Completing AI Humans Through Interactive Domains

At the 14th Future ICT Forum, Genesislab AI Research Lab Director Yoo Ji-hyeong presented how interactive domain technology enables digital humans to conduct holistic evaluation through real-time interaction, gesture recognition, and gaze analysis.

Share
Completing AI Humans Through Interactive Domains
Photo by Zach M / Unsplash

Source: Financial News (파이낸셜뉴스) — Yoo Ji-hyeong, Lab Director: “AI humans completed through interactive domains” [14th Future ICT Forum] Original in Korean

At the 14th Future ICT Forum co-hosted by Financial News and Korea’s Ministry of Science and ICT, Yoo Ji-hyeong, AI Research Lab Director at Genesislab, presented on the AI human domain. His central argument was that interactive domain engagement represents the key to creating digital humans that move and respond like real people.

Yoo explained: “Much like the stages of human cognition, AI humans must progress beyond simple learning to incorporate analysis, evaluation, and generation capabilities.” He linked this progression to the concept of interactive AI as Genesislab defines it.

During the digital human showcase conducted at the forum, Genesislab unveiled its AI human prototype. The system demonstrates real-time responsiveness to questions and conversation, expressing not only verbal responses but also gestures and facial expressions. The foundation is interactive AI technology. Genesislab plans to extend the technology beyond AI and digital interview contexts to new applications including mental health specialists and content creators.

Yoo stated: “Interactive AI-based digital humans can analyze gaze patterns and conduct holistic personality evaluation just as human interviewers assess candidates.” The system creates a two-way exchange: it reads the user’s body language and speech in real-time, returning contextually appropriate expressions, gestures, and dialogue in response.

Read more

단일 LLM의 한계를 넘어서: Multi-Agent System은 왜 필요한가

단일 LLM의 한계를 넘어서: Multi-Agent System은 왜 필요한가

단일 LLM으로 복잡한 비즈니스 문제를 해결하는 접근은 현실에서 쉽게 한계에 부딪힌다. 이 글에서는 단일 프롬프트부터 멀티 에이전트 시스템에 이르기까지 AI 아키텍처의 발전 단계를 분석하고, 각 구조가 왜 실패하거나 부족했는지 그 이유를 짚는다. 그리고 그 흐름 속에서 도출되는 멀티 에이전트 스케일링 법칙이 B2B 플랫폼 설계에 어떤 시사점을 주는지 살펴본다.