Intergenerational Tacit Knowledge Transfer: Leveraging AI

Abstract

The growing number of senior experts leaving the workforce (especially in more developed economies, such as in Europe), combined with the ubiquitous access to artificial intelligence (AI), is triggering organizations to review their knowledge transfer programs, motivated by both financial and management perspectives. Our study aims to contribute to the field by analyzing options to integrate intergenerational tacit knowledge transfer (InterGenTacitKT) with AI-driven approaches, offering a novel perspective on sustainable Knowledge and Human Resource Management in organizations. We will do this by building on previous research and by extracting findings from 36 in-depth semi-structured interviews that provided success factors for junior/senior tandems (JuSeTs) as one notable format of tacit knowledge transfer. We also refer to the literature, in a grounded theory iterative process, analyzing current findings on the use of AI in tacit knowledge transfer and triangulating and critically synthesizing these sources of data. We suggest that adding AI into a tandem situation can facilitate collaboration and thus aid in knowledge transfer and trust-building. We posit that AI can offer strong complementary services for InterGenTacitKT by fostering the identified success factors for JuSeTs (clarity of roles, complementary skill sets, matching personalities, and trust), thus offering organizations a powerful means to enhance the effectiveness and sustainability of InterGenTacitKT that also strengthens employee productivity, satisfaction, and loyalty and overall organizational competitiveness.

1. Introduction

Building on and adding to the growing number of studies that analyze the developments of artificial intelligence (AI) in business settings, we are exploring the opportunities of integrating AI in intergenerational tacit knowledge transfer (InterGenTacitKT) contexts. A research gap was identified here as knowledge transfer, sharing, and hiding have been focused on before, but not in the context of AI, and we specifically focus on junior and senior tandems (which we call JuSeTs).

For intergenerational knowledge transfer, mentoring, coaching, advising, and tutoring—in their varied interpretations—are established formats as they offer personalized learning and facilitate relationship-building. Collaboration, job/role shadowing, and storytelling have emerged as particularly effective methods for transferring tacit knowledge (for a detailed typology, please see [1]).

In a tandem format, the partners jointly occupy a position for a specified timeframe and share responsibilities and decision-making competencies.

Our study triangulates the research on InterGenTacitKT with the results from a set of qualitative interviews with the aim of exploring senior experts’ contribution to organizational success. As a foundation for our study, we present a definition for InterGenTacitKT and its importance, methods, and challenges. Drawing on our study’s findings, we present success factors for one specific format of intergenerational knowledge transfer: junior/senior tandems. These findings we then overlay with our analysis of available research on artificial intelligence (AI) in InterGenTacitKT situations.

AI nowadays is already an important conversational partner for both older and younger employees. It has the potential to actually bridge a gap between generations, as both older and younger employees have to relate to this new technology and evaluate how to integrate it into their daily tasks [2]. In our study, we followed Russel and Norvig’s (2022) definition of artificial intelligence (AI) as agents capable of learning from the environment and performing actions that reflect human-like intelligence while applying rationality based on processed data [3].

Specifically, we build on the fact that AI systems have become an inherent part of the white-collar jobs that our study focuses on [4,5,6]; with teams already experimenting how to integrate AI in a meaningful way into their operational responsibilities.

In this study, data was collected through 36 semi-structured interviews with participants selected from knowledge-intensive industries.

Repeated reading and evaluation of this material revealed a subset of data that was used for this article: namely the experience of some interviewees with intergenerational tandems in transferring their tacit knowledge to younger colleagues.

Tacit (or implicit) knowledge refers to a form of personal, experience-based knowledge of an intangible and unconscious nature that is difficult to articulate or formalize. It matures over time through repeated application, reflection, and social interaction, often embedded in routines and shared practices [7]. Nonaka and Takeuchi (2021) emphasize the dynamic interplay between tacit and explicit knowledge as a core mechanism of knowledge creation and its contribution to organizational resilience [8]. Resilience describes the capability of an organization to adapt effectively to challenging or disruptive conditions [9].

Collins (2010) distinguishes between relational, somatic, and collective tacit knowledge, adding the socio-cultural dimensions that make tacit knowledge challenging to transfer [10].

Decision-makers faced with a potentially critical amount of knowledge loss due to retiring senior experts are looking for innovative ways of capturing not just their explicit but also their tacit knowledge: from relationship and conflict management to their intuition regarding problems with complex engineering systems. Advances in technology and AI systems might be about to give the possibility to mitigate this challenge: facilitating new ways to codify tacit knowledge, which would then be available for future generations, independent of time and the number of knowledge seekers.

With digitalization and the growing use of AI, the interpersonal interaction between human experts decreases, and thus the opportunities for tacit knowledge transfer [11]. At the same time, human experts might feel reluctant to share tacit knowledge with AI due to ethical concerns [12] or fear of the loss of their jobs [13]. For the purpose of our study, the latter is less relevant, as we will focus on how organizations can retain tacit knowledge from senior experts that are actively preparing to depart from the organization or have already retired.

