Your startup has an AI strategy. Does it have a human strategy?

At a recent conference in Saudi Arabia, the discussion surrounding AI in business focused heavily on how the technology can accelerate tasks such as research, drafting, analysis, and team productivity. While these efficiency gains are evident, the story raises a critical question: what becomes of the workload once tasks become faster?

If a report that once took an hour now takes only 20 minutes, does that translate to 40 minutes of extra capacity, or does it merely allow for two additional reports to be completed? This distinction may seem minor, yet psychologically it is significant. If every efficiency gain is immediately met with additional output, the working day does not become easier; it becomes denser.

Startups, already known for their fast-paced culture, must pay particular attention to this issue. Teams often wear many hats, priorities shift rapidly, and employees are frequently expected to take on new responsibilities as the business grows. AI can streamline this environment, yet it can also blur the lines of what a person's role entails.

A person may maintain the same title while their responsibilities are increasingly managed, summarized, or analysed by AI, with turnaround and volume expectations changing almost overnight. While AI adoption can boost confidence and empower individuals to take on broader responsibilities, it can also lead to role ambiguity. Research from China involving 541 employees of technology firms found that while greater use of generative AI increased employees' confidence in taking on more diverse duties, it simultaneously created role confusion.

This blend of heightened capability and unclear role boundaries is a factor worth considering. When AI enables faster production and broader responsibilities, the risk is that people may experience cognitive overload as they attempt to navigate the evolving nature of their work. This is especially pertinent in startups, where roles are often broad and formal job descriptions may not fully capture all the tasks involved.

Additionally, the issue of control is crucial. Studies have shown that partial AI assistance can support autonomy, competence, and meaningfulness, whereas more comprehensive automation can erode these experiences over time. The manner in which AI is introduced can significantly impact whether people feel they are exercising judgement or merely supervising output.

Over time, even useful technologies can affect how much ownership individuals feel over their work. Therefore, treating productivity alone as the measure of successful AI adoption is insufficient. If a team is producing more, but the workday has become compressed, responsibilities are less clear, and meaningful judgment is diminished, the benefits of efficiency are only part of the story.

Instead of focusing solely on speed, founders must consider how the time saved is being utilized. Is it creating genuine capacity, fostering better thinking, providing recovery periods, or allowing space for tasks that require human judgment? Alternatively, is the organization simply increasing the volume expected from individuals without altering the nature of the work?

Looking ahead, the question extends beyond immediate efficiency gains to the long-term development of expertise. As AI increasingly handles early-stage thinking in roles, organizations must consider how they will cultivate the judgment required for more advanced responsibilities. Expertise typically emerges through repeated exposure to problems, mistakes, uncertainty, and decision-making.

Removing too much of this developmental work may yield short-term efficiencies but may introduce new challenges in the future. While the focus on AI's future capabilities is understandably intense, founders need to reflect on the practical implications of using AI in their organizations. Namely, when AI makes work faster, what are you choosing to do with the time it saves?