[Everyone's AI, Divided AX]③Focus on Lowering Barriers to AI Entry..."Support Remains Unfelt by Many"

From One-Time Infrastructure Support to Long-Term Operations Management
Diversifying the Focus of AX Assistance

Reducing Trial-and-Error with AX Cooperatives
Establishing Industry-Specific Data Standards
Expanding Regional Shared AX

Editor's NoteArtificial intelligence (AI) is transforming corporate productivity and ways of working, but there is now a growing gap between large conglomerates and small- and mid-sized enterprises (SMEs) in how quickly they are adopting these technologies. While companies with dedicated AI teams and infrastructure are accelerating company-wide AI transformation (AX), other businesses and institutions without such resources are being left behind, excluded from 'AI for all.' The Asia Business Daily looks into the AX divide hidden beneath innovation and explores the conditions and solutions needed to boost productivity through inclusive AI where no one is left out.

The government has introduced programs to support AI transformation (AX) in corporations by helping them build AI, data, and cloud infrastructure. Meanwhile, large companies are also extending support to their partners. However, as each company finds itself at a different stage of readiness and has distinct needs, there are often disconnects between the support they receive and tangible outcomes. This is why there are increasing calls to shift the focus of support from simply spreading AI adoption to providing ongoing assistance in practical utilization to eliminate AX blind spots.

According to the Ministry of Science and ICT on August 13, the government is providing corporations with essential technology resources for AX—such as AI development, data infrastructure, and cloud services—through a 62.5 billion won voucher program this year. Following AI vouchers (25.2 billion won), data vouchers (7.2 billion won),and cloud vouchers (4.1 billion won), the government has added an AX One-Stop Voucher (26.0 billion won) that integrates support across different fields, reorganizing the structure to now offer four types of vouchers.

Large corporations are also supporting their partners' AX journeys. According to a survey conducted by Korea Enterprises Federation through the polling agency Mono Research on the top 800 manufacturers by sales, 45% of responding companies said they support their partners' AX efforts. Among various activities to assist partners in AX, the most common was integration and standardization of production, quality, equipment, and logistics data (34.4%), followed by joint pilot/demonstration projects (23.8%), AX capability assessments and roadmap consulting (16.4%), and education/professional workforce dispatch (14.3%). 

However, many point out that current support is primarily focused on lowering the initial AI adoption barriers for businesses, falling short when it comes to operating and upgrading systems in accordance with each company's circumstances and industry specifics. In fact, the Korea Enterprises Federation's survey found that the biggest roadblocks in partner AX cited by large companies were insufficient digital infrastructure and AI utilization capabilities (43.4%), as well as difficulties in data integration, standardization, and security management (22.7%). Given that ongoing investments in data management and maintenance are required even after AI is implemented, there are calls for supplementary measures to help reduce the burdens on individual companies.

Experts recommend enhancing existing policies with solutions such as building industry-sector AX cooperatives, establishing standardized data protocols, and expanding the pool of local, shared AX professionals.

Goo Inhyeok, professor at the School of Global Service at Sookmyung Women's University, suggested forming 'AX cooperatives' at the industry level to share AI capabilities that individual companies may struggle to develop on their own. Within the same sector, companies would jointly establish data standards, security systems, and shared expertise to validate AI solutions together. Professor Goo said, "Instead of repeatedly funding the setup costs for each company, investments should be directed toward establishing a collaborative foundation that enables companies to carry out AX together." He added, "By building a structure in which participating companies invest together and jointly own services, we can significantly reduce redundant costs and trial-and-error for individual companies."

A system supporting the entire AX process, not just one-off setup costs, is also needed. This will prevent AX initiatives from stalling after government support ends due to a lack of ongoing operational funds or management personnel—while also enabling AI models to be retrained or new data to be managed as work processes change. Professor Goo noted, "There must be policy mechanisms to support data management, maintenance, and model retraining for two to three years after initial AI adoption," adding, "Policy outcomes should also be evaluated not only by the number of companies introduced but also by actual utilization and investment levels by companies one or two years down the line."

Park Jinho, professor of Computer and AI at Dongguk University, stressed the need to develop an AX package market that allows companies to source the capabilities they need from external providers. "A continuous system for AI maintenance and education must be in place after adoption," Park said. "In the future, we expect to see a subscription-based market where companies can freely access the AX capabilities they require." He also emphasized the need for industry-specific data standards, as well as shared protocols for security and access rights suitable for AI training.

Jeong Seokchan, director of the Grand ICT Research Center for Artificial Intelligence at Dong-eui University, suggested supporting the expansion of on-site AX professionals who work closely with businesses in regional hubs to set AX goals, build data infrastructure, and apply AI. This would involve jointly leveraging expertise from local universities and research institutes.

This content was produced with the assistance of AI translation services.