We see an opportunity to combine two powerful approaches for knowledge continuity. Tandems are already a proven method for InterGenTacitKT [14,15,16,17,18], [Interviewee F10], and organizations are becoming ever more creative and proficient in effectively using AI [4,19]. It seems to us that now would be the time for conducting further research on combining these two for knowledge continuity.

Based on our analysis, we will present possible synergies and challenges as well as potential implications for revitalizing workplaces. This article also advances the body of literature on the role of AI in InterGenTacitKT by identifying the gaps through a systematic literature review (SLR) and suggestions for further research.

Contribution to UN Sustainable Development Goals

Our study reflects on the possibilities of integrating artificial intelligence in intergenerational tandems for transferring tacit knowledge from senior or retired experts to junior employees. Collaboration in tandem arrangements can enhance participants’ enjoyment of their work, significantly contributing to employee commitment, productivity, and job satisfaction. Current research indicates that senior as well as retired experts are notably motivated by opportunities to both acquire new knowledge and share their expertise [20].

Incorporating AI into tandem scenarios necessitates participants to actively manage both the benefits and complexities arising from AI integration, as AI systems themselves are also subject to continuous evolution. Thus, actively involving senior experts in identifying efficient strategies for utilizing contemporary tools such as AI to facilitate their tacit knowledge transfer could increase senior experts’ enjoyment and well-being at the workplace and thus prevent early retirement.

By integrating AI in tandem arrangements, juniors gain enhanced learning opportunities, which could lead to improved skills, confidence, and career growth. The reciprocal learning dynamics involving junior employees, senior experts, and AI systems can therefore substantially contribute to the long-term effectiveness and sustainability of intergenerational tacit knowledge transfer processes.

In addition, by actively integrating (early) retirees into knowledge transfer tandems, organizations contribute to recognizing the expertise and lifetime contributions of senior or retired experts. Providing them with greater opportunities for engagement and influence contributes to the resilience and cohesion of society [21].

We believe that this approach contributes to UN Sustainability Goal 03, “Ensure healthy lives and promote well-being for all at all ages”, and Goal 08, “Decent work and Economic growth” (see Figure 1 and [22]) by offering new perspectives and actionable advice to revitalize workplaces and sustainable well-being in organizations, even though the content of this publication has not been approved by the United Nations and does not necessarily reflect the views of the United Nations or its officials or Member States.

2. Theoretical Background

In the following section we offer short summaries of the terms we will be using.

2.1. Intergenerational Tacit Knowledge Transfer (InterGenTacitKT)

2.1.1. Definition and Importance

The projected consequences for organizations of knowledge loss have been extensively researched [23,24]. This article focuses on tacit or implicit knowledge. To facilitate reading we will use the term “tacit” in this article.

As has been established by several authors, tacit knowledge evolves over time through the repeated application, integration, and sharing of explicit knowledge across diverse situations, interactions, and contexts [7,8]. This process enhances skill sets by fostering intangible competences, like contextual awareness, reflective understanding of the individual’s knowledge, and relationship-building [7,8,25,26].

We build on the extensive body of literature on the concept of “generation” [23,24,27] by selecting a fragment of the discussion: how can tacit knowledge be transferred from an older, more experienced employee to a younger, less experienced employee? This article focuses on the experience differential between senior and junior individuals, which, for the purposes of this study, is aligned with age.

We follow the research and analysis of Pfrombeck et al. (2024) regarding the definitions of “younger” or “junior” employees to be 35 years old or younger [28].

Age classifications for “older” or “senior” individuals differ across studies, ranging from the mid-50s to the mid-70s [29,30], with some sources considering individuals aged 50 and above as part of this group [31,32,33,34].

In this article, we adopt a working definition of senior experts as those aged 55 or older. This category specifically includes individuals who exited their primary employment before reaching the official retirement age—a group referred to as early retirees—and who may find it challenging to contribute to organizational success in new formats or contexts.

Empirical evidence indicates that senior team members contribute to shaping junior employees’ cognitive and social capabilities by influencing their approach to complex tasks, cooperative dynamics, and conflict resolution practices [1,35,36,37]. The ability to deal with ambiguity, stress, as well as the social aspect of a role’s responsibilities, which includes sensitive interactions with specific stakeholders, such as bodies of government, key customers, suppliers, specific employees, or senior managers, is crucial for success in certain positions [38,39,40].

2.1.2. Challenges for InterGenTacitKT

Gerpott et al. (2017) recommend that, in order to be able to absorb tacit knowledge, the receiver needs a framework of explicit knowledge into which they can integrate the additional knowledge [41]. Also, the maturity of the persons involved has as much impact on the approach to knowledge transfer as it does on dealing with potential conflicts and the approach to their solution [41]. Ongoing research emphasizes that establishing and maintaining a foundation of trust and mutual respect for the partners’ competencies and contributions is essential for the effective transfer of tacit knowledge [42,43].

Different studies have inferred that a significant age difference between colleagues may result in strongly dissimilar perspectives on a broad range of topics that might hinder the building of a trusting relationship between seniors and juniors [44,45]. This is reflected in the attitude towards AI in the workplace, which younger employees might perceive differently than older employees [46,47].

