MiHoYo burned through nearly $300,000 worth of AI tokens in just 13 hours during an internal AI experiment.
The story came from Zheng Yinhe, head of the AI NPC speaking at the 2026 Alibaba Cloud Summit.
According to him, an employee set up dozens of AI agents to work together on a project. The agents kept talking to each other, generating outputs, and reviewing each other’s work, causing token usage to skyrocket overnight.
The expensive experiment didn’t pay off. Despite the massive cost, the final results were reportedly full of hallucinations and unusable.
MiHoYo burned through nearly $300,000 worth of AI tokens in just 13 hours during an internal AI experiment.
The story came from Zheng Yinhe, head of the AI NPC speaking at the 2026 Alibaba Cloud Summit.
According to him, an employee set up dozens of AI agents to work together on a project. The agents kept talking to each other, generating outputs, and reviewing each other’s work, causing token usage to skyrocket overnight.
The expensive experiment didn’t pay off. Despite the massive cost, the final results were reportedly full of hallucinations and unusable.
Java devs: you might be missing a key tool in your toolkit - #DataFrames.
Why consider them?
✅ Transform, organize, and analyze data directly in Java
✅ Improve efficiency, flexibility, readability, and developer productivity
🔗 https://t.co/x2zyk33N3I
#Java
Spring AOP lets you inject logic (logging, security,
transactions) into methods WITHOUT touching the actual
code.💛
It wraps your bean in a proxy and intercepts calls.
#CtrlPlusCode#Spring
🧵 Thread on how it works 👇