Introduction

Organisations across different sectors had experienced the advent of technology in each function among which, artificial intelligence (AI) is regarded as the disruptive technological development (Faix, 2022; Kong et al., 2021). Also, Prakash et al. (2022) had confirmed that the progress in technology not only changed the way organisations operate but it also disrupted the conventional way of work and lifestyles. According to Almeida (2022), disruptive technology refers to the advancement in the science and technology that leads to the fundamental restructuring of existing methods. AI is defined as a complex machine learning application that has a data processing approach such as deep neural networks, simple algorithms relying on regression analyses, natural language processing or voice recognition (Hunkenschroer & Luetge, 2022). Since AI is an emerging concept the definition of AI lies within the scope of the tasks it currently handles. AI is an emerging concept, and its abilities and futuristic tasks would modify the present definitions used to describe AI (Kaplan & Haenlein, 2019; Yarger et al., 2020). Technological advancements in data management have become an essential part of every business with the ability to measure productivity, efficiency and operational excellence (Minbaeva, 2021). Currently the data management for the multi-utilitarian reasons in business is assisted by the existing tools that have become the preform of the developments in AI research. However, the role of AI in the business management is considered novel and in a nascent stage on the timeline of technological evolution. Implementation of AI in the primary activities of business did not assume the full scale yet. This could be due to the organisations technological needs met by the technologies that have intense human intervention. Confirming this Fountaine et al. (2019) say that many companies limit AI to secondary projects and struggle to scale up the use of AI in their regular and key activities. AI can be often seen implemented in the areas, viz. virtual merchandising, experiential marketing, customer experience management and supporting the expert systems. Abdeldayem and Aldulaimi (2020) say that AI is still believed to be at the modest pace in blending with the human resource management (HRM) operations. Nevertheless, the growing use of intelligent systems in HRM at the different layers of organisation structure is seen in several organisations in some way. This growth is modest due to several technical and behavioural reasons that prevail among the HR personnel and departments. However, research and experimentation related to the HR related AI tools is widely happening (Abdeldayem & Aldulaimi, 2020; Fountaine et al., 2019). To discuss AI concept in detail, this paper followed the sequence covering the objectives of the study, methodology adapted for the study and review of literature that includes subsections discussing the evolution of AI, AI and business and AI in HRM and sub-functions. The next section in the review of literature is the existent and emerging AI tools in HRM functions. The following sections include conclusion, managerial implications, limitations and further scope. Setting this as the context, this study works on the below objectives.

Objectives

To study the evolution of AI. This paper would try to study the origins of AI and track its evolution over the decades.

To study the adaptation of AI into the key business functions. While the technological advancements have changed the distinct functions of business, the status of the adaptation of AI in such functions has not been widely reviewed yet. Considering these, this study will explore the adaptation of AI among the organisations.

To study the role of AI in HRM and its sub-function. With the specific interest to explore the role of AI in the HRM activities, this study investigates the AI’s impacting role in HRM operations.

To study the application of AI tools in the HR function. Studying the emerging AI tools and technologies in the field of HR is another objective of this study.

Methodology

A conceptual method is adapted to study the selected subject extensively. Research and implementation of AI are studied form the secondary data from the authentic sources. To find pertinent publications, Boolean operators and keywords relating to AI’s involvement in HRM functions were used to search academic databases, such as PubMed, Google Scholar and relevant journal archives. Predetermined inclusion criteria, such as publication date (not prior to 2000, except a few articles that were related to the theories), academic credibility and relevance to the research topic (AI and HRM), were used to pick articles and sources. Critical analysis was employed to assess the literature’s reliability and applicability, with a focus on finding recurring themes, patterns and contradictions in the context of AI’s role in HRM functions. The synthesis of the data was used to derive broad conclusions and insights that served as the foundation for this paper. This helps in developing the knowledge on AI and HRM from both the academic research and the industry.

