AIToday
Large Language ModelsTop Companies' AI MovesAI Business & IndustryTop Companies AIPublished: Nov 21, 2026, 06:31 JST

Toyota's Yamada turns meeting review over to AI, scores rise

Send on LINE
Toyota's Yamada turns meeting review over to AI, scores rise

3 Key Points

  1. What happened

    Yamada Hikaru of Toyota's advanced product development division completed a "meeting quality evaluation prompt" in December 2025 for the internal TAA tool; pasting in minutes yields a 5-point score and three improvement tips.

  2. Why it matters

    The prompt asks the AI to check meeting duration and whether an agenda was shared beforehand, and it avoids naming individual participants when critiquing.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

Yamada's route into generative AI was accidental: in 2024 she was bedridden with mycoplasma pneumonia and a 40-degree fever, and while passing time on her phone she asked ChatGPT to personify the pathogen. She expected a spear-wielding demon but got a cute five-year-old girl playing in the snow, and the gap between her expectation and the output showed her how much the wording of a request changes the answer. That lesson carried into her job, where she had long been bothered by meetings that ran an hour when scheduled for 30 minutes and ended without decisions. Rather than propose a top-down fix alone, she asked Toyota's internal generative AI tool TAA how to build what she wanted, going back and forth 10 to 20 times and also consulting Copilot to combine the best parts of each suggestion. The resulting meeting-quality prompt deliberately stays gentle: it rates on a 5-point scale but avoids being too harsh, and it does not single out participants by name. The companion speaking-analysis prompt drops scores entirely, focusing instead on strengths, filler words like "eh" and "ano," and whether she states conclusions first, with advice phrased positively. Both prompts now sit on Toyota's internal Prompt Forest sharing site, and colleagues have said they want to try them and pass them to their groups. Yamada, who once found it hard to submit improvement ideas because her administrative engineering role offered no obvious targets, now defines improvement more broadly: anything that makes herself or those around her a little happier or a little more able to head home early counts.

FAQ
How does the meeting quality evaluation prompt work?
You paste the meeting minutes at the end of the prompt and let the AI read them. The AI asks for missing details like meeting duration and whether the agenda was shared in advance, then gives a 5-point score and three improvement tips.
What did Yamada change after seeing the AI's feedback?
For her regular meeting with familiar members, she began creating and sharing an agenda with the topic and duration beforehand. She also started saying "let's wrap up" as the end time approached and confirming who would do what.
Why did the second prompt not use scores?
Yamada said speaking habits take years to form and cannot be fixed quickly. She felt gentle, forward-looking advice suited her better than a numeric score, and she prioritized psychological safety so no one would feel hurt or lose motivation.
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 articleDaiwa House Industry proposes prefab data centers to bypass power wait