Korea Microsoft is focusing on the rapid adoption of artificial intelligence (AI) in the country, proposing a strategy to leverage accumulated data and knowledge within companies to enhance operations. The company aims to expand AI's role from merely assisting human judgment to coordinating and executing various tasks, emphasizing the need to verify that this leads to tangible business outcomes such as cost reduction and time savings.
On September 30, Korea Microsoft held the 'Microsoft Industry Summit' at COEX in Seoul, featuring over 50 sessions and more than 30 customer and partner case studies across five industries: finance, retail, manufacturing, healthcare, and education.
Korea Microsoft highlighted the rapid increase in generative AI usage in South Korea. According to a report from the MS AI Economy Institute, as of June, 40.6% of South Koreans were using generative AI, ranking the country 12th globally. This figure represents a 3.5 percentage point increase from the first quarter, the largest rise among surveyed nations.
The summit focused on how to apply AI to real business operations and connect it to measurable outcomes. Korea Microsoft presented a process for AI adoption, organizational integration, operation, and performance generation through industry-specific examples.
Choi Kang-rim, head of manufacturing and mobility at Korea Microsoft, noted, "The South Korean manufacturing sector is at a critical juncture for change." He pointed out that while manufacturing contributes over 27% of the country's GDP, the industry faces challenges such as a shortage of skilled labor, making it difficult to retain accumulated experience and expertise. Maintaining competitiveness solely through cost reduction is becoming increasingly challenging.
Additionally, with growing uncertainties in supply chains and markets, companies must effectively utilize their knowledge and respond quickly to changes while maintaining production and operations in unexpected situations.
Choi emphasized that the questions surrounding AI have shifted from when and how much to implement it to how quickly and accurately it can be executed through AI.
He further explained that the use of AI in manufacturing evolves from connecting dispersed data and knowledge within companies to optimizing tasks with AI assistance. Ultimately, this leads to the development of 'autonomous enterprises' where multiple AI agents coordinate tasks and AI autonomously performs work within defined parameters.
The summit also introduced a 'Physical AI' strategy, which applies AI to manufacturing equipment and robots. Aaron Schneider, vice president of Microsoft Physical AI Engineering, explained how data collected from cameras and sensors is used in digital twins and simulations to train and validate AI, which is then applied to actual equipment. Microsoft plans to connect AI development, simulation, and on-site deployment through its 'Physical AI Toolchain.'
During the customer case presentations, examples of AI application in real business settings were shared. Amorepacific has built a 'data highway' that connects and standardizes over 70 years of internal research data, including raw materials, formulations, and test reports. Based on this, the company operates an AI service called 'Lemon,' allowing researchers to search for and utilize necessary information in natural language. This has reduced the time required for a team of three to compare formulations from a week to just one day, and data exploration time from over half a day to under ten minutes.
Lee Seung-hyun, CEO of AI solution startup Enhance, emphasized the importance of verifying whether AI truly delivers 'token value.' He stated that businesses should assess the actual value created by AI after excluding operational and management costs from increased revenue and reduced expenses.
Lee also shared a case where AI was applied to the cost review process of construction projects under the Ministry of National Defense, explaining its practical application. He concluded by stressing that in manufacturing and procurement, the focus should not solely be on achieving automation rates but rather on how much purchasing costs have been reduced and how supply stability and operational speed have improved, using actual business outcomes as the basis for evaluating AI's value.
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.