Level-5 Apologizes For AI Use In Event Footage, Will Avoid Direct AI In User-Facing Video
Level-5 is keeping AI in its pipeline while pulling it out of the images fans actually see, which is the compromise a publisher reaches when the efficiency gain is real but the audience resistance is louder.
Reporting from 1 source: ASCII.jp.
Level-5 apologized on September 17 for errors caused by generative AI and human mistakes in promotional videos shown at LEVEL5 VISION 2026 II. The company said it will limit AI to efficiency improvement and avoid direct use in footage delivered to users. Representative Akihiro Hino said on X on September 12 that he added AI processing to make the presentation flashier, and said careless AI output data is not in the games.
The disputed material came from concept art for Yo-kai Watch 2: Hadou, where power lines and vending machines were called unnatural and possibly AI-generated. Level-5 said the power lines were hand-drawn by a designer and the vending machines were human error, details omitted by mistake. It still acknowledged careless AI use in some footage. Hino framed the AI as a way to make the presentation flashier, and drew the line at the event itself: scenario, character designs, and basic settings stay human, and the games will ship without careless AI output. The company wants major titles built in two years instead of five.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.