DistantNews
Support us
VIDRAFT Launches AI Foundry to Custom-Develop AI for Companies and Institutions

VIDRAFT Launches AI Foundry to Custom-Develop AI for Companies and Institutions

From Dong-A Ilbo · () Korean

Translated from Korean and summarized by DistantNews. Read the original for the full story.

At a glance

News Official statement New plan
  • South Korean AI startup VIDRAFT is launching an AI Foundry business that customizes existing AI for companies and public institutions using customer data and objectives.
  • The company targets security-sensitive sectors including public services, defense, healthcare and finance, using Model MRI, Darwin and Chimera to assess and optimize AI for deployment on internal networks.

South Korean deep-tech startup VIDRAFT is pursuing an “AI Foundry” business that develops artificial intelligence for companies and public institutions and delivers systems ready for use in the field.

The company compares the model to semiconductor contract manufacturing. It will examine existing AI using customer-provided data and goals, then apply breeding, knowledge transfer and optimization processes. VIDRAFT says, “Intelligence is not created, but cultivated.”

The startup is targeting public services, defense, healthcare and finance, where data security is a core requirement. These institutions need dedicated models that can run on internal networks, but building them independently can take considerable time and money. VIDRAFT says its Model MRI solution diagnoses an AI system before its Darwin and Chimera technologies adjust performance for specific domains.

Intelligence is not created, but cultivated.

— Kim Min-sik, CEO of VIDRAFTThe company used the phrase to explain its approach of adapting and developing existing AI rather than building every system from scratch.
About this summary

Originally published by Dong-A Ilbo in Korean. Translated, summarized, and contextualized automatically by DistantNews, with a note on how the source frames the story. Not individually reviewed before publishing. How this works.