2.1.3. Methods and Success Factors for InterGenTacitKT

The practice of knowledge transfer continues to evolve and is heavily influenced by digital innovations. Given the context of an aging workforce and increasing shortages of skilled professionals, it is inviting to evaluate the application of AI specifically for the successful capture and transfer of tacit knowledge. In this section, “Methods and Success Factors for InterGenTacitKT,” we will present an overview of knowledge transfer practices before focusing on the formats that are deemed especially useful for the transfer of tacit knowledge across generations.

Mentoring, tutoring, advising, and coaching are habitually used interchangeably outside of academic contexts. These dyadic one-on-one formats are designed to support the learner in acquiring knowledge through a more individualized approach compared to traditional teaching methods [24]. Their design and objectives can differ significantly depending on context and intention [48,49,50,51].

When the focus is specifically on tacit knowledge transfer, several approaches have emerged as particularly effective.

Collaboration across generations provides junior staff with opportunities to engage with senior experts and is recognized as a potent instrument for facilitating the transfer of tacit knowledge from senior experts to junior employees, as that knowledge transfer takes place in an environment that promotes active engagement and communication, thereby optimizing the absorption and retention of tacit knowledge [35,36,43,44,52].

Storytelling has also been identified as a powerful medium for intergenerational knowledge sharing. It provides context, encourages emotional resonance, and conveys complexity in a form that is both memorable and accessible [52]. This narrative technique is applied in the Serious Games industry, where immersive scenarios help translate abstract knowledge into experience [53]. Nevertheless, experts caution that while Serious Games show promise, their effectiveness and acceptance as knowledge transfer tools remain inconsistent, and the results are context- and user-dependent [54,55,56,57,58].

Role or job shadowing allows receivers to absorb knowledge through observation [59,60,61] and effectively supports the transfer of tacit knowledge by exposing junior employees to real-time workflows, decision-making, and interpersonal dynamics. It can foster cross-functional interaction, enhance understanding of organizational culture, and bridge generational divides through inclusive engagement [62].

These diverse formats underline the insight that different types of tacit knowledge require tailored transfer methods, and individuals vary in how they best absorb such knowledge [59].

Summarizing the above research, the success of InterGenTacitKT can be supported by the following factors:

• The maturity of the receiver: this includes a solid subject matter foundation and an advanced level of self-awareness regarding their own values, emotional intelligence, and biases;
• The level of self-awareness of the senior, including a certain level of pedagogic skills;
• The complementary soft skills of the partners;
• Matching individual characteristics;
• The willingness and active contribution of both parties.

2.2. Artificial Intelligence

Artificial intelligence (AI) is a branch of computer science focused on building systems capable of performing tasks such as learning, reasoning, decision-making, and forecasting [63,64].

AI systems rely on computational procedures (algorithms) to process data, identify patterns, and make predictions or decisions. The quality and quantity of data are central to AI performance. AI also depends on statistical models, which are mathematical representations built by training algorithms on data, and on feedback loops that help refine performance through continual learning [5,6,65,66].

The existing literature highlights a broad spectrum of AI applications within the field of knowledge management, including process and workflow analysis and optimization, big data analytics, and various advanced computational methodologies. Applications include natural language processing (NLP), facilitating effective knowledge extraction and retrieval; machine learning algorithms for predictive analytics and knowledge discovery; semantic analysis and ontology-building tools for structuring and organizing complex information; expert systems designed to encode and replicate expert knowledge; intelligent search engines enhancing accessibility and usability of stored knowledge; and conversational agents such as chatbots, which support knowledge dissemination and interactive knowledge retrieval [19,63,67,68].

The anticipation that tacit knowledge could be captured and codified using AI, thus making it not just easier to transfer but to multiply this valuable resource without time constraints, has led to numerous studies across a broad range of fields.

3. Methodology

This study follows a qualitative, inductive, and interpretivist approach [69] to study social phenomena. In “an interpretivist epistemology, […] you will be concerned to understand the meanings that participants ascribe to various phenomena” [69], p. 324.

Grounded theory [70] was the process relied upon, which fosters ongoing reflection, and the literature is never really set aside, as new discoveries from the field work arise. Theory is built from the qualitative data collected.

An original set of 36 semi-structured interviews [20] led us to this current research effort, based on what 11 of those interviewees revealed. This subset of our interviewees brought attention to junior/senior tandems.

We also aimed to understand if and under which conditions AI could be deployed to boost junior/senior tandems’ success.

This brings us to our research question: if tandems are so effective in tacit knowledge transfer (according to [14,15,16,17,18]) and AI is ever more prolific in organizations (please see [4,19,64] for a discussion on this), how can we use AI to enhance junior/senior tandem outcomes?

The methodology followed for this study is best described by [71], p. 151, in that we listened to the “…multiple voices [and perspectives] in the data…”. And much as [72] demands (p. 915) that the knowledge of accounting researchers be combined “with the craft knowledge of accounting practitioners”, we compared the findings from the literature review with the findings from our qualitative research effort [71,72].