Review of Literature

Evolution of AI

AI was defined in different types based on the research objectives and the technological purpose that AI was developed for (Palos-Sánchez et al., 2022; Welsh, 2019). McCarthy (1956) and Minsky (1968) stated that AI is the science of creating intelligent machines that can imitate the humans with precision. Later studies have agreed that AI is the technology that makes machines become self-learners and possess the cognitive abilities of the human being (Paesano, 2021; Palos-Sánchez et al., 2022; Tambe et al., 2019). Cox and Mazumdar (2022) confirm this by reiterating that AI is a tool that has the ability to perform the tasks that humans normally do. Several scholars in the past had contradicting opinions on how AI could replace the role of humans in decision-making. Few scholars and scientists have strongly believed that technology would take over the decision areas, while others believed that technology could only leverage the human decision-making process but would not replace the human mind entirely (Russell, 2021). In spite of the conflicting opinions of the stakeholders, data processing and predicational abilities of systems have dramatically multiplied in the last two decades.

The beginning of AI could be dated back to 1960 when researchers in the computer science and mathematical logicians had developed methods for data assertions and algorithms for data inferences (Russell, 2021). Since then, AI had started assuming the role in the automation of manufacturing and operations management not widely though.

In the 1980s, AI researchers had started empowering AI in dealing the uncertainties in real-time tasks by acquiring human thinking abilities which is later called as machine learning (Pearl, 1988; Russell, 2021). During the 1980s computers also have started learning from the experience, a technique called ‘deep learning’, which was popularised by the John Hopfield and David Rumelhart (Anohya, 2017). The further developments till 1990s in the AI can be asserted by the govt of Japan, which invested heftily in researching expert systems, system processing, logic programming and other AI-based technologies.

After the failure of the Fifth Generation Computer Project in meeting its commercial objectives in the early 1980s, concerned individual scholars and the engineers shifted their research area to explore AI extensively (Anohya, 2017). In the later years, AI could be seen being involved in key functions, such as manufacturing, customer experience management, big data analytics and other key business functions.

AI and Business

Coexistence of AI and business could be seen from the late years of the 19th century. AI had forayed into the most industries lately by making its mark on different business functions (Zhang et al., 2016). AI-based technological tools and applications started assuming the key roles in managing the backend operations of the business while delivering the customer experience at the frontend (Karimi et al., 2009). Palos-Sánchez et al. (2022) say that the developments AI-based technologies have taken the new dimension in organisational excellence by enabling speed and precision in the decision-making. AI tools have helped organisations to concentrate on value-centric activities across all functions by extenuating the mundane tasks from the routines of the managers. For instance, HR managers could evaluate an applicant based on the data extracted by AI tools from social presence of the candidates they hire. Also, employees’ performance could be evaluated and managed without spending much time on the manual processing. Tools like IBM Watson help the managers to handle such routine tasks effectively (Palos-Sánchez et al., 2022). Customer service in marketing function has become real time by providing online support and suggestions with the help of predictive analytical capabilities that AI had added to the (CRM) systems (Chen et al., 2012). Supply chain operations are disrupted due to the contingencies and the irregularities involved in the business. This leads to delays in the material flow, failed fulfilment of orders and challenges in the stock management. Implementation of AI in the supply chain operations had addressed these challenges by efficient management of data and making the supply chain functions proactive (Bodendorf & Zimmermann, 2005). Also, other back-end operations of the business such as resource planning, warehouse management and logistics are effectively optimised, resulting the cost savings on the overheads (Karimi et al., 2009). Apart from its implementation in the business, AI applications also have assumed an important role in making weaponry more autonomous and deadly in the defence sector (Heller, 2019). Unmanned aircraft and drones, machine-learning solutions targeting the specific tasks and autonomous intelligent radar systems, etc., are the few examples.

Figure 1. AI Timeline: 1970–2000.

Source: Anohya (2017).

