Seoul: LG AI Research, the artificial intelligence lab under South Korea's LG Group, has announced its vision to expand the EXAONE AI ecosystem, encompassing efforts from foundational models to enterprise-level AI application services. During a media event in Seoul, Lim Yoo-hyung, co-head of LG AI Research, emphasized the lab's commitment to securing global competitiveness by applying their proprietary foundation model across various industrial sectors.
According to Yonhap News Agency, EXAONE, first introduced in 2021, is LG AI Research's multimodal AI model based on a large language model. As part of its strategy to enhance the EXAONE lineup, the lab has recently launched EXAONE 4.0 and a specialized pathology model, EXAONE Path 2.0, alongside agentic AI solutions designed for enterprise use.
LG AI Research also unveiled EXAONE 4.0 VL, a vision-language model capable of understanding both text and images. Additionally, three new enterprise-focused AI tools were introduced: ChatEXAONE, an employee-tailored agentic AI program; EXAONE Data Foundry, a platform for producing high-quality AI data; and EXAONE On-Premise, a full-stack solution for building secure, in-house agentic AI systems.
Co-chief Lee Hong-lak stated that the evolution of agentic AI would soon lead to physical AI, a technology poised to change the real world. He highlighted the lab's focus on accelerating the development of physical AI technology that enables AI to recognize, judge, and alter the actual environment.
Following the promotion of Lee and Lim to co-lead LG AI Research, they plan to concentrate on developing core AI technologies and applying them in real-world industries to create tangible value. Lee expressed the lab's dedication to advancing EXAONE to a world-class level and expanding it as a growth driver by specializing it as an industrial AI tool.
In its efforts to expand partnerships, the lab aims to launch a business intelligence service with the London Stock Exchange Group (LSEG) in the third quarter of this year. The service will leverage LSEG's financial data and unstructured sources, such as news articles and corporate disclosures, to predict investment asset returns and generate reports to aid global investors.
Lee Hwa-young, head of the AI business unit at LG AI Research, noted that the service offers data on mid- and small-cap companies, which are typically more challenging to track. While currently available for LSEG customers, the lab plans to make the service accessible via platforms such as Google or Yahoo in the future.