Taisei Corporation shapes the next generation of talent with ChatGPT

Taisei Corporation shapes the next generation of talent with ChatGPT

Taisei Corporation’s HR team is leading the rollout of ChatGPT Enterprise to drive AI-powered talent development across the organization.

Founded in 1917, Taisei Corporation is one of Japan’s leading construction companies. For more than a century, it has delivered projects in Japan and around the world, helping to build the social infrastructure that supports modern life.

Recently, a new question has come into focus: What should Taisei build next? The company began to ask whether its most important investment should be not only in buildings and infrastructure, but in people.

With this in mind, Taisei’s HR organization decided to introduce ChatGPT Enterprise as a cornerstone of its talent development strategy. Rather than treating generative AI as a simple productivity tool, Taisei positioned it as a technology for expanding human potential. Employees now use AI to push past their own perceived limits, reshape how they work, and unlock meaningful efficiency gains. Since rolling out ChatGPT Enterprise, Taisei has seen:

• 90% weekly active usage of ChatGPT Enterprise
• More than 5.5 hours of work saved per employee each week
• 3,300 custom GPTs created

While many companies look to AI mainly to improve efficiency, Taisei started from a different goal. The introduction of ChatGPT Enterprise was led by the HR organization with a clear focus on people development.

At the center of this decision was a belief that AI is more than a tool. It is something that can extend human potential and reshape how people work. Yasuo Tanaka, Chief Manager of Human Resources Training Center, explains, “Technology eventually becomes available to everyone. That is why we believed our real advantage would come from the people who can use it well, and from the culture that supports them.”

People development has long been a core management priority at Taisei, and ChatGPT Enterprise fits directly into that strategy. The aim was not simply to increase the number of AI users, but to create an environment where employees learn with AI and grow independently over time.

Reflecting on why Taisei chose ChatGPT Enterprise, Tanaka says, “OpenAI Japan moved quickly and made an effort to understand our culture and our challenges. They stood beside us in our aim to develop people who can truly make AI their own.”

For the company wide rollout of ChatGPT Enterprise, Taisei adopted an approach it calls “Middle Out”, combining direction from leadership with energy from the front lines. This concept is Taisei’s own, created to describe how change should be driven from the center of the organization. It was proposed by Kenta Nukazawa, Assistant Manager HR Development / GenAI Project Lead, drawing on his experience working closely with both on site teams and HR systems.

Nukazawa had long felt that neither top nor bottom alone was enough. He explains, “Decisions made at the top do not automatically move people on the ground. At the same time, passion at the front line alone rarely gains full support from management and does not spread across the whole organization. You need someone in the middle who can stand between the two and translate both sides.”

From his point of view, the hardest part of bringing AI into an organization is not the technology, but aligning how people feel about it. Looking back, Nukazawa recalls, “Some people see AI as something that looks useful, while others feel anxious about it. If you ignore those different reactions, AI will never truly take root in the organization.”

The idea of “Middle Out” grew out of this concern and was quickly shared inside HR. Tanaka and other leaders then adopted it as a formal model for driving change across the company.

Based on this model, Taisei designed a wide range of initiatives, including training programs, internal events, communities of practice, and hackathons. AI began to spread not as “a tool handed down by the company”, but as “something that supports my own work” for employees on the ground.

Tanaka reflects on the outcome, “Because we had the idea of Middle Out, messages from the top reached the front line, and ideas from the front line spread across the entire organization.”

As AI started to appear in more day to day work, a new challenge emerged at Taisei: how to handle information scattered across the company in a way that was both useful and safe.

Taisei saw strong potential in ChatGPT apps as a way to raise productivity. The ability to draw on information spread across services such as Microsoft 365 and Box in a single interface promised real value for people on the ground. At the same time, concerns surfaced around topics such as information control and the risk of incorrect answers.

Looking back, Nukazawa says that his first priority was not convenience. “I kept asking myself whether people would really feel safe using this. That question mattered more to me than how useful the system might look on paper.”

Working closely with the Information Planning Department, Nukazawa helped design a governance framework step by step. The team introduced access controls, organized usage logs, built an education program, and set up regular monitoring.

Through this steady effort, ChatGPT apps have taken root across the organization and have become tools that many employees rely on in their everyday work.

What mattered most, however, was not the restrictions themselves. Nukazawa explains that the goal was to create understanding, not just rules. “We were never trying to tie people down with policies. We focused on explaining why each restriction was in place and on creating an environment where everyone could experiment with confidence.”

By combining department level controls on what information could be accessed with ongoing education, Taisei created an environment that balances safety with active use. Employees can now rely on AI in their daily work, knowing that the right safeguards are in place.

Since introducing ChatGPT Enterprise, Taisei has seen clear, measurable impact:

• 90% weekly active usage rate
• More than 5.5 hours of work saved per employee each week (equivalent to roughly 260 hours per person per year)
• 3,300 custom GPTs created
• 3,800 projects created
• 99% of users want to continue using ChatGPT Enterprise

The more meaningful change is happening on the ground.

Younger employees are using AI to deliver work on par with their more experienced colleagues, and veteran employees are beginning to recognize the value of AI in developing the next generation. It is now common to hear comments such as, “Without ChatGPT, my work just does not move forward.”

Working with AI is steadily becoming a normal part of life at Taisei, not a special exception. Reflecting on this shift, Yasuo Tanaka shares, “We were able to reach this point because we kept taking on new challenges without giving up, and because so many employees chose to support those efforts and turn them into something much bigger.”

At the center of this journey is a human-centered model for using AI.

“Introducing AI is not about promising a more convenient future, but about creating an environment where people can think with confidence and continue to engage with AI over time.”

At the core of Taisei’s vision is a future where every employee is comfortable using AI, talent grows directly from work on the ground, and those changes are embraced as part of the company’s culture. In this future, AI is not just a tool but a partner that quietly supports daily work and helps create new value.

Taisei now wants to take this human centered approach to AI beyond its own organization. The company aims to share the model it has built with OpenAI across the construction industry and, eventually, with society at large. Making this way of working available to others is the next chapter in Taisei’s mission.