AIToday
Large Language ModelsTop Companies' AI MovesAI Business & IndustryTop Companies AIPublished: Nov 10, 2026, 06:30 JST

Japan's AI that replaces humans failed; connective AI sticks, says PKSHA Technology's Katsuya Uenoyama

Send on LINE
Japan's AI that replaces humans failed; connective AI sticks, says PKSHA Technology's Katsuya Uenoyama

3 Key Points

  1. What happened

    PKSHA Technology CEO Katsuya Uenoyama said after 14 years of AI deployments at over 4,800 companies, replacement-style AI failed to take hold while connective AI stuck; at JR East's JRE POINT call center, AI transcribes calls as human operators handle customers.

  2. Why it matters

    AI that replaces a person end to end does not embed in operations; AI that sits between people and handles the surrounding work — call records, meeting minutes, clinical notes — gains a foothold, Uenoyama found across 4,800 deployments.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

Uenoyama's account is built around a contrast he observed over 14 years of deployments at more than 4,800 companies: AI meant to substitute for a person whole tends to fade, while AI placed between people — taking over the notes, records and minutes around a human interaction — settles in. The JR East JRE POINT call center is his concrete example, where the customer conversation stays with a human operator and the AI writes the record beside them.

The framing he draws from this is that the tasks AI absorbs first are not the human-facing work itself but the layer around it: care workers' nighttime records, sales minutes after a meeting, doctors' notes between consultations. The automation of routine clerical and standard white-collar work still proceeds, in his telling; what he treats as the open question is whether people can adapt to the pace at which the content of the work changes.

FAQ
What is 'connective AI'?
It is an approach where AI does not replace a person but sits between people and handles the work around them, such as writing up call records or meeting minutes. Uenoyama says this pattern emerged from repeated successes and failures in real deployments, not from an initial philosophy.
How did PKSHA Technology arrive at this idea?
PKSHA Technology says it has spent 14 years working on AI deployments with more than 4,800 companies, and the pattern surfaced as a common thread across those projects, including the failures.
Top Companies AIRead Original Article
Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleSakai AI Co-Innovation Kickoff confirmed startup lineup