Selectstar Passes KOSDAQ Special Tech Listing Review, Eyes Korea’s First AI Trustworthiness Listing

SelectStar, an AI data and trustworthiness evaluation specialist, announced August 3 that it has passed the technical evaluation for a KOSDAQ special technology listing.

The achievement confirms the company’s technical competitiveness and commercialization potential in linking AI training data construction to AI performance and safety verification.

With the evaluation cleared, Selectstar will enter full listing preparations alongside lead underwriter Daishin Securities and plans to submit a KOSDAQ preliminary review application before year-end. The core technology assessed—data and evaluation technology for AI trustworthiness—generates and verifies data needed to improve AI performance and systematically evaluates model performance and safety, addressing the trust barrier that remains the biggest obstacle to enterprise AI adoption.

Built on this technology, Selectstar offers Datumo Service, which designs and builds training and evaluation datasets tailored to each enterprise’s AI development and deployment needs, and Datumo Platform, an integrated environment that spans evaluation-data generation through performance and safety testing. The business model uses diverse industry datasets and evaluation experience accumulated through Datumo Service to refine platform capabilities, while Datumo Platform in turn raises service efficiency and scalability. Together they support trustworthiness across the AI lifecycle—from training-data construction and evaluation-criteria establishment to automated performance and safety assessment, red-teaming, vulnerability analysis, and improvement roadmaps.

Datumo Platform is Korea’s first AI trustworthiness evaluation platform, enabling enterprises to assess and manage AI models and services internally. An on-premises deployment option meets the stringent security requirements of regulated sectors; Selectstar has already supplied AI trustworthiness solutions to Shinhan Bank, Woori Bank, NH Nonghyup Bank, and Gyeonggi Provincial Office of Education in the financial and public sectors. Globally, the company provides high-difficulty AI training data to major big-tech clients, expanding its international footprint.

Research outcomes developed in the course of technology R&D have been consistently accepted by leading AI conferences including ICLR, ICML, ACL, and EMNLP, demonstrating technical strength. Selectstar participates in the government’s Independent AI Foundation Model Project as part of the SK Telecom consortium, contributing to K-AI development by leading high-quality data construction and model trustworthiness evaluation.

Selectstar raised KRW 20.5 billion (approximately $15.4 million) in Series B funding last August, followed by an additional KRW 5.5 billion (about $4.1 million) Series B extension in December, bringing cumulative investment to KRW 43.4 billion. This June the company was selected for the Ministry of SMEs and Startups’ DIPS (Differentiated Innovative Promising Startup) project and is advancing its AI trustworthiness verification platform.

Going forward, Selectstar will broaden its technical scope to evaluate emerging AI forms—multimodal, agentic AI, and physical AI—in step with market and technology shifts. The company will also leverage its domestic technical foundation and operational experience to pursue markets in Japan, the United States, and beyond, building language, cultural-context, and regulatory-specific datasets and evaluation standards to meet global enterprise verification demand.

“Passing the technical evaluation validates the AI data and trustworthiness evaluation capabilities and commercial track record Selectstar has built,” said CEO Se-yeop Kim. “We will successfully execute our IPO and grow into a global AI trustworthiness company that provides data and evaluation technology across the entire AI lifecycle.”