AI Foundation Model Project: SK Telecom Ranks First in Second Evaluation, Motif Files Official Complaint

SK Telecom has emerged as the leader in the second phase of South Korea's ambitious independent AI foundation model project, known as Dokpamo. The team achieved an overall score of 70.6, narrowly beating out competitors in a closely watched evaluation.

The Ministry of Science and ICT, responding to a request for transparency from Motif Technologies, released detailed scores for the four participating teams: SK Telecom, Upstage, LG AI Research, and Motif. The evaluation encompassed benchmark tests, expert assessments, and user feedback, aiming to gauge the capabilities of these advanced AI models.

“

This evaluation is not about ranking or eliminating teams, but about driving the growth of domestic AI companies and qualitatively expanding the ecosystem.

While SK Telecom secured the top spot overall, the individual component scores reveal a more nuanced picture. Motif Technologies excelled in the AAII benchmark, and LG AI Research led in the expert evaluation. This suggests a competitive landscape where different teams showcase strengths in various aspects of AI development.

However, the release of the detailed results has been accompanied by a formal complaint from Motif Technologies. The company has officially requested the Ministry of Science and ICT to review the evaluation, citing concerns about the process. The ministry has stated it will handle the appeal through formal procedures and provide a decision within 15 days, underscoring the ongoing scrutiny and competitive tension within South Korea's AI development initiatives.

“

The evaluation criteria for this second phase were also established through multiple discussions with the elite teams. Following official notification of the final draft before the evaluation, there were no disagreements from the elite teams, and the evaluation proceeded based on that.