A thematic narrative analysis was utilized, emphasizing the identification and exploration of central themes and topics [69]. As Braun and Clarke (2016) state, thematic analysis demands the flexibility to find patterns based on study participants’ contributions beyond the original research question [73].

3.1. Participants

As the research is qualitative, the sample is small and purposive, involving participants chosen as they were seen to be able to contribute to answering the research question [71,72].

Let it be noted that when selecting the interview participants for the employees’ and retirees’ perspectives, we chose individuals free of financial need when they consider working beyond legal retirement age [20]. At the outset of this research paper, we decided to focus on white-collar experts and exclude physically demanding jobs from our study, as these deserve their own dedicated research [74,75]. Our study, instead, is building on research that concluded that older employees’ preferences and skills are best suited to office work [76].

For the employees’ perspective, we selected participants roughly ten years before they would reach legal retirement age. This reflects the framework established in the German Semi-Retirement Act, introduced in 1996. The primary objective of this law is to provide a framework for both employees and employers to manage a phased transition from full-time work to retirement, ensuring the employees’ financial stability in this process through the payment of specific benefits. Eligibility begins at age 55, which often prompts preliminary discussions about retirement options when employees enter their early 50s [77].

The perspective of the employer/recruiter was also added to the qualitative study. The distribution of the research participants is presented in

Table 1. Distribution of study participants based on perspective and gender.

Gender/Perspective	Employee Perspective	Employer/Recruiter Perspective	(Early) Petiree Perspective	Total	Percentage Across Genders
Female	7 (42%)	5 (29%)	5 (29%)	17 (100%)	47%
Male	6 (32%)	6 (32%)	7 (36%)	19 (100%)	53%
Total	13 (36%)	11 (31%)	12 (33%)	36 (100%)	100%

.

Table 2. Overview of study participants with the “employee” perspective, including industry, gender, and age at the time of the interview.

Employee Perspective
Industry	Age at Time of Interview	Gender
Finance/Leasing	50	M
Education	54	M
Life Sciences	54	M
IT Solutions	55	F
Service Industry	55	F
IT and Process Consulting	56	F
Transportation	56	F
Executive Search	58	F
Management Consulting	58	M
Management Consulting	60	M
Social Services	61	F
IT and Engineering Consulting	65	F
Manufacturing	67	M

,

Table 3. Overview of study participants with the “employer/recruiter” perspective, including industry, gender, and age at the time of the interview.

Employer/Recruiter Perspective
Industry	Age at Time of Interview	Gender
IT Solutions	48	M
Finance/Leasing	50	M
IT and Engineering Consulting	52	M
IT and Engineering Consulting	55	M
Transportation	56	F
Life Sciences	57	F
Executive Search	58	F
Public Administration	58	F
Social Services	61	F
Manufacturing	67	M
Executive Search	70	M

and

Table 4. Overview of study participants with the “(early) retiree” perspective, including industry, gender, and age at the time of the interview.

(Early) Retiree Perspective
Industry	Age at Time of Interview	Gender
Manufacturing	48	M
Life Sciences	57	F
IT and Engineering Consulting	63	M
IT and Engineering Consulting	65	M
Automotive	67	F
IT and Engineering Consulting	67	M
Manufacturing	67	M
FMCG	68	F
FMCG	68	F
IT and Engineering Consulting	69	M
Financial Services	73	F
IT and Engineering Consulting	76	M

each give an overview of the study participants’ perspectives as well as their (former) industries and positions.

The study participants were coded alphanumerically with “F” or “M” to identify the gender, “em” for the employee perspective, “er” for the recruiter or employer perspective, and “rt” for the retiree perspective. Citations will look as follows (examples):

• emF01 represents a female employee;
• erF06 represents a female employer/recruiter;
• rtM05 represents a male retiree.

3.2. Instruments

The semi-structured interview format enabled the unfiltered collection of the participants’ experiences, ideas, and suggestions. This qualitative method demands continuous attention to objectivity, reliability/dependability, generalizability/transferability, and validity/credibility [69].

Data collection and analysis occurred over a 12-month period sequentially and concurrently, facilitating continuous data verification and allowing for adjustments to initial interpretations, which took place in meetings with the research team that authored this article. All participants were previously informed about the study objectives and the voluntary nature of their participation; following their consent, they were afterwards invited to the interview at an agreed-upon time. The interviews lasted between 30 and 90 min and were mostly conducted online using Zoom, and—with the consent of the interviewee—they were recorded and fully transcribed.

The three distinct interview scripts (presented in

Table 5. Interview script 1—Questions to persons responsible for strategic decisions regarding HR and overall workforce deployment, e.g., CEO or head of HR—English Version—excerpt.

Interview Script 1 Questions to Persons Responsible for Strategic Decisions Regarding HR and Overall Workforce Deployment—Excerpt
Interview Question/Script	Justification for the Question from the Literature/from the Test Interviews
What benefits would you see in keeping experienced experts on board after legal retirement age?	[78,79,80,81] and test interviews
Under which circumstances would you be willing to hire employees who have already reached the legal retirement age?	[78,79,80,81] and test interviews
What do you see as the added value of very experienced employees?	[78,79,80,81] and test interviews
Where do you see challenges in hiring employees beyond the legal retirement age?	[78,79,80,81] and test interviews
What kind of commitment would you be prepared to give interested candidates regarding the form of the employment?	[78,79,80,81] and test interviews

,

Table 6. Interview script 2—Questions to employees—English Version—excerpt.

