Over the years, study upon study has shown that senior leaders are slower to adapt to new technology – email, the Internet and social media – than younger employees. That’s not necessarily so with AI, according to the University of Delaware’s Saleem Mistry.
Mistry, associate professor of management at UD's Alfred Lerner College of Business & Economics, recently conducted a survey of more than 200 university alumni, 75% of which had more than 16 years of professional experience. He found that senior leaders are actively adopting AI to solve their biggest challenges. However, they are doing so largely without structured support or guidance.
Here are four findings from Mistry's survey that shows how AI is actually being used at the top.
Senior Leaders Are Overwhelmingly Self-Taught
Mistry said his most glaring finding is the gap between high AI adoption among senior leaders and the near-total absence of formal corporate support. Although a majority use these tools, they are almost entirely self-taught, which highlights visible opportunity that organizations aren’t really steering the AI conversation for leaders:
• High usage. 62% of all senior leaders surveyed use AI tools "regularly" or "occasionally" in their work.
• Training gap. Of those users, an overwhelming 80% report their organization provides "Never" or only "Sometimes" (mostly never) adequate training.
Mistry said this shows that leaders from VP level down are using tools like ChatGPT and Copilot informally to keep up with heavy workloads, without any real organizational guidance. The stakes are high. In the survey, a vice president of legal was using AI for compliance tasks and a manager of three was using it for performance reviews, both with no formal training. “These are senior leaders handling sensitive work while essentially figuring it out on their own,” Mistry said.
There is a clear ladder of AI use
Leaders are not using AI randomly. There is a clear progression in how they use it, moving through three levels.
• Tier 1 (The Drafters)
This is the most common starting point. Leaders use AI to improve writing and communication. They draft emails, shape documents, and refine tone. As one Director of Product put it, it helps him "polish phrasing" and adjust tone and voice.
• Tier 2 (The Synthesizers)
At this stage, leaders use AI to manage information overload. They summarize meetings, condense documents, and pull together research so they can keep up with large volumes of input. As one leader managing a team of 200 said, "Being a leader requires attention in a variety of areas. AI helps me manage the vast amounts of information I need to consume."
• Tier 3 (The Architects)
Here, leaders move beyond writing and summarizing. They use AI to automate parts of their work. This includes building agents, creating custom GPTs, or designing tools that track work and performance. One leader managing 300 people said, "It will eliminate half or more of my overhead."
Managers and individual contributors use AI for different reasons
People managers and individual contributors (IC) are using AI for very different reasons based on their roles.
• For people managers, their main challenge is scale. They are overloaded with communication and administration, so they use AI to reduce noise and keep up. They lean heavily on summarization and tone adjustment tools.
• For project leads and ICs, their focus is output. They use AI to produce work faster, including drafting content, building decks, writing code, or generating ideas. This difference reflects their jobs. One group is trying to keep up, the other is trying to produce more. It also shows that AI is not a single-use tool. Its value depends on the problem it is being used to solve.
This difference reflects their jobs. One group is trying to keep up, the other is trying to produce more. It also shows that AI is not a single-use tool. Its value depends on the problem it is being used to solve.
Resistance to AI is mostly intentional
Among the 38 percent of leaders who do not use AI, resistance is usually not based on lack of awareness. It falls into three groups:
• The Ethical Objectors. Some avoid AI due to concerns about its broader impact.
• The Quality Skeptics. Some do not trust the output and feel it is not reliable enough for important work.
• The Blocked. Some are not allowed to use AI due to company policy.
Mistry concludes that there is a clear overall pattern: Leaders are using AI in practical ways, but mostly without structured support or guidance.
“If it feels like you are figuring this out as you go without much help from your organization, that is consistent with what most leaders are experiencing,” Mistry said.
To connect directly with Mistry and arrange an interview, visit his profile page and click on the "connect" button. Interested reporters can also email MediaRelations@udel.edu.
