Only 11% of companies get meaningful financial impact from their investments in AI (BCG, 2021). These 11% have something in common, and it’s not about fancy algorithms or complex technology. It’s about how they leverage the human factor and get people to work in partnership with AI. This was also supported by the top management interviewed by our research.
Fueled by access to an unprecedented amount of data and ever-increasing computing power, artificial intelligence (AI) is reshaping the world and the way we live. Fully 84% of global businesses believe that AI will give them a competitive advantage (Statista 2018), and 72% believe that AI will be the most significant business advantage of the future (PwC 2017). But how for ist the implementation in practice and what are the limiting factors. To find out more about this we talked to top managers.
AI in practice: What Do Managers Say?
In order to gain practical insights into the future of AI from the highest levels of management, we interviewed five top managers regarding their understanding of the current status of AI and its potential in their industries, as well as the managerial challenges leaders are facing. These managers came from five different industries: (1) food, (2) commodities, (3) technology, (4) finance, and (5) healthcare. Importantly, all interview partners a) are responsible for the highest managerial decisions in their firms and b) operate in different industries.
From the interviews we gathered substantial evidence that the use of AI in practice, especially autonomous AI, is still in its infancy. Three managers stated that they use AI but that they perceive their companies’ implementation of AI to be in its early stages, whereas two managers said that their companies do not yet use AI. The interviews thus showed that the state of AI varies across industries. All of the managers whose firms have adopted AI stated that AI does not take over decision-making processes in their organizations but instead acts as an advisor for certain decisions, suggesting that autonomous AI is not in use. All five managers stated that they see potential for AI in their industry.
Technology has far outpaced practical implementation. But why?
When asked about the managerial view of the main implementation problems associated with AI, the answers of the managers whose firms already use AI differed markedly from those whose firms do not. The latter group of managers named the availability of data and liability issues as the primary managerial concerns, whereas all three of the managers whose firms have adopted AI identified the human factor as a primary concern.
When the managers of the two firms that did not implement AI yet were explicitly asked if the human factor is a managerial issue, both managers stated that because they had not yet implemented AI, they had no experience. In contrast, among the AI adopters, the most commonly stated issue was users’, investors’, and employees’ behavior in regard to AI, such as the uncertainty that comes with the implementation. This finding accords with what research increasingly suggests: the main challenge to AI adoption is not technology but cultural obstacles, processes and people (Borges et al. 2021; Duan et al. 2019; Gursoy et al. 2019; Raisch and Krakowski 2020). This claim was strongly supported by our interview partners.
The words of all the managers interviewed indicated that the rosy view of certain studies regarding the benefits of implementing AI at all levels of a firm are questionable and that the reality is far more challenging than a simple performance analysis. In none of the interviews were the capabilities of AI questioned. The trend towards AI-driven decisions was clear, and AI’s capabilities were seen as indisputable. However, in the interviews the managers’ need to understand how and when users’ trust can be gained and how systems must be designed to work in collaboration with the users became apparent.
The challenge to AI implementation is not technology. It’s about how you get your people and AI to collaborate.
Among the top management, the human factor was frequently mentioned to be among the biggest managerial challengesin AI implementation. Human factors are important and under-recognised considerations in the design and development of AI and the reason why only 11% of all companies achieve substantial financial benefits from implementing AI (BCG, 2021). In view of this finding, it is important to note that AI is currently most commonly studied by computer scientists, roboticists, and engineers. While these are experts in technology and aim to maximize algorithmic performance, they often disregard the role of the humans involved and the wider organizational implications of AI adoption. The fact is, however, that human and AI agents do not inhabit separate worlds but interact and are interdependent.
The human factor can actually be leveraged to competitive advantage if executives are able to design systems to work in partnership with people. This is a complex and challenging task that requires human factors specialists to design and develop user-centric systems.