Interview Script 2 Questions to Employees—Excerpt
Interview Question/Script	Justification for the Question from the Literature/from the Test Interviews
How do you envision you will spend your time once you have reached legal retirement age?	[78,79,80,81] and test interviews
What are your thoughts regarding your physical and mental fitness once you retire?	[78,79,80,81] and test interviews
How does your private [and close] personal environment think about your plans?	[78,79,80,81] and test interviews
How does your professional environment think about your plans?	[78,79,80,81] and test interviews
Where do you see the difference between volunteer work and a job for which you receive a salary?	[78,79,80,81] and test interviews

and

Table 7. Interview script 3—Questions to retirees—English Version—excerpt.

Interview Script 3 Questions to Retirees—Excerpt
Interview Question/Script	Justification for the Question from the Literature/from the Test Interviews
While you were still working: how did you imagine filling your time once your retired?	[78,79,80,81] and test interviews
Now that you are retired: which activities do you find most fulfilling?	[78,79,80,81] and test interviews
What would you recommend to someone who is still actively working regarding their retirement?	[78,79,80,81] and test interviews
What do you miss most about your working life?	[78,79,80,81] and test interviews
Under which conditions would you consider sharing your years of experiences, your network and expertise with the younger generation?	[78,79,80,81] and test interviews

) tailored for each participant group were derived from an extensive review of the existing literature and refined through feedback obtained from preliminary test interviews.

In the course of the semi-structured interviews based on the above interview script excerpts, some of the interviewees discussed their experiences with forms of tandems. This led to asking them about the success factor of said tandems.

4. Literature Review

For our study we were interested in research that evaluated if and how artificial intelligence is being deployed to support tacit knowledge transfer across generations.

As a basis we build on the literature available on this topic, applying the PRISMA method [82,83].

The systematic literature review (SLR) was conducted in the online database Scopus in March 2025. Papers published and added to Scopus after that date were not included in this article. Only peer-reviewed articles in English were evaluated.

As keywords used for the literature review, we chose “tacit knowledge” and “artificial intelligence”, which produced 130 journal articles and conference papers. A first analysis of the abstracts led us to concentrate on publications after 2020 for our specific research focus due to the dynamic developments in this topic and the boost in digitalization across industries triggered by the pandemic in 2020. Figure 2 presents our interpretation of the PRISMA flow diagram for systematic reviews.

Thus, the literature we analyzed for our research fulfilled the following criteria:

• Peer-reviewed articles or conference papers published in Scopus between 2020 and 2025;
• Published in English;
• Having a business and managerial focus or transferability;
• Focusing on private sector industries and services;
• Adult sample groups.

Before we present the results of our SLR, we present the findings of our previous qualitative study, which are relevant for the selection and evaluation of the literature analyzed for this article.

5. Empirical Contribution to This Study: The Junior/Senior Tandem Strategy

For this study we are exploring particularly the data that was collected in the semi-structured interviews by asking the following questions whenever the interview flow led to it:

• Employers/recruiters were asked, “Could you imagine using retirees in tandem with younger colleagues in a job or position and how could this work?”.
• Employees and retirees were asked, “How do you feel about the concept of tandems of one ‘younger’ and one ‘senior’ sharing a job in a 70:30 or 60:40 ratio for a certain period of time?”.

A second analysis of the data with the focus on success factors for transferring knowledge via junior/senior tandems revealed that 11 participants (6 women and 5 men; all above 52 years of age at the time of the interview; 3 employees, 5 retirees, and 3 employers/recruiters) provided success factors for junior/senior tandems. Participants based these on their own experiences with similar formats or by drawing on their profound knowledge of evaluating and developing people.

In the following sections of this article, we will present the results and discuss the possibilities of whether and how AI might support in facilitating these success factors.

6. Results

6.1. Summarizing the SLR: The Use of AI for Tacit Knowledge Transfer

The prospect that AI could capture and codify tacit knowledge—thereby facilitating not only its transfer but also enabling the multiplication of this asset independent of time constraints—has stimulated extensive research across diverse disciplines.

This section provides an overview of these capabilities, including opportunities and challenges in relation to tacit knowledge transfer.

Tacit knowledge continues to be seen as a key component for improvisation, innovation, and new domain knowledge [11,68,84,85,86,87]. While one stream of literature posits that (intergenerational) tacit knowledge transfer combined with change adaptability are crucial for organizational innovation [86,87], Arias-Pérez et al. (2022) argue that for organizational improvisation, tacit knowledge is losing relevance when compared to AI, which can adjust in real time to organizational challenges [11], especially when companies may struggle to make tacit knowledge available without time limits and train their workforce fast enough to keep up with real-world dynamics [88,89,90].

