This study aims to help MSMEs in understanding the barriers to adoption of AI and how they are interlinked with each other. There is a general consensus amongst researchers and corporates that artificial intelligence (AI) can help in better and faster decision-making. However, many medium, small and micro enterprises (MSMEs) have not adopted AI in the context of automation.
The authors have assigned weights to the barriers of AI implementation by using Best Worst Method (BWM). In this study, the authors argue that the more alternatives there are, the greater the chances of making an error in judgment. This is a drawback of the traditional comparison matrices that are used for assigning weights to attributes/alternatives/factors. Subsequently, we have computed the influence of the barriers on each other using the DEMATEL method. This helped us understand what the drivers of the bottlenecks are and the interlinkages between them.
The barriers identified are top management support, change resistance from employees, lack of knowledge about AI in automation, Lack of resources to implement the solution, uncertainty regarding the future of benefits, perception of job loss and subsequent conflicts and lack of perseverance. Price of automation, uncertainty regarding future and top management support were identified as factors that drive other bottlenecks.
The study has theoretical implications, as it corroborates the technology organization environment (TOE) framework in the context of AI automation in MSMEs. This is the first study of this type, to the best of the authors’ knowledge, on MSMEs in India. Factors such as lack of top management support, change resistance, perseverance and price of the automation can be viewed as organizational blocks to AI implementation. Lack of knowledge about AI in automation and Lack of resources are technology-related barriers, whereas uncertainty regarding the future of technology and job loss are related to the environment. This has implications for the top management of MSMEs. The results did show that some bottlenecks act as drivers of other bottlenecks. Thus, the MSMEs, which inherently have limited resources, can prioritize their focus on the ones that have high driving power.
The authors have used the TOE framework to refer to the technological, organizational and environmental factors that affect the decisions to adopt a new technology. The decision of the business and government, to participate in the diffusion of AI, may be contingent upon monitoring the bottlenecks and designing mechanisms to enhance use of AI for automation by MSMEs across India. Thus, this study has wide policy implications.
