How should universities teach leadership now that teams include humans and autonomous AI agents?

Unlike teaching students how to use artificial intelligence (AI) and other technologies, which often feels like fighting a losing battle to keep up with constant change, teaching leadership used to be far more stable. The popularity and effectiveness of certain leadership styles undergo gradual shifts, but nothing too dramatic. The past 20 years, for example, have seen a transition from more authoritarian styles towards people-focused and motivating leadership, such as transformational leadership. 

As an academic, it was comforting to know I just had to tweak my slides every year rather than change whole lectures. There were no alarms and no surprises when teaching leadership. 

Generative AI has shattered that calm. Within a short time, GenAI has gone from being a tool in the workplace to being a teammate. The modern leader no longer leads only humans, as they have done for centuries, but also autonomous AI agents. An autonomous AI agent, which uses GenAI and other information systems, can operate independently (as the name suggests) as part of a team, communicating with other team members and “thinking” for itself. For example, the autonomous AI agent can participate in a meeting and receive verbal instructions for complex tasks. As educators, our comfort zone of reciting stories of great leaders from history has ceased to be sufficient. How would Napoleon or Churchill have managed autonomous AI agents? We don’t know. 

• Solving the soft skills crisis using artificial intelligence
• Faced with GenAI, educators’ engagement capacity matters more than ever
• How to empower your university to integrate generative AI using tools and talent you already have

So, how should university teachers prepare a new generation of modern leaders to approach these mixed teams? Teaching leadership styles that are effective at motivating people is no longer enough. In addition, students must now learn how to build their team’s trust in AI, then they will need to know how to combine leadership styles in a way that gets the most out of both humans and AI.

Build trust with a clear vision of what the role of AI is

First, leaders need to become ambidextrous, as comfortable leading humans as they are leading on which of various technologies are used. Traditionally, some leaders have been focused on day-to-day operational and tactical decisions, while others focused on more strategic decisions. To lead autonomous AI agents, the leader must build trust in them among the team. The team members must have confidence in the autonomous AI, meaning they must believe in its ability. The leader must be clear on its use and build a consensus around this. The team must be put on a sustainable trajectory for change. Failing to identify the right role for AI will keep the team stuck in uncertainty. 

Leadership students must learn the different forms of AI, their strengths and weaknesses. They must also learn the typical concerns that various stakeholders have and how to take specific steps to build trust. These topics should be discussed in class first so students appreciate the context in which they will lead. Students can practise analysing a case study and developing a vision for the role of AI and how to build trust.

Choose the right combination of leadership styles

While more time must be spent on understanding emerging technology, the established approaches to leadership still need to be covered in class. 

The most popular leadership styles taught at university today are servant, transactional and transformational. All three are effective in motivating a team. The servant style focuses on providing support; transactional focuses on an enticing reward for the work done; and transformational creates a shared vision of the future. These popular leadership styles are still the ones to focus on when leading mixed teams of humans and autonomous agents, but the leader needs to think about how to combine them to get the best result (see Table 1). 

Transactional and transformational leaderships’ combined impact on AI and trust

Given the volatile times we live in, a leader may find themselves in a situation where they know how they will use AI, but they are not entirely clear on the goals and journey. In a teaching context, students can be given scenarios where they must lead a team, including autonomous AI agents, to achieve goals. They can then analyse the situations and decide what leadership styles to apply and how to build trust in their human team members. Educators can illustrate this decision-making process using a table (see above).

They may need to combine transactional leadership with transformational leadership, for example. Transactional leadership focuses on planning, communicating tasks clearly and an exchange of value. This works well with both humans and automated AI agents. 

Transformational leadership prioritises creating a shared vision to change something, inspiring and motivating people to go beyond their narrow personal interests. As AI and automation replace some human interaction, class discussion could cover the emotional void and isolation left over and how, in the age of AI, a leader needs deep, emotionally engaged relationships powered by servant or transformational leadership. These strong emotional bonds can complement the well-crafted exchange of value arranged with transactional leadership. 

Some clarity on the future of teaching leadership

An effective course on leadership today must convey the importance of leading on technology as well as people, building trust in the technology, and finding the best combination of leadership styles to get the most out of humans and automated AI agents.

Alex Zarifis is a lecturer in information systems at the University of Southampton, UK. His latest book is Leadership with AI and Trust: Adapting Popular Leadership Styles for AI (De Gruyter, 2025).

If you would like advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the Campus newsletter.