thereby strengthening the core components of PE (Fugate and Kinicki, 2008; Rothwell et al., 2008) and increasing willingness to adopt AI technologies (Wang et al., 2023). In H&T, AIR is considered essential for bridging the gap between graduates’ competencies and industry needs (Stylianou and Pericleous, 2025), serving as a key determinant of employability and career sustainability in the era of Technology 5.0 (Hussain et al., 2023). Thus, AIR is posited to enhance PE.

PE is closely linked to self-efficacy, a central determinant of confidence in achieving career goals and influencing job choices (Bandura, 1986; Lent et al., 1994). Klassen et al. (2023) found that self-efficacy significantly impacts job choice intentions, while Atitsogbe et al. (2019) observed that PE precedes self-efficacy among individuals with limited work experience. Empirical evidence further indicates that PE positively impacts adaptability and job search behaviors (Rothwell et al., 2008) and fosters proactive job-related decision-making (Onyishi et al., 2015), including a tendency toward wage employment (Neneh, 2020) or entrepreneurship (Atitsogbe et al., 2019). In other words, PE is assumed to influence JI.

Within CCT, AIR—as an advanced form of technology readiness—acts as an adaptivity resource, while job SC represents adaptability. Technology readiness enhances adaptability through concern, control, curiosity, and confidence (Kumi et al., 2024). Individuals with high technology readiness are more open to technological innovation, optimistic about its benefits, and confident in performing AI-based tasks, thus perceiving a stronger fit with technology-intensive roles in H&T. Moreover, technology readiness strengthens the personal meaning of work (Abdul Hamid, 2022), implying that AIR positively influences job SC.

Empirical studies further confirm the direct effect of SC on behavioral intention across diverse contexts—mobile apps (Wu et al., 2020), cosmetics (Khalid et al., 2018), online retail (Das and Khatwani, 2018), and tourism (Tran et al., 2022). Similarly, person–job fit and occupational congruence significantly shape JI (Kristof-Brown et al., 2005; Sheldon et al., 2020).

Based on the above arguments, this study proposes the following hypotheses:

EL plays a pivotal role in bridging the gap between theoretical knowledge and practical application, enhancing students’ responsibility and alignment between academic training and industry needs (Kong, 2021). IEL offers enactive mastery through role-playing (Huertas-Valdivia, 2021), field trips (Arcodia et al., 2021), internships (Yang et al., 2016), and simulations (McGrath et al., 2020). Greater engagement in these activities enhances confidence and competence in handling professional challenges (Bandura, 1986). IEL also involves vicarious learning, such as job shadowing (Yan and Cheung, 2012) and team coaching (Azanza et al., 2022), where observing others’ success reinforces self-belief (Bandura, 1986). Together, these experiences enhance PE—the belief in one’s ability to perform effectively in potential roles (Chen et al., 2021)—which is strongly associated with self-efficacy (Berntson, 2008). Participation in improvement-oriented groups similarly boosts confidence and readiness for broader job roles (Axtell and Parker, 2003).

Beyond skill acquisition (Pitan and Muller, 2019), EL cultivates critical reflection (Kolb and Kolb, 2005) and supports self-determined learning (Deci and Ryan, 2000). Active engagement in meaningful experiences fosters self-awareness and a clearer understanding of vocational preferences and capabilities (Hirschi and Freund, 2014). As Kristof-Brown et al. (2005) note, person–job fit evolves not only from individual traits but also through active participation in career-relevant contexts. Building on this foundation, the study posits that deeper engagement in EL helps H&T students better identify their values, interests, and competencies, enhancing perceived job fit and employability.

Building on these arguments and the previously discussed links among PE/SC and JI, this study proposes the following hypotheses:

Beyond cognitive evaluation, job choice decisions are also shaped by affective factors—particularly work passion. JP refers to a strong, enduring positive emotion individuals develop toward their aspired occupation (Zigarmi et al., 2009). According to the psychological transformation logic, clear perceptions often evoke emotional attachment (Tran et al., 2023). PE, defined as individuals’ self-assessment of their knowledge, skills, and attitudes relative to labor market demands, determines their confidence in obtaining suitable employment (Pitan and Muller, 2019; Rothwell et al., 2008). As students perceive higher employability, they proactively manage their careers and recognize their market value (Jackson and Wilton, 2017; Tomlinson, 2017). Drawing on expectancy-value theory (Wigfield and Eccles, 2000), PE represents expectancy beliefs that enhance the perceived value of desired jobs—expressed affectively through JP.

