AIToday
Large Language ModelsZenn AI/MLPublished: Oct 13, 2026, 22:00 JST

Ceatec 2026 attendee on Zenn: AI is just a calculator, not a forecaster

Ceatec 2026 attendee on Zenn: AI is just a calculator, not a forecaster

3 Key Points

  1. What happened

    A programmer who attended Ceatec 2026 wrote on Zenn that almost every booth was AI-related, that he was surprised to find a staffing agency there, and that his local LLM (an AI model run on a company's own machines) plan for internal AI had been branded "AI knowledge".

  2. Why it matters

    He argues the industry now prioritizes the buyer's image over technical specifications, and that "AI says wrong things" is disclosed largely because a company could face huge damages if something happened and it were not disclosed.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Context & Analysis

The Zenn post is framed around a visit to Ceatec 2026, where the author found almost every booth AI-related and was surprised to see a staffing agency among them. He contrasts this with past shows where visitors came specifically to touch products like the Mac, and says today's exhibits are mostly parts of larger systems, prompting image-based pitches rather than technical specifics.

The author ties this to a broader concern that both buyers and sellers let image take priority over understanding what is real. In that context, he notes that current AI cannot explain the mathematics of Transformer Self-Sttension's KQV, and that it is used because validation studies found its answers accurate. That, he writes, is the real reason AI is labeled as sometimes wrong: without such disclosure, a lawsuit could cost an AI company massive damages.

He also describes his own experience using AI as a programming aid past retirement age, saying his form system ACR took a year to build with AI and required what would likely have been over 1,000 programmers and testers, and hundreds of millions in yen, 30 years ago. He closes by reflecting on a TV story of an 85-year-old woman with three months to live who talks with AI daily, and on the hope that machines, originally meant to help people, are never used to harm or deceive.

FAQ
What does the author say about AI and future prediction?
He writes that AI cannot predict the future and is essentially just a calculator.
Why does the author say AI discloses that it can be wrong?
He argues the disclosure exists largely because, if something happened and it were not disclosed, an AI company could face massive damages in court.
What did the author build with AI?
His form system ACR took one year to build with AI's help, and he has started releasing it.

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 articleMicroLLM lab runs seven tiny AI models in your browser