AI in HRM and the Sub-functions

Kim et al. (2015) state that ability, motivation and opportunity (AMO theory) effect the HR personnel in delivering the best possible results by effectively handling all HR tasks and practices. Minbaeva (2021) concurs with this and states that disruptive technological advancements have become essential in managing organisations and HR tasks. Kaur et al. (2021) predict that the use of intelligent systems will become indispensable in the near future of HRM operations, though the present pace of research in the AI and HRM is considerably low. This section discusses how technology and intelligent systems have discovered innovative ways of performing the tasks in the below different sub-functions of human resources management:

AI and HR planning and workforce design

AI’s role in digital selection and recruitment

AI in performance management

Implementation of AI in training and development

AI and HR Planning and Workforce Design

HR planning and workforce design is a strategic practice that takes the leverage of data analytics to make resource optimisation recommendations for the organisational development (Doll, 2022). Predicting the requirement and planning to balance the hiring with the optimised cost is the key essential for effective HR planning and workforce design (Kurniawaty & Rosyadi, 2022). The growing organisational activity and the subsequent need to cut the costs throw challenges to the HR for effective hiring. Effectiveness in hiring and employee engagement lead to the productivity, whereas failing to recruit people with the appropriate competencies will lead to business failure (Čanković, 2015). Technological advancements in the business had given advantage to the HR managers to assess the candidates even before the personal interaction happens. For instance, the online social presence of the individuals had enabled the HR to conduct the social media screening with the help of data posted by the candidates on their social media pages (Jeske & Shultz, 2016). However other earlier studies have expressed concern about the reliability of the information provided by the candidates on the internet (Davison et al., 2011; Bentley, 2013). Amidst all the speculations about the impact of technology on HR planning and workforce design, intelligent systems had started their course of action. AI-based tools are helping HR executives in making strategic decisions related to the planning and workforce design. AI tools have been implemented in sorting the multitudes of the job applications received, assessing the gestures of the candidates during the online interviews, and updating the applicants of their hiring status. This had helped the HR function to reduce the grunt work on HR personnel (Zielinski, 2017). A San Francisco based HR tech company FirstJob had created an AI tool ‘Mya’, which works as an HR assistant to answer varied queries of the candidates on organisation policies, benefits and the organisational culture (Zielinski, 2017). Intelligent applications of this kind have trimmed off the non-value-added tasks from the activities of the HR personnel. Another tool that is in the testing phase is X.ai, which, if successfully launched, could be deployed in the interview scheduling and calendar management of the HR executives.

IBM Watson adds new competencies to the HR as an intelligent talent management tool. It enables HR to the social media listening for understanding the aspirant insights, provides predictive analytics on the success probability of the candidates and prioritises the positions to be filled by performing the requirement analysis fed in the HR systems (Feuss, 2015; Maity, 2019; Zielinski, 2017). While such intelligent reflexes have been performed by AI applications, several HR and technology experts still opine that AI tools have a risk of failing due to the patterns they have learnt from the historical data and inadvertently ignored the recent developments which are important and sensitive to factor in. Implementation of AI tools in the real-time selection tasks of HR had been initiated in the global companies like Unilever, Amazon, Proctor & Gamble, etc. Also, SAP claims that its clients like Standard Charted, Microsoft, Whirlpool Corporation, etc., have been impacted immensely with their HR analytics and workforce planning application.