Person-job fit, which reflects the compatibility between individuals and job attributes (Lam et al., 2018), plays a central role in shaping job love, encompassing passion, intimacy, and commitment (Bibi et al., 2022). When employees perceive strong alignment between their jobs and personal attributes, they become more motivated and emotionally connected to their work (Bibi et al., 2022). Since SC reflects a deeper psychological dimension of person-job fit (Kristof-Brown et al., 2005), it is likely to foster JP.

Cardon et al. (2009) emphasize that passion for a specific field fuels motivation to pursue related goals. Similarly, Biraglia and Kadile (2017) and Wang et al. (2021) found that occupational or industry passion significantly drives career choice intentions. Supporting this, Bibi et al. (2022) confirmed that JP encourages individuals to pursue personally meaningful work consistent with their self-concept and long-term aspirations.

Building on these arguments and the previously discussed links among AIR/IEL and PE, AIR/IEL and SC, PE/SC and JI, this study proposes the following hypotheses:

The conceptual model is shown in Figure 1.

All procedures involving human participants complied with institutional and national research ethics standards and adhered to the principles of the 1964 Helsinki Declaration and its subsequent amendments. Respondents were informed of the study’s purpose and their right to withdraw at any time. Anonymity and confidentiality were strictly maintained throughout the data collection process.

Validated scales from prior research were employed to measure all constructs. IEL was assessed using four items from Vo et al. (2025), and AIR was assessed using four items from Karaca et al. (2021). Four dimensions of PE —GC, Professional competency (PC), Career planning and confidence (CPC), and WA—were measured using 8, 4, 3, and 3 items, respectively, adapted from Chen et al. (2021). Two dimensions of SC—ASC and ISC—were measured using three items each, as described by Tran et al. (2022). JP was measured using six items from Biraglia and Kadile (2017), and JI was measured using four items from Tang et al. (2020). All items used a five-point Likert scale (1 = strongly disagree, 5 = strongly agree). Three experts in tourism behavior reviewed the instrument for content validity and translation accuracy, and a pilot test with 30 students confirmed clarity before the final survey.

Given the study’s aim to capture students’ job-related intentions, an online survey targeting Vietnamese H&T students in their third or fourth year who had participated in EL activities was conducted as a data collection instrument, supported by purposive sampling, given the time and financial constraints that precluded the implementation of simple random sampling (Tang et al., 2020). In practice, undergraduate programs in Vietnam typically last three to four years. During their third and fourth years, students commonly participate in professional placements, complete graduation internships, and may even receive offers of full-time positions before graduating. Therefore, focusing on these two cohorts aligns well with the purpose of this study, as they represent the group most likely to be in the critical stage of forming concrete job intentions. Therefore, to ensure that only qualified respondents were included, three screening questions were administered, covering their major, academic year, and the EL activities they had participated in. Between April and June 2025, a total of 603 responses were collected, of which 521 were valid. Among these, 35.5% were third-year students and 64.5% were fourth-year students; 63.3% majored in hospitality and 36.7% majored in tourism; and 80.8% of the respondents were female. Anonymity and the absence of “right or wrong” answers helped minimize social desirability bias (Podsakoff et al., 2003).

Data were analyzed using partial least squares structural equation modeling with SmartPLS 4.0, which is suitable for testing complex models with multiple constructs and exploratory objectives (Hair et al., 2019; Rigdon, 2012). The analysis included assessments of the measurement model, common method bias (CMB), and structural relationships.

This study assessed both first-order and second-order measurement models. The first-order model included ten factors and 42 items, evaluated through indicator loadings, reliability, and validity tests. In the initial analysis, all item loadings exceeded the 0.708 threshold except for GC6 (“Native language proficiency,” 0.656), GC7 (“Foreign language proficiency,” 0.657), and GC8 (“Emotional control and tolerance of pressure,” 0.693). Due to insufficient discriminant validity between PC and general competency (GC) (HTMT >0.9), these three items were removed.

The revised model (Table 1) met all reliability and validity criteria, as evidenced by Cronbach’s Alpha (CA) and Composite Reliability (CR) values exceeding 0.7, and Average Variance Extracted (AVE) values surpassing 0.5, which confirmed internal consistency and convergent validity.

Discriminant validity, as tested using the Fornell–Larcker criterion and the HTMT ratio, was also acceptable (Table 2). Although the HTMT value between PC and GC (0.906) slightly exceeded the cutoff, the constructs were conceptually distinct, and other indices consistently supported discriminant validity (Henseler et al., 2016).

The second-order model evaluated the hierarchical structure of PE (GC, PC, WA, and CPC) and SC (actual self-congruence, ideal self-congruence). All first-order loadings exceeded 0.708, and CA, CR, and AVE values satisfied the recommended thresholds (>0.7 and > 0.5). These results confirm that PE and SC are best represented by their first-order components (Table 3).

