“For manufacturing and defense sites, general-purpose AI is not enough”… Seongho Yoon, CEO of Machinarax, reveals industry-specific AI strategy

Seong-ho Yoon, CEO of Machinarax, said at a press conference held on the 6th, “The era of physical AI has already begun, but the first places where it becomes a reality are not humanoids, but manufacturing industrial sites and combat sites. Unlike general cloud environments, industrial sites have high requirements for precision, reliability, and security, so there are limits to responding with general-purpose AI alone.” Machinarax cited the closed network environment and field data issues as key reasons for the difficulty in introducing AI in industrial sites. Manufacturing plants and defense facilities often have restrictions on exporting external data, making it difficult to apply general cloud-based AI services as is, and the data structure and operation method of each industrial equipment are different, so it is difficult to achieve the level of accuracy desired by companies without sufficiently learning actual field data.