The development: version-specific releases.
LG AI Research released EXAONE 4.5 in April 2026 as an open-weight vision-language model. Its official repository identifies the EXAONE AI Model License Agreement 1.2–NC. Separately, the August 2026 K-EXAONE 2.0 technical report describes that release under Apache 2.0. These are different releases with different stated licence frameworks; they should not be collapsed into one blanket permission.
Source: LG AI Research · EXAONE 4.5 official repository and licenceOur perspective: distribution can be a commercial strategy.
Publishing model weights can expand adoption, technical scrutiny and an ecosystem of developers. It does not necessarily mean that the underlying company has abandoned its commercial rights or that every downstream use has the same permissions.
The relevant question is how access is combined with service, support, enterprise deployment, data and customer agreements. Different model releases can support different commercial routes.
Read the licence of the asset actually deployed.
A company should identify the exact release it uses and the terms attached to that release. The K-EXAONE 2.0 report is a useful example of why a later release can have a different distribution framework from an earlier model family. Current source files, notices and applicable agreements still need to be checked before deployment.
Investment diligence should connect that licence map with training-data permissions, customer contracts and restrictions on reuse. A broad description such as “open” is not enough to establish the issuer’s commercial rights.
Source: LG AI Research · K-EXAONE 2.0 technical report, August 2026What a credit assessment needs to know.
The business must be able to provide the product it has promised without depending on rights it does not possess. Licence continuity, deployment costs, servicing and customer ownership become part of the economics.
The broader Korean AI story is one of increasingly sophisticated technical and distribution choices. Our financing lens remains precise: identify the commercial right, verify the obligations and assess the route to collections. This article is a research perspective, not legal advice on a particular licence.
This article reflects KGCF's investment philosophy and research perspective. Public market and industry evidence is context, not an investment recommendation, forecast or indication of fund performance.