To assess CMB, the Partial Correlation Procedure with a general factor variable (GEN-FACT) was applied (Tehseen et al., 2017). A single random-item variable (0–1) was introduced to test its effect on the dependent constructs. Bootstrapping with 5,000 resamples revealed negligible changes in R² values (PE and JP remained unchanged; SC increased by 0.003; JI increased by 0.002), suggesting that CMB was not a significant concern (Figures 2 and 3).

The structural model was evaluated for model fit, hypothesis testing, and explanatory power. The model achieved satisfactory fit indices (SRMR = 0.054 < 0.08; normed fit index = 0.844 > 0.8). Bootstrapping procedures (Zhao et al., 2010) were used to test the indirect effects of AIR and IEL on JI. Results (Table 4) partially supported the proposed mediation: SC significantly mediated the effects of AIR and IEL on JI (H1b, H2b), whereas PE did not (H1a, H2a). This may reflect that students perceive employability in H&T as more dependent on interpersonal, practical, and teamwork skills rather than advanced technological competence. Hence, students high in AIR may not feel more employable, but instead more aligned with AI-related roles, which in turn influences their job intentions.

JP exhibited a strong mediating role, with all four indirect paths (H3a, H3b, H4a, H4b) supported. This highlights JP as a core affective mechanism linking EL and AIR to JI—underscoring its function as a motivational force sustaining students’ career engagement in emotionally intensive H&T professions.

The model’s explanatory and predictive power were further confirmed. The R² value for JI was 0.657 (Figure 2), indicating substantial explanatory strength (Cohen, 1988). Predictive validity was assessed using PLSpredict/Cross-validated Predictive Ability Test (CVPAT), with Q² values for JI, JP, PE, and SC all positive, demonstrating predictive relevance (Hair et al., 2019). Moreover, the model’s mean prediction errors were lower than the indicator-average benchmark, and no meaningful loss relative to the linear model benchmark was observed, confirming strong predictive capability (Table 5).

The adoption of AI in customer-facing operations has become an irreversible trend in H&T. Given the real-time, service-oriented nature of these industries, graduates must not only perform effectively upon entering the workforce but also seamlessly integrate AI into their roles. Drawing on the integration of CCT and the C-A-C model, this study tested the framework on 521 Vietnamese H&T students. The findings reveal that both EL involvement (IEL) and AIR significantly shape students’ job choice intentions through two primary pathways: (1) a cognitive route, by strengthening perceptions of person–job fit (AIR/IEL → SC → JI); and (2) a cognition–emotion route, by jointly reinforcing employability beliefs and SC, together with JP (AIR/IEL → PE/SC → JP → JI). SC emerges as a key mediator linking AIR and IEL to JI, while JP is essential for sustaining these intentions. Overall, the study highlights the interaction between technological readiness, professional learning, and individual identity in shaping career intentions, offering both theoretical advancement and educational insights.

A clearer picture emerges when comparing this study’s findings with those of previous studies and examining the distinct roles of PE, SC, and JP in the cognitive-affective pathway. Although many previous studies have consistently shown that PE predicts career-related intentions (Atitsogbe et al., 2019; Neneh, 2020; Onyishi et al., 2015), the current study did not provide evidence supporting PE as an independent mediator between AIR/IEL and JI. This conclusion is based on the statistical results, which show that both indirect effects through PE were insignificant (H1a, H2a). In contrast, the results showed that PE only became influential when placed in the mediation chain involving JP (PE → JP → JI). Both mediation chains (H3a, H4a) were strongly supported. This implies that employability beliefs, while cognitively meaningful, must trigger an emotional response, namely JP, before they translate into career intentions. In the highly emotional H&T industry, self-efficacy in finding a job is not enough; it is the emotional attachment to the job that drives career choice intentions. This finding complements previous studies by showing that PE influences career choice indirectly and conditionally, rather than directly as traditionally assumed.

In contrast to PE, SC emerged as a strong and stable mediator of both AIR and IEL (H1b, H2b), confirming its proximate role in the formation of career intentions. This is consistent with the broader research on person-job fit and career identity (Kristof-Brown et al., 2005; Sheldon et al., 2020), and extends the understanding that SC serves as a core cognitive anchor in both traditional learning (IEL) and AI-enabled environments (AIR). For H&T students, perceiving their career as reflecting their true self and desired future self is more crucial than general perceptions of employability. This reinforces the identity-based argument that SC reflects deep psychological compatibility, which is particularly important in emotionally oriented service sectors. Across all mediational chain hypotheses, JP consistently transmits the effects of both cognitive factors (PE and SC) to JI (H3a–H4b), underscoring JP’s centrality as an emotional mechanism in the cognition → emotion → behavior chain. JP does not operate in isolation; instead, it is activated when students’ cognitive evaluations – either of employability or of self-fit – reach a level sufficient to create an emotional attachment to the profession. This multi-layered conduction mechanism of JP helps clarify the formation of career passion before entering the labor market and reinforces the results of previous studies that passion is an essential driver in H&T career decisions (Biraglia and Kadile, 2017; Bibi et al., 2022; Wang et al., 2021).

