[Edaily Reporter Kim Hyun-ah ] Major South Korean companies are staking their very survival on the race for artificial intelligence (AI). They are spending tens of billions to hundreds of billions of won to secure NVIDIA GPUs and are accelerating the adoption of generative AI and Agent AI. On the surface, it appears they are keeping pace in the global AI competition.

However, the reality as seen by technology leaders on the ground is quite different. Chief Technology Officers (CTOs) and AI executives from various companies I recently met with cited organizational change and innovation in work methods—rather than AI technology itself—as the greater challenges. They noted that while they have secured cutting-edge AI at enormous cost, the organizational structures and work methods necessary for AI to deliver proper results remain stuck in the past.

AI Has Been Adopted, but Work Practices Remain the Same

Many companies are declaring an “AI Transformation” (AX). However, a closer look at the front lines reveals that the scale of change is not as significant as expected.

In many cases, efforts are limited to identifying “tasks that can be replaced by AI” as directed by management or automating simple, repetitive tasks that were previously handled in Excel. Even while pushing forward with data integration projects, data silos between departments remain intact, and although AI has been introduced, decision-making methods and work processes have not changed significantly from the past.

In essence, companies are merely adding AI to existing workflows rather than utilizing it in new ways.

A company’s competitiveness depends not on which cutting-edge AI model it chooses, but on how prepared the organization is to adapt flexibly so that AI can transform existing work practices.

The AI Race Is a Race for “Speed”

In the AI industry, the market landscape can shift in a matter of months. What global AI companies prioritize is not perfect consensus among all members, but the ability to make quick decisions and take action at critical moments.

After recognizing OpenAI’s potential, Microsoft strategically reallocated the computing resources needed for the AI race and focused on supporting OpenAI. The company determined that securing a market lead was more important than internal organizational interests.

Google drives innovation by pitting multiple teams with the same goals against each other and consolidating or terminating underperforming projects. OpenAI is also well known for its culture of flexibly bringing researchers together across departmental and hierarchical boundaries to drive projects forward.

Although their approaches differ, the common thread is clear: making quick decisions at critical moments and concentrating resources where they are needed most.

The AI transformation is not merely an information technology (IT) project. It is a management strategy that involves redesigning business structures and resource allocation. Ultimately, top management must set the direction and lead the organization through this transformation.

The fact that Naver (NAVER(035420)) has been strengthening founder-led leadership since last year and increasing its strategic focus on AI and future businesses is another example that demonstrates the importance of leadership in setting the direction for the AI era.

In contrast, many domestic companies still require multiple levels of reporting and coordination even for minor decisions. While caution is a virtue in corporate management, in the AI industry, excessive caution can cause companies to miss opportunities to secure a first-mover advantage.

The statement by Koo Hyun-mo, former CEO of KTCorporation(030200) —“I don’t write books on AI because things change so quickly that they become outdated in no time”—symbolically illustrates the pace of change in the AI era.

Talent in the AI Era: “The Questioner”

The limitations of organizational culture are also evident in the talent acquisition process.

Major Korean corporations are actively recruiting researchers from global AI companies. However, there are quite a few cases where key talent, who were recruited with great effort, fail to settle into the organization and end up leaving.

The causes cited include existing organizational practices, a closed approach to information sharing, and an evaluation culture that does not tolerate failure. This is because, in some cases, the organization actually demands that experts—recruited with the expectation of innovation—“learn our company’s way first.”

However, the talent needed in the AI era is not someone who simply follows the organization’s existing ways. It is someone who can redefine problems, question existing processes, and, when necessary, present alternative views even to senior management.

In an era where AI generates answers, Saramin who defines the right questions to ask holds a greater competitive advantage than Saramin who simply finds the right answers quickly.

In organizations that are uncomfortable with questions, the most valuable talent is ultimately the first to leave.

You can buy AI, but you cannot buy organizational culture

. In the future, AI will take on a significant portion of corporate activities, moving beyond repetitive tasks to include data analysis, forecasting, and decision-making support.

Therefore, the role left for humans is not to quickly calculate the correct answer, but to ask the right questions and propose new directions. The same applies to companies.

Future competitiveness is not determined by how many additional GPUs a company acquires. A corporate culture that embraces diverse opinions, the flexibility to leverage external talent as catalysts for change, and leadership that acts swiftly while accepting the risk of failure—these are the true sources of competitiveness in the AI era.

It is not only perfectly controlled organizations that drive innovation. A culture that allows for a certain level of autonomy and experimentation is equally important. Just like SK’s open organizational culture, as mentioned by Yoo Young-sang ( SK(034730) ), Chair of the AI Committee, an organization’s autonomy and flexibility can serve as the driving force behind new initiatives.

AI systems can be bought with money. However, an organizational culture that transforms AI-generated ideas into innovation cannot be bought with money.

In the AI era, the most vulnerable companies are not those lacking AI technology. They are the ones unable to accept the uncomfortable questions raised by AI or the differing opinions of their employees.

The outcome of the AI competition will likely be determined not by the number of GPUs, but by an organization’s ability to listen to uncomfortable questions until the end and translate them into action.