The different perceptions can be explained with the different forms of tacit knowledge that are being discussed—relational, somatic, or collective [10]—as they will lead to different attitudes and expectations as to the successful codification [87,91,92,93]:

• Relational tacit knowledge refers to insights into somebody’s probable reaction due to an intuitive understanding of that person’s unspoken expectations. It develops over time through interpersonal interactions. It could, e.g., influence a solicitor’s decision regarding when and how to present information to a judge. A possible approach to codify this might be an AI analysis of past documents, e-mails, or video calls between the participants, which could be used to predict future reactions and suggest personalized approaches.
• Somatic tacit knowledge refers to physical or motor skills that have been acquired through repeated practice. Consider the trained hearing of a mechanic when listening to the noise of a faulty engine before reaching for their tool. A possible approach to codify this might be to record sounds, register possible vibrations and similar signals, and use AI to analyze the captured data to suggest solutions.
• Collective tacit knowledge refers to “unwritten rules” within groups that develop through shared experience, practices, and norms, thus giving intuitive guidance in new situations. It can be observed at informal company gatherings that follow different patterns from organization to organization. A possible approach to codifying this might be to use AI to analyze past videos, photos, and any past internal communications, creating AI-powered guidelines for effective team building.

Studies suggest a variety of approaches for AI agents to absorb tacit knowledge, like a simulated environment [94], including AI in serious role-playing [95], the analysis of associations and nuances in languages and texts of documents, reports etc. [96], or using a sensor framework that analyses users’ faces, speech, and app usage to combine this with context awareness to absorb tacit knowledge and make it available for other users [97,98]. For optimal results, de Souza et al. (2023) emphasize that AI should also be in the loop when human experts collaborate [84]. This ties in with studies advising that the knowledge transfer needs to be a two-way road between experts and the use of AI for improving outcomes and the effective usage of information [99,100].

Various forms of AI support users not just by accessing stored tacit knowledge, but by providing tailored knowledge recommendations building on the analysis of each user’s individual knowledge and implicit logical reasoning [101,102], enabling, e.g., doctors to accurately diagnose diseases and suggest appropriate treatments based on analyzing stored data that includes “tacit knowledge” but cannot replace human judgement [19,103].

Authors that analyze the behavior of AI are asking if these systems themselves display tacit knowledge when their decision-making process seems to defy statistical analysis [104,105,106], while other authors argue that AI lacks the potential for intuitive decision-making, as intuition is based on individual experience and subconscious knowledge and competences, and AI relies on structured methods to turn a stream of data into decisions [4,87,93]. This is supported by studies that find that in fields with potentially high variability in the outcomes, even after capturing tacit knowledge to streamline complex workflows, human judgment and intuition need to be in the loop [68,92,100,107,108].

Several studies have thoroughly researched limitations in the use of AI and emphasize the investments AI necessitates. We recognize the importance of topics like data privacy and security, a user-friendly interface, and the appropriate functionality of the applied AI; however, in the context of this study, these aspects, as well as financial investments, will not be addressed.

Trust in AI is repeatedly highlighted as crucial for its effective use, but it is limited by a lack of confidence in the AI’s algorithms, the risk of proliferating biases, and the awareness that AI lacks ethics and accountability [4,19,90,99,109,110]. Transparency and reliability of AI systems can positively impact trust in AI [6,13]. As trust has been identified as a core feature of successful tacit knowledge transfer [42,43], addressing this issue is paramount, especially if experts are expected to rely on AI to make high-risk, high-cost decisions (faster) based on AI recommendations [6,111,112].

6.2. Thematic Analysis: The Organizational Actor’s Lens

A second analysis of the data collected through the interviews on the success factors for transferring knowledge via junior/senior tandems revealed that 11 participants highlighted the following factors as crucial for the success of JuSeTs:

• Clear definition of the roles of each member of the tandem;
• Complementary skills of both;
• Matching the personalities of both;
• Trust between both.

Table 8. Study participants on success factors of junior/senior tandems.

Theme or Pattern Identified	Perspective	Highlight/Interview Excerpt	Code for Interviewee
Clear definition of roles	employee	“A tandem in which I do a lot of the work would almost be like having two part-time jobs. It would depend a lot on the content whether it works. We need very clear agreements, and the organization has to be very well coordinated.”	emF03b
	employer	“… that one senior is matched with 2 or 3 juniors… if I formed [just one 1:1] tandem…, the older person might… try to dominate the younger person, and that conflicts would arise as a result.”	erM14
	employer	In a client meeting, the senior “… knows when to keep quiet, when to listen, when to dig deeper, when to let things slide, when to tighten the reins, and so on. You don’t know this if you’ve learned the method… but haven’t yet applied it…”	erM15
	retiree	“… precise contracting…” between the junior and the senior.	erF06
	retiree	“… as senior I had the standing [with other partners] to call-out no-gos…[giving the Junior the example and confidence when and how to do that in a similar situation]…”	rtM08
Complementary skill sets	retiree	“… he was more pragmatic in many ways, I was more precise, and we were good as a team….”	rtM07
	employer	The senior with the experience, the younger bringing additional subject matter expertise and the potential to grow into the role.	erF06
Matching personalities	retiree	“… at least someone who initiates it and sees if the two of them get along well enough…”	rtM09
	retiree	“… both need to choose and want this [tandem]… or it will never work…”	rtF07
	retiree	“… we understood each other perfectly, even though we were different”	rtM07
	retiree	Both need to say “… it fits, or it doesn’t…otherwise it will just be painful…”	rtM10
Trust	employee	“I had assumed that [the senior I was working with] was not in competition with me. It has since become clear that he was not so well disposed towards me and…adorns himself with my feathers…”	emF09
	employee	“… a proper alignment …openness and trust…and then deliver on the promises…”	emF12

presents the identified success factors for JuSeTs, including interview excerpts.

