Teach it to AI! Trial Exhibition to Collectively Train Capybara AI Behavior Analysis Tool via Customer Participation

AQUARIUM×ART átoa and Kobe Digital Lab have started a trial exhibition of a prototype AI behavior analysis tool for capybaras. Visitors participate by annotating images on tablets, helping the AI learn to evaluate environmental enrichment.
イベントNQ 75/100出典:PR Times

📋 Article Processing Timeline

  • 📰 Published: March 30, 2026 at 23:00
  • 🔍 Collected: March 30, 2026 at 22:56
  • 🤖 AI Analyzed: April 25, 2026 at 04:11 (605h 15m after Collected)
AQUARIUM×ART átoa (hereafter, átoa) will hold a trial exhibition of an 'original tool that automatically records and analyzes the behavior of reared animals using image AI (hereafter, this tool),' which is being jointly developed with Kobe Digital Lab Co., Ltd. (Chuo-ku, Kobe City, Hyogo Prefecture; President Shinichi Tamaki; hereafter KDL).

This tool is being developed to objectively evaluate the effects of the environmental enrichment (*) conducted by átoa. At the same time, a major feature is that it is a participatory exhibition that provides interactive learning, where the AI grows through visitor participation, allowing visitors to naturally enjoy and touch upon animal behavior and ecology.

Through this trial exhibition, we plan to verify usability for visitors and necessary functions, and feedback the results into development for full-scale exhibition and operation.

[Image: Participation image]
[Image: átoa's capybara]

## Exhibition Overview
Now that the prototype of this tool is complete, we will conduct a trial exhibition targeting capybaras. Visitors will enclose the capybara in the image from a tablet installed at the exhibition corner (this is called annotation), and select (record) its behavior at that time. As the input image data increases to 100, 200, 500, and 1000 images, the AI's ability to judge when, where, and what the capybara is doing from the image will improve. By using this AI analysis tool raised by visitors, it will become possible to track capybara behavior in detail over long periods, creating a system that can objectively evaluate what kind of environmental enrichment is effective.

[Image: iPad input screen]
[Image: iPad input screen part 2]