AI’s Role in Digital Selection and Recruitment

Challenges involved in finding the competent candidates can be addressed by the advancements and innovations in technology. Mehrotra and Khanna (2022) describe that AI helps recruiters to identify, screen, assess and select the right candidates. This helps in creating a significant talent pool for the organisation. This is done by automation of workflow, eliminating the repetitive high-volume tasks with reduced turnaround time and cost. Companies are extensively deploying AI tools in their selection and the recruitment process to increase the pace and efficiency of operations (Hunkenschroer & Luetge, 2022). People Management (2023) reports of Amelia, which is a virtual HR agent, a quick learner, multilinguist and manages multiple tasks efficiently. This tool was developed by IPSoft who is expecting that eventually Amelia will be responsible for all the recruitment related tasks. The selection and recruitment process are becoming complex with the increasing number of applications and the complex parameters to shortlist the candidates. Intelligent tools like Amelia are helping HR in sorting and managing the information efficiently. Agreeing to this, Rąb-Kettler and Lehnervp (2019) argue that the adaption of technology in the selection and recruitment process gives the best experience to the recruiter and candidates participating in the process. Interestingly the AI-based systems had gone to the extent of decerning the candidates’ personal details, viz. age, gender, confidence levels and the chances of selection through the voice and image recognition systems (Dattner et al., 2019; Oswald et al., 2020; Simbeck, 2019). However, few scholars have argued that intense use of intelligent systems that provide too many personal details of the candidates may cause biased opinions among the selectors (Fernández-Martínez & Fernández, 2020). Screening algorithms have been a part of the selection process, which would ask the candidates a set of predetermined questions during structured interviews. AI-based systems not only evaluate the responses but also assess the expressions, emotions and the body language of the candidates using audio and facial recognition tools (Chamorro-Premuzic et al., 2016; Fernández-Martínez & Fernández, 2020; Köchling et al., 2021; Tambe et al., 2019). Despite the benefits that AI provides in the selection process HR has to handle several other challenges pertaining to the use of technology. First, data literacy of the hiring managers is essential in using the AI tools effectively as the results have a massive impact on the applied candidates (Fernández-Martínez & Fernández, 2020; Hunkenschroer & Luetge, 2022; Simbeck, 2019). Second, the programmers developing the algorithm must understand the business processes and the standards that hiring organisations follow for assessing the applicants. Failing to comprehend such requirements while developing the tools may incapacitate the tools from doing the job it is meant for (Hunkenschroer & Luetge, 2022; Oswald et al., 2020). Third, factoring in the important ethical factors is a greater challenge when the human minds are prone to follow the old patterns and beliefs with the excess amount of data available. For instance, to overcome the ethical issues some AI companies have removed the code that identifies the CVs on the basis of gender to mitigate the unconscious bias among the selectors (Hunkenschroer & Luetge, 2022; Yarger et al., 2020). Mehrotra and Khanna (2022) had seen a significant impact of the AI recruitment tools on the experience of the HR personnel and candidates. Such tools are widely used in the organisation multinational firms like IBM, Earnst and Young, Infosys, L’Oréal, Unilever, Bajaj Allianz, and Mindtree (Mehrotra & Khanna, 2022). Also, a few HR start-ups, namely HireVue, Pymetric, Para.ai, Hiretual and Darwinbox, have started implementing the AI tool in the selection process. If HR could integrate with the programmers in developing the tools, AI applications could become the most reliable technology for selection and recruitment in the future.

AI in Performance Management

Performance management effectiveness is the alignment between the organisational goals and the individuals employed (Awan et al., 2020). The basic purpose of performance management systems (PMS) is to help the managers in decision-making by ensuring the right information is accessible to them as the decision inputs (Sahlin & Angelis, 2019). As a part of the technological evolution data-driven management is adapted widely that is called algorithmic-management, which helps organisations in automating the tasks and processes (Sahlin & Angelis, 2019; Schildt, 2017). In this context, performance management tools developed in recent years have proved to drive the HR interface through the conventional realm. Buck and Morrow (2018) argue that performance management is not a one-time activity but rather a continuous data-driven conversation that requires more touch points and the real-time information management to give the employees the transparent and constructive feedback they deserve. Wang (2021) says that managing the objectives and key results (OKRs) and key performance indicators (KPIs) have reached almost precision in quantification with the implementation of technology. With their ability in collecting, processing and analysing the voluminous data AI tools have brought the new efficiencies in the PMS (Wang, 2020).

AI-based performance management tools help in mitigating the gaps in the communication between the employee and the management by creating a real-time or near real-time face-to-face interaction and enabling the interaction through the chatbots subsequently. This results in retaining the talent by realigning the predicaments created among them for any reason not identified by the organisation earlier (Buck & Morrow, 2018). The intelligent systems like ‘Plause’ work on data analytics and try improving the relation between the managers and their teams by enabling the managers to understand their teams’ insights on different time frequencies before arriving to the appraisal time (Buck & Morrow, 2018). Also, other similar applications, viz. ‘Humu’, help the individuals to measure the key indicators and direct the employees and managers for self-evaluation towards a positive behavioural change, thus leading to the productivity and efficiency. The younger generation who represents the majority of work workforce in all the industries expects such an innovative method in the performance management where precision, transparency and integrity are the key elements (Buck & Morrow, 2018; Sahlin & Angelis, 2019). However, the AI-based applications in PMS are yet to occupy significant space in many organisations.