Participants based these reflections on their own experiences with similar formats or by drawing on their profound knowledge of being in a role of evaluating and developing people.

In view of the complexity of tandem situations, it is logical that clarity of roles was the factor that was most emphasized by the interviewees, as it is the most easily controlled factor in that format of InterGenTacitKT. Depending on the urgency of the situation, complementary skill sets and matching personalities of juniors and seniors might be considered a luxury, while achieving trust takes time [113]. It is also apparent that the empirical data reveals requirements regarding relational and collective tacit knowledge. In the next section of this paper, we present approaches to how the tools that AI offers could be used to support all four of these intangible success factors.

7. Synthesis of Current Research Findings: The Possible Role of AI in JuSeTs for Successful InterGenTacitKT

In the following we present suggestions for the integration of AI in InterGenTacitKT with the focus on JuSeTs. We merge insights from the systematic literature review (SLR) with our empirical findings, concentrating on the success factors for JuSeTs (clarity of roles, complementary skill sets, matching personalities, and trust) to present opportunities for AI to provide relevant contributions to enhance InterGenTacitKT. Depending on the form of tacit knowledge—relational, somatic, or collective—different approaches may apply.

As we documented, trust between the involved parties of tacit knowledge transfer is crucial for its success [42,43,114,115]. AI adds complexity if the organization encourages the use of AI, but trust in AI is lacking [6,110]. AI is seen as negatively impacting communication between colleagues [46], but it can be used to support social exchange when face-to-face meetings are challenging [90]. Adding to this, potentially fundamentally different attitudes that tandem partners may have towards AI and its contributing role in InterGenTacitKT make its use even more questionable. Thus, we start our list with AI’s possible contribution to building trust between tandem partners:

7.1. Trust

The influence of interpersonal trust on successful InterGenTacitKT is quite simple to illustrate, as is shown in Figure 3.

It becomes more complex when we add trust in AI, as is shown in its most basic combination in Figure 4.

Part of building trust between juniors and seniors is a mutual respect for the partners’ competencies and contributions. AI could provide both partners with additional context and insights personalized to the users’ needs, preferences, and expectations. This could contribute to understanding the tandem partners’ perspectives and enhance mutual respect, and, thus, facilitate smoother interactions.

Working on the assumption that the organization provides transparency to bolster trust in AI, the tandem’s acceptance of technology-supported tacit knowledge might also increase.

7.2. Clear Definition of Roles

If the senior of the tandem is—or was a former—employee of the junior’s organization, their contribution might start by revealing where relevant information is stored and how it can be accessed—whether these sources are other experts or dormant data repositories [116,117]. The senior’s organizational knowledge (including its history) may steer the junior and AI to relevant sources, dependent on the maturity of the digitalization of this organization. While the junior starts building their own relationships and network guided by the senior, the AI can complement and accelerate tacit knowledge availability by accessing digitized (personnel) files, correspondence, contributions, evaluations, and other relevant data repositories [102].

7.3. Complementary Skills

Empirical data underscored the importance of complementary skills within JuSeTs. AI can analyze the partners’ detailed educational and work history as well as their preferred style of working, communicating, and learning. This might highlight complementary skills and diagnose skill gaps, thus suggesting task division in the tandem that provides targeted training opportunities for the junior. The empirical and SLR data highlight the importance of applying absorbed tacit knowledge for long-term retention. AI systems can support real-time decision-making processes within JuSeTs by providing access to relevant historical insights and recommendations tailored to specific scenarios, thus complementing the senior’s intuitive expertise with data-driven suggestions [67].

7.4. Matching Personalities

Both empirical results and the literature emphasize the importance of effective matching for successful JuSeTs. Understanding and utilizing the contribution each participant (senior, junior, and AI) brings to the table can lead to a powerful combination [5]. AI could support the matching process of tandems, building on AI systems that are being used in recruiting. By analyzing personality traits and interpersonal dynamics when combining tandems, compatibility can be significantly improved. AI insights might even highlight experts within the organization that have been less visible to the decision-makers and thus add resources to the knowledge transfers [100].

To summarize, AI can be a facilitator in the transfer of tacit knowledge, but it also adds complexity. As the tandem partners may have different attitudes towards AI in the workplace, laying a foundation of trust in the available AI needs to be a priority for the organization [6,13].

