Indie Foundation Model Second Phase Scores Released... SKT Ranks First

Detailed Evaluation Scores Disclosed Following Motif's Objection

The detailed scores for the second phase of the Independent Artificial Intelligence (AI) Foundation Model (Indie Foundation Model) Project were released on August 27.

Deputy Prime Minister and Minister of Science and ICT, Paik Kyunghun (center), is seated on the 27th at the Data-Related Ministers' Meeting held at the Government Seoul Office, presided over by Prime Minister Han Seongsuk. Photo by Yonhap News

This disclosure of the detailed results was made at the request of Motif Technologies, the team eliminated in the second phase evaluation, and with the consent of the other elite teams.

Based on the overall results of the second phase evaluation, the elite team from SK Telecom ranked first with a score of 70.6 points.

They were followed by Upstage with 69.9 points, LG AI Research Institute with 69.0 points, and Motif with 65.8 points.

This evaluation was scored out of 100 points, consisting of benchmark tests (25 points for the Artificial Analysis Intelligence Index (AAII), 15 points for the National Information Society Agency (NIA)), expert evaluation (35 points), and user evaluation (15 points from experts, 10 points from the general public).

By category, in the Artificial Analysis Intelligence Index (AAII) evaluation, which adopted global standards, Motif ranked first with 11.9 points.

They were followed by Upstage with 9.4 points, SK Telecom with 8.8 points, and LG AI Research Institute with 7.8 points.

In the National Information Society Agency (NIA) benchmark evaluation, which covered seven areas including Korean language and reliability, SKT (13.4 points) and Upstage (13.3 points) took the lead with a narrow margin. LG AI Research Institute scored 12.8 points, while Motif recorded 12.7 points.

In the expert evaluation, which involved 10 external experts from the industry, academia, and research sector, LG AI Research Institute received the highest score at 29.5 points, followed by SKT (29.3 points), Upstage (29.1 points), and Motif (27.1 points).

From the user evaluation, which measured actual usability, SKT ranked first in the expert user category with 11.6 points, based on responses from 49 AI startup CEOs. In the public evaluation category, LG AI Research Institute took first place with 7.6 points, as determined by a panel of 185 members of the general public.

The Ministry of Science and ICT stated, "The Indie Foundation Model is not simply about ranking or eliminating teams; it is aimed at the qualitative growth and expansion of the domestic AI ecosystem." The ministry explained, "Considering the rapid evolution of AI technology, we change our development goals every six months (moving target), and the evaluation criteria and methods were also finalized and implemented after prior consultations among the elite teams without any disagreements."

Previously, Motif filed an objection with the Ministry of Science and ICT regarding the results of the second phase selection for the Indie Foundation Model.

In response, the Ministry of Science and ICT announced that Motif's objection would be processed in accordance with established procedures, and that a final decision would be made within 15 days.

This content was produced with the assistance of AI translation services.