Utilizing AI Technology from a Management Perspective

Nakamura Hiroki
5 min readDec 21, 2023


I often find myself wondering about the future of work. Particularly, the rapid development of AI technologies like Large Language Models (LLMs) seems to have a significant impact on how we will work in the future. I remember reading reports about this a while ago.

It discribed how AI technologies, such as ChatGPT, might reduce working hours. In the US, it’s predicted that within 10 years, 28% of workers will have a 4-day work week, and 71% will experience at least a 10% reduction in work hours. In the UK, similarly, 28% of workers are expected to move to a 4-day work week within the same timeframe, with 88% seeing at least a 10% reduction in working hours.

In my parents’ generation in Japan, working six days a week was the norm. It seems that the concept of having two days off per week only became widespread in Japan in the late 1980s. It’s been over 40 years since then, so irrespective of the spread of AI technology, I personally feel it’s about time for another day off to be added.

Recently, I wrote about how ChatGPT has changed my way of working.

Around me, I see people expanding their job scopes and reducing their working hours. This seems particularly true for those who enjoy breaking down tasks with a programming-like approach.

However, the changes we’re seeing still feel like they’re at an individual level. I believe that increasing individual productivity is not enough to significantly impact the performance of an organization or team. From a manager’s perspective, it’s important to understand how AI technology is changing individual work styles and how to integrate these changes to enhance team performance, how to design organizations accordingly is crucial. While it may be challenging to accurately predict future changes, I’m going to try to base my thoughts on my own experiences of trial and error in my role.

Become Smaller Teams

Using ChatGPT, I’ve strongly felt that an individual can now handle a broader range and volume of tasks. Currently, AI isn’t very helpful for tasks that one completely doesn’t know how to approach. However, for tasks that one can judge the quality of and knows how to do, but are time-consuming, it’s often feasible to utilize AI.

According to a survey by Google Japan, Bard is frequently used in Japan for seeking specialized advice, research, and issues related to programming.

When it comes to gathering information, if one can accurately judge its correctness, or in programming, if one can read and verify the code produced, LLMs greatly contribute to the efficiency of these tasks.

If an individual can handle a wider range of tasks, it naturally follows that smaller teams can achieve the same outcome. Intuitively, I feel that the team size could be at least half of what it used to be. If the number of people involved reduces to a fraction, the cost of communication dramatically decreases, leading to a more flexible team that can make decisions and act quickly.

Become More Dynamic

As specialization progresses, the boundaries of what can and cannot be done become clearer, both in terms of roles and skills. As a result, tasks that go beyond one’s range increasingly become impossible to handle, and tasks outside one’s domain are typically delegated to specialists in that area. However, I think that what is outside the domain for one person often does not require the full expertise of a specialist in that field.

For instance, conducting a simple data analysis doesn’t always require the full skill set of a data scientist, nor does creating a small operational tool always need a senior engineer’s skills. I frequently encounter such tasks that are beyond one’s capability but don’t require a professional’s full expertise.

As a PM, I often find myself having to ask for tasks to be done that don’t require specialization. Each time, I feel apologetic as these tasks generally lack rewards for professionals. They are necessary, yet often unfulfilling for the expert.

AI can fill this gap between “can’t do but necessary” and “can do but rewarding.” This prevents the unnecessary use of time by people with specialized skills. As a result, such expertise can be utilized more in truly necessary situations across various projects and organizations. Consequently, in organizations that utilize AI technology from a holistic optimization perspective, I believe the nature of teams will become more dynamic.

The Increasing Importance of Management’s Role

I believe that simply utilizing AI technology at an individual level will not naturally lead to smaller, more dynamic teams that are optimized as a whole. Naturally, to drive this change, the role of management becomes more crucial.

First, managers themselves need to adapt to the change. When understanding the scope of what each team member can do, it’s necessary to comprehend not only what they can do themselves but also the maximum they can achieve with AI. This requires an ongoing understanding of what can and cannot be done with the continually evolving AI technologies.

Furthermore, to achieve results with fewer people, managers themselves need to expand their scope of capabilities. There’s a view that managers should focus solely on management tasks and not engage in hands-on work. However, I believe that this perspective changes with the proliferation of generative AI. Asking someone to do something easily done with user-friendly AI feels akin to asking someone else to send an email for you.

In other words, not only the team members but also the managers themselves should utilize AI and assess each person’s abilities with the assumption of AI use. They need to understand where human expertise is truly necessary and ensure that the valuable time of highly skilled individuals is not wasted. I think this kind of management will be required.

Moreover, when considering beyond one’s own team and looking at the bigger perspective, whether a manager can hold a new perspective of overall optimization utilizing AI becomes very important in assigning managers.

At the End

In this post, I’ve considered how AI technologies like LLMs can be utilized from a management perspective. The key points are:

  1. Become Smaller Teams
  2. Become More Dynamic
  3. The Increasing Importance of Management’s Role

Using technological advancements for personal optimization is a matter of individual skill, but leveraging them from a team or organizational perspective depends on management. I don’t believe these changes are entirely discontinuous or entirely different from what we’ve seen before. However, considering the capabilities of technologies like ChatGPT, the pace of change feels much faster than before. Working in a startup, I’ve had to increase the volume and speed of work, whether I wanted to or not. As a result, the amount and scope of work I’m handling now is at least twice what it was a year ago, and the roles of each team member have significantly expanded.

The positive changes that have occurred in myself and my current company over this year were driven by external factors, making them necessary. Going forward, I want to drive change that is reproducible based on internal factors. I hope to contribute, even in a small way, towards achieving a 4-day work week, or perhaps even more.