Building on that, AI can offer strong complementary services for InterGenTacitKT by fostering the identified success factors for JuSeTs (clarity of roles, complementary skill sets, matching personalities, and trust), thus offering organizations a powerful means to enhance the effectiveness and sustainability of InterGenTacitKT that also strengthens employee productivity, satisfaction, and loyalty and overall organizational competitiveness.

8. Discussion

A key distinction exists between general knowledge sharing and the specific case of knowledge transfer from senior experts or (early) retirees: Generally, employees are motivated to share knowledge due to the potential for career advancement through the acquisition and subsequent application of new knowledge [118]. However, when the explicit intent of such knowledge transfer is to render the senior employee redundant, organizations face the challenge of identifying effective incentives to encourage these senior employees to actively participate in the transfer process [117].

JuSeTs can be an efficient and effective approach to transferring tacit knowledge and harvesting the experience and know-how of experienced employees. Working in tandems can contribute to the participants’ enjoyment of their work, which is a considerable contributor to employee commitment and productivity [14,47,90].

Learning something new as well as sharing their expertise is today a well-established motivation for senior experts and retirees when engaging in intergenerational knowledge transfer [20,35,41,44,50,118,119,120,121,122], and the different intergenerational skills and knowledge allow many venues of learning for both parties.

Organizations have long realized that even with modern technology like data analytics, Serious Games, and AI, tacit knowledge requires creative and individual ways of being transferred [8,36,59,123,124,125,126], and these interactions need to be well prepared, structured, and evaluated [51,59,127]. So, involving seniors and (early) retirees in finding efficient ways to use modern tools like AI in transferring their tacit knowledge might be a promising approach.

Trust within organizations has been repeatedly emphasized as an important factor for organizational performance [114,115]. Considering the importance of trust in the successful transfer of tacit knowledge, the attitude each participant in the process brings towards AI will influence the effectiveness of AI [109,128]. With AI becoming more and more an integral part of any organization, trust in AI needs to be addressed [4,110].

Working under the assumption that AI is giving organizations a competitive advantage, for instance by enabling organizations to extract information from dormant data like maintenance reports and technical literature, rapidly synthesize numerous medical files, or standardize processes [11,109,111,129,130], engaging motivated senior experts to find ways to codify their own tacit knowledge could represent a successful strategy for long-term success.

For the contribution of AI to be experienced as useful, accurate, and relevant, authors suggest the collaboration of multiple experts, adding to each other’s domain and tacit knowledge when “feeding” AI [91,108,131]. Applying a network analysis of human experts’ decision strategies [132] can be used to create curated repositories that enable efficient transfer of high-quality knowledge to individuals, to groups, and into the organization, and ideally further facilitate the integration of AI.

9. Conclusions, Limitations, and Recommendations

In the quest of organizations to increase efficiency, AI has proven in a very short period to be a universal tool and phenomenon, running across all age groups, from young to old. Hence, AI can serve as a facilitator in the transfer of tacit knowledge, although it also introduces additional complexity. Since senior and junior employees may hold varying perceptions regarding the use of AI in professional settings, establishing a foundation of trust towards available AI solutions must become a strategic priority for organizations.

Building upon this rationale, AI can deliver substantial complementary support, e.g., in the decision-making process for InterGenTacitKT by enhancing the key success factors identified for JuSeTs, namely clarity of roles, complementary skill sets, matching personalities, and trust, as we have demonstrated in our results.

It is important to emphasize that AI does not replace human tandems but rather improves their efficacy. Just as AI, in isolation, does not sufficiently address the challenges of tacit knowledge transfer, neither do JuSeTs alone guarantee comprehensive knowledge continuity within organizations.

A limitation of this study is that while thematic analysis within a grounded theory framework offers a systematic approach to qualitative data, it necessitates explicit acknowledgment of the context-specific nature of our interpretations and their settings. Due to the complexity and dynamics of this topic, our study’s limitation is related to the very specific format of the JuSeTs we studied for InterGenTacitKT. Our study still may serve as a building block for future research, which could address the following areas.

Ethical considerations and the establishment of best practices regarding AI integration in intergenerational workplace contexts could be explored. Empirical studies could be conducted to assess the impact of AI-driven interventions on tacit knowledge transfer effectiveness. Also, building on our results, future research might focus on the specific perspective of younger employees.

It could also be valuable to establish an overview and evaluation of available systems in the market that offer how to capture and make use of tacit knowledge within organizations.

Adding AI into a tandem situation can achieve more than merely transferring knowledge. As tandem partners navigate the advantages and challenges associated with AI integration, they simultaneously advance their capabilities in cooperation and collaboration and their proficiency in leveraging AI, thus strengthening employee productivity and satisfaction. At the same time, AI systems themselves evolve, refining their ability to effectively support humans in intergenerational tacit knowledge transfer contexts. This reciprocal learning process between senior experts, junior employees, and AI can enhance the sustainability of InterGenTacitKT. AI could function much as having a coach or objective expert opinion at hand, 24/7. Is that the future role and essence of AI? Only time will tell how AI will evolve [133] and what role humans will play in steering that evolution. This is fascinating, but also a true challenge for human beings.