South KoreaChemicals and materials2026 – ongoingDongjin Semichem
In-house LLM Dongjin AI Ask for patent and paper search before R&D
Dongjin Semichem built an internal-network LLM for patent and paper search, at a cost of several hundred million won
Problem
Dongjin Semichem makes photoresist and other semiconductor materials. Researchers needed to check papers and patents for similar existing work, and chemical and process data had to stay inside the company.
Approach
The company built Dongjin AI Ask, an in-house large language model that runs only on its internal network. It is mainly used to search papers and patents before R&D starts. Herald Economy reported the build cost as several hundred million won.
Results
Takeaway
The company kept the model on its own network to protect process data. No performance figure was reported.
Categories
- Industry
- Chemicals and materials
- Business function
- IT and R&DInternal work and knowledge management
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Built in-house
- Evidence
- Press reporting
Sources
Source summaries
The article describes Dongjin Semichem's in-house LLM, which runs only on the internal network and is used to search papers and patents. It gives the build cost as several hundred million won and reports no usage or performance figures.
Full details are in the original sources.