Implementation of AI in Training and Development

Salas et al. (2012) say that organisations spend billions of dollars on the training and development activities that enable organisations to adapt the change. This helps organisations to compete effectively, excel and innovate in the operations and develop sustainability across the organisational activities. Huselid and Becker (2011) affirm that well-organised training and development activities create the competitive workforce, optimisation of human resources and results in value and productivity across all the functions of business. For this reason, management contemplates on what to train, how to train, what are its implications and how to measure the training results. The increasing specialised skill requirements in the organisations started demanding the training and development activities to be more challenging and result yielding (Salas et al., 2012). Such complex requirements in the training and development have been addressed by the intelligent tools powered by the AI which are sustainable in nature (Maurya et al., 2023). These tools have initiated the personalised learning, intelligent training methods, automated content creation, adaptive testing, virtual assistance to the learner and trainer and predictive analytics and assessments. Fountaine et al. (2019) argue that organisations must start AI academies to promote AI internally and implement the AI tools to identify the taining needs at every organisational level. Maity (2019) says that AI has the ability to assess employees and map their performance to the required training needs. Also, AI could customise the learning methods that suit the learner and identifies the right trainer to the trainee. These AI training tools also help the HR and the employees to determine the training features such as duration, training pace, frequency and the assessments. Thus, organisations could mitigate the manual intervention in identifying the training needs and the selection of trainees, while enhancing the effectiveness of training by creating interactive platforms that make trainings more trainee centric (Maity, 2019). In addition to assisting in the training and development activities, AI is also used in the creation of the knowledge management environment to make the information accessible to the target groups in different context. By creating the AI-based collective knowledge management systems organisations could increase the information accessibility through the repositories that are loaded with the data in the forms of charts, case, scripts and other user-friendly visual formats.

Srimannarayana (2011) says that most organisations do not measure the impact of training on the individuals and the organisations in a routine, as they believe that measuring the results is expensive and several other pre-existent challenges make it difficult to measure. These challenges are addressed by the AI systems effectively by measuring and interpreting the results of the training into the actionable items at the individual trainee level (Maity, 2019). SAP Litmos is one such application that offers AI-driven learning management solutions to the organisations. Litmos serves a verity of clientele like LaborMAX, Epping Forest, Celigo and Ford Direct. Degreed is another California-based upskilling and technology company that uses AI for skill development of its employees. AI systems in the training and development define and develop the parameters by which training could be conducted by integrating the requirements of the trainee and the organisation.

Existent and Emerging AI Tools in HRM Functions

Several AI tools such as genetic algorithms, fuzzy sets and artificial neural networks (to name a few) are now being used in various functional areas of organisations (Arrieta et al., 2020; Holland, 1992). Ong (2019), which had stated that AI is growing rapidly in the areas, viz. Intelligent Talent Acquisition, employee engagement. Tools like Voice of Employees (VoE) and Virtual Assistant (VAs) have become a great help for HR personnel in this context. However, according to other few scholars (in the Asian perspective) AI’s scope has also expanded to the digital recruitment, candidate’s employment choice decisions, etc. (Linscott & Raghuraman, 2020; Mearian, 2023; Pillai & Sivathanu, 2020). Precisely, AI technology is revolutionising the HR function to emerge futuristic by digitalising the tasks at every layer of HRM (Huan et al., 2006). In the recent years, HR personnel are being supported by the intelligent systems in operations by an AI-based tool ‘Mya’. This tool was developed by a San Francisco based HR tech company ‘FirstJob’. As a VA in the knowledge management system, Mya handles interaction with the candidates and helps them understand the organisation in detail with ease (Zielinski, 2017). Empirical evidence from a study by Murugesan et al. (2023) reveals that implementation of AI tool in most of the IT and ITES organisations in India has resulted in the greater employee satisfaction.

