Beijing Engineer Sentenced to 5 Years for Deleting 89 TB of AI Training Data
In September 2024, a Beijing algorithm engineer illegally accessed his company's server cluster, executed a notorious delete command that erased over 89 TB of AI training data and 3D model assets, causing system paralysis and more than 200,000 yuan in losses, and was sentenced to five years and ten months in prison for destroying a computer information system.
In September 2024, Wang, an algorithm engineer at a technology company in Beijing's Dongcheng District, illegally logged into the company’s server cluster to use computing resources for personal projects and entered a widely recognized dangerous "delete‑all" command, wiping out more than 89 TB of AI game‑department training data and several proprietary text‑to‑3D AI models in a single night. The action crippled the model training system, halted all R&D projects, and resulted in economic losses exceeding 200,000 yuan.
Investigation revealed that Wang colluded with an external individual, deliberately bypassed security controls, and used a command whose meaning is “no confirmation, force delete all files,” demonstrating clear subjective intent.
During the case, the Beijing Procuratorate assigned a special assistant prosecutor with dual roles as a technical advisor to review evidence. Experts in criminal law and computer science were consulted to assess the economic damage, including direct data reset costs and expenses required to restore data and repair model functionality. They concluded that AI model training data constitutes a core component of a computer information system, giving it the attribute of a “computer information system” under the law.
The court convicted Wang of the crime of destroying a computer information system and sentenced him to five years and ten months in prison.
The case, the first in Beijing involving illegal deletion of AI training data, clarifies the legal status of AI models and training systems, extends legal protection to computing‑power losses, and provides a new basis for quantifying the amount of crime in similar incidents.
Black & White Path
We are the beacon of the cyber world, a stