"What's the Point of Adoption?"... SMEs Identify Obstacles to AX

AX Success Equation 2026 Conference Held
81.4% Cite "Issues with Data Quality and Quantity"
Cost Prediction Failures Due to Underestimating Labor Costs
"Data Governance and Practical Training Needed"

"What's the point of just putting it there? The utilization rate is low."

On the 25th, at the opening session of the 'Artificial Intelligence Transformation (AX) Success Equation 2025' conference held at POSCO Tower Yeoksam in Gangnam-gu, Seoul, Hwang Sungjae, CEO of XYZ, is giving a presentation. Venture Business Association

At the 'Artificial Intelligence Transformation (AX) Success Equation 2026' conference held by the Venture Business Association at POSCO Tower Yeoksam in Gangnam-gu, Seoul, on the 25th, the CEO of a small and medium-sized enterprise expressed this concern. The CEO said, "We are using artificial intelligence (AI) to analyze defective products at an actual manufacturing site, but after nearly a year of collecting data, it feels like it will take forever to achieve artificial intelligence transformation (AX) at this pace," adding, "Although we have introduced AI, the actual utilization rate seems to be low."

Small business owners and SMEs in Korea commonly cited 'data collection' as an obstacle to AX. The event, hosted by the AX Bridge Committee under the Venture Business Association, gathered small business owners and SME CEOs aiming for AX, as well as industry experts such as Sanghoon Oh, CEO of Luxrobo, and Cheolmin Pyo, CEO of AI3, to analyze why AI is struggling to take root in the field and to seek solutions.

During the opening session, Taebyeong Jeon, CEO of Manna CEA, said, "Due to the nature of agriculture, it takes three to four months for a single crop cycle from growth to harvest, so collecting data takes a long time," emphasizing, "Even if the government implements AI distribution projects, there is an absolute lack of usable data, so we need ways to increase it." Seongjae Hwang, CEO of XYZ, stated, "The quality of data determines the success or failure of AI transformation."

Joo-Wan Lee, Chairman of the AX Bridge Committee, is delivering a welcome speech at the "Artificial Intelligence Transformation (AX) Success Equation 2025" conference held on the 25th at POSCO Tower Yeoksam, Gangnam-gu, Seoul. Venture Business Association

According to a survey of 221 domestic venture companies conducted and announced by the Venture Business Association on the same day, the 'five major barriers' to AX experienced by venture companies were identified as lack of data (81.4%), failure to predict costs (73.3%), incorrect problem definition (73.3%), solution mismatch (68.8%), and the PoC trap (64.3%). Sungbae Kong, Vice President of Megazone Cloud, analyzed, "Companies are facing difficulties in terms of data quality, quantity, and accessibility," and added, "A fundamental reason is the lack of a culture that recognizes data as an asset and the absence of data governance to refine data quality."

Vice President Kong also pointed out that perceiving AI adoption as a 'one-time project' and underestimating the additional costs required for ongoing maintenance and management is problematic. He further noted tendencies such as applying external solutions without sufficient consideration of internal circumstances, and failing to move from the initial adoption and proof-of-concept stages to actual expansion due to unclear expected outcomes. Kong explained, "63.8% of surveyed companies remained at the initial adoption or proof-of-concept stage," and "Among them, 85% were small and medium-sized venture companies with fewer than 50 employees, indicating a structure where pilot-based experiments are repeatedly conducted."

Experts summarized the field's challenges and proposed 'three key solutions for realizing AX.' First, they emphasized that the government should strengthen support for data governance and quality management so that company-owned data can be refined, processed, and labeled for real business operations through data policies. In addition, they explained that talent development should shift from focusing solely on training advanced AI developers to practical education for venture CEOs and employees to learn what and how to solve with AI. They also stated that the government should provide continuous support across all stages-PoC, adoption, operation, and expansion-and establish a packaging support system from a total cost of ownership (TCO) perspective.

Joo-Wan Lee, Chairman of the AX Bridge Committee, said, "It is only possible for Korea to become one of the top three AI powers when there are many companies that can properly use AI in the field," adding, "In 2026, the core of government AI policy must shift from 'technology supply' to 'bridging the execution gap.'"