IBM Watson is an advanced tool developed based on AI technology and adapted by HR managers in planning staffing and optimising the recruitment. IBM Watson also does analytics to make recommendations on the prioritisation of the required profiles. It predicts the right candidature and analyses the probability of resignations and exits with 95% of confidence. Also, IBM Watson assists the HR personnel in handling complex data management tasks and creates the enhanced employee experience in the organisation (Feuss, 2015; Jeude, 2020; Maity, 2019; Zielinski, 2017). Payroll management is empowered by the tools like ‘AccountingSuite’, which also had extended its functioning to the Cloud platform (PR Newswire, 2023). Amelia is another conversational AI-based virtual HR assistant that helps recruitment processes, manages employee learning and provides the cordial learning experience to them. Amelia is developed by ‘IPSoft’, which has also diversified Amelia’s abilities in different product versions for supporting the customer service and experience (People Management, 2023). Buck and Morrow (2018) mention in their study an intelligent system named Plause, which works on the data analytics to provide organisations better knowledge on the employee insights and perception towards the job and the company. Similarly, Humu is an AI-based tool that helps individuals to keep track of their performance and provides support for consistent self-evaluation (Buck & Morrow, 2018; Sahlin & Angelis, 2019). Another tool that is in the experimentation and testing phase is X.ai, which if successfully clears the testing phase could be deployed in the interview scheduling and calendar management of the HR executives. Industry offers the HR function a wide variety of conversational AI-based tools that handle tasks at different levels of HR functions. These tools include ‘Skillate’ a decision-making application, ‘Marvin’ a virtual recruiting consultant, ‘Harver’ a tool designed for mass hiring, ‘Loxo’ an application tracking system and several other tools in the development and testing phase.

Conclusion

AI is emerging rapidly beyond the conventional technology and has already been into the lives of individuals and organisations (Manning, 2020). However, its role as an intelligent application as it is meant to be is still in the nascent stage in several industries. The progress in the research and implementation of AI in the HRM in the next few years shall certainly see its prominence in the decision-making roles of HRs. Challenges in the development and the implementation of AI tools are persistent like in any other technology implemented by organisations. In spite of such challenges, AI is set on a course to intertwine into HR operations consistently.

Managerial Implications

The technological progression leads to the disruption of the conventional mode of functioning in the business. Considering the increasing pace of technological advancement organisations are required to make wise decisions in adapting the trends in technology with foresight (Jeude, 2020). Discussion from this study could help the practitioners, scholars in comprehending the latest developments in the AI with reference to the activities of HRM. Information related to the AI tools discussed in the literature helps the practicing HR managers in adapting the AI tools for HR planning workforce design, recruiting, training and development and performance appraisal. Most of the tools cited in this study are meant for the implementation of the process management in HR functions. Thus, the strategic leadership in the HR function could take insightful information related to the AI tools and evaluate their need for implementation for the organisational change towards the digitalisation (Quaquebeke & Gerpott, 2023). Industry practitioners, academicians and the research scholars could consider the inputs from the study to comprehend the availability of the AI tools and the applications that are in the nascent stage of development. This study had included detailed discussion on AI in HRM in spite of the dearth in scholarly data.

Limitations and Further Scope

While most of the content related to the existent status of the AI in the HRM function is discussed in this study there are few limitations that leave scope for the further research in this area. This study did not engage any empirical analysis as that methodology is not relevant to the identified objective. Researchers in future could conduct the empirical analysis to understand the behavioural impact of AI tools on the users. For instance, ease of performance evaluation, impact of AI hiring experience on the candidates, and impact of AI on employee productivity are few studies that could measure the impact of AI application on the behaviour of its stakeholders. This study had examined the status quo of the AI tools that are existent or in the development stage. However, future studies could evaluate the efficiency of these tools by studying their technical abilities. Comparative studies related to the AI tool belonging to similar domains can be done for further investigation. Also, the impact of AI tools on the HRM function could be studied by comparing it to the organisations that use conventional methods.

Acknowledgement

The authors are grateful to the anonymous referees of the journal for their extremely useful suggestions to improve the quality of the paper. Usual disclaimers apply.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Mohsin Khan  https://orcid.org/0000-0001-7420-2455

Vijay Kumar Reddy  https://orcid.org/0000-0001-7422-0151