mign CEO Presents at 88th AI Seminar

Speaking Engagements

CEO of mign, a Developer of Generative AI Software for the Architecture and Civil Engineering Industries, Speaks at the 88th AI Seminar [Archive Video to Be Released]

Archive video coming soon: Seminar "Understanding What ChatGPT Can Really Do — A Technical Overview and Real-World Case Studies" — Covering the fundamentals of generative AI and how it is being applied across industries —

Masahiro Tsuma, CEO of mign Inc. (pronounced “mine”; headquartered in Bunkyo-ku, Tokyo), a company that develops software for the architecture and civil engineering industries using advanced technologies such as AI and VR, spoke at the 88th AI Seminar, “Understanding What ChatGPT Can Really Do — A Technical Overview and Real-World Case Studies,” hosted by the Japanese Society for Artificial Intelligence (JSAI) on June 19, 2023.

A recording of the seminar is being edited and will be released either as bonus content for paid seminar members or on the official JSAI (Japanese Society for Artificial Intelligence) YouTube channel.

The seminar is aimed at anyone who wants an expert-level technical overview of text-generating AI, as well as engineers in the field who are considering how to put generative AI tools such as ChatGPT to work in their own industries. If that sounds like you, we encourage you to watch the video once it is available.

▼ Details of the 88th AI Seminar (event has concluded): https://www.ai-gakkai.or.jp/event/ai-seminar/no88_jsai_seminar/

• 88th AI Seminar: "Understanding What ChatGPT Can Really Do — A Technical Overview and Real-World Case Studies" — Session Overview (Archive Release)

The rapid spread of generative AI, exemplified by ChatGPT, has prompted more people than ever to start exploring how they might put AI to use. ChatGPT, for instance, has been reported to handle not only established NLP (natural language processing) tasks such as translation and summarization within a single model, but also tasks that were never anticipated before.

Even so, much about how to use generative AI remains unclear—from distinguishing the tasks it can perform from those it cannot, to finding the best instructions (prompts) to give it. Enthusiasts continue to experiment through trial and error, but examples that have made it all the way to real industrial application are still few and far between.

This seminar combined a technical explanation from a natural language processing expert with case studies of industrial applications. The featured expert was Hiroya Takamura of the National Institute of Advanced Industrial Science and Technology (AIST), a leading authority on natural language processing and machine learning. Beyond a technical overview of text-generating AI such as ChatGPT, he discussed what ChatGPT has newly made possible and offered an academic perspective on the path toward industrial applications.

In the case study portion, several engineers from companies that have already begun applying text-generating AI in their businesses took the stage to explain how they are using the technology and the trial-and-error process that led up to launching their services. Through these talks, the seminar aimed to give participants deeper, practical insight into how generative AI can be applied.

• Speakers

Technical Overview

Hiroya Takamura (National Institute of Advanced Industrial Science and Technology)

Tatsuya Ishigaki (National Institute of Advanced Industrial Science and Technology)

Corporate Case Studies

Shunsuke Kozawa (Gunosy Inc.)

Masahiro Tsuma (mign Inc.)

Yasuhiro Sogawa (Hitachi, Ltd.)

Takayuki Sangawa (City of Yokosuka)

• mign: Software Built for the Architecture and Civil Engineering Industries

At mign, we develop software for the architecture and civil engineering industries that brings cutting-edge technologies—chiefly AI, VR, and AR—to the field ahead of the curve. We place particular emphasis on harnessing generative AI, including the AI chatbot ChatGPT and the image-generation software Stable Diffusion, and we move fast to turn these technologies into working products.