Overall, the findings suggest that PE plays a conditional, emotion-dependent role, SC plays a direct, identity-based role, and JP is the core emotion catalyst that sustains the transformation of cognitive appraisals into career-choice intentions.

This study advances the understanding of career choice intentions in H&T through four main theoretical contributions. First, it develops an integrated explanatory framework by linking CCT with the C–A–C model. While CCT conceptualizes levels of career adaptation, the C–A–C model specifies sequential decision-making mechanisms. By aligning CCT’s four layers—Adaptivity (AIR, IEL), Adaptability (PE, SC), Adapting Response (JP), and Adaptation Result (JI)—with the cognition → affect → conation process, this study extends Wang et al. (2024)‘s work by elucidating how Adaptivity transforms into JI through psychological mechanisms. Few studies have combined these frameworks; thus, this dual integration enriches both by clarifying who adapts (CCT) and how adaptation occurs (C–A–C).

Second, the study introduces AIR as a novel antecedent at the Adaptivity level. Building on technology readiness theory (Parasuraman, 2000), this work extends beyond previous studies that have merely assessed students’ AIR (Karaca et al., 2021; Stylianou and Pericleous, 2025; Wang et al., 2023), positioning AIR as a behavioral antecedent in career formation. By examining AIR alongside IEL, the study identifies two complementary sources of Adaptivity: technological readiness and professional practice orientation—offering a theoretical expansion beyond traditional competence-based models.

Third, the results highlight a distinctive mediating mechanism where SC surpasses PE in transmitting the effects of AIR and IEL on JI. Unlike prior studies that emphasize PE (Onyishi et al., 2015), this finding highlights the emotional demands of the service sector (Hochschild, 2012), in which the alignment between self-identity and job roles becomes more critical than confidence in employability. This distinction refines career adaptability theory by differentiating between market-oriented adaptability (PE) and identity-oriented adaptability (SC).

Finally, the study confirms JP as an essential adaptive response that shapes job intentions before career entry. Extending Vallerand et al. (2003) and Biraglia and Kadile (2017), JP is shown to act as an emotional bridge between cognition (PE, SC) and behavior (JI), reinforcing that intrinsic motivation—manifested as JP—emerges early and profoundly influences job choice in emotionally intensive fields like H&T.

The findings offer valuable practical implications for both H&T educational institutions and businesses. Beyond external interventions, they emphasize the importance of students understanding their internal psychological mechanisms—from cognitive appraisal (SC, PE) to emotional attachment (JP) to behavioral intention (JI). Such awareness fosters self-reflection, professional identity formation, and long-term commitment to the H&T industry.

Grounded in CCT and the C–A–C model, this study identifies four sequential levels in students’ career intention development: (1) Adaptivity (IEL, AIR), (2) Adaptability (career identity, PE), (3) Adapting Response (JP), and (4) Adaptation Result (JI). Each level plays a distinct role and requires coordinated support among educational institutions, businesses, and students, guided by reflective learning.

To operationalize these insights, Table 6 proposes targeted recommendations for each level, integrating mechanisms for monitoring and evaluating effectiveness. This framework provides a systematic roadmap for stakeholders to enhance students’ career intentions and sustain workforce development in the H&T sector.

This study has several limitations that should be considered in future research. Firstly, the sample primarily consisted of third- and fourth-year students in Vietnam, which may limit the generalizability of the findings. Future studies should expand the sample to include international students for broader representation. In addition, although screening questions were used to ensure that all respondents had participated in EL activities, differences in the amount or depth of EL exposure between third- and fourth-year students may still exist. Such variation could introduce unobserved heterogeneity, and future research may consider stratified sampling or multi-group analysis to better account for these differences.

Secondly, although the tested model demonstrated statistically significant relationships among the variables, the cross-sectional design limits the ability to infer causality. Future research may employ longitudinal or experimental designs to better capture the dynamic process of career intention formation. Finally, this study treated AIR as a unidimensional construct. Future research should explore how different dimensions of AIR may differentially influence the formation of JI among H&T students.

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent was obtained from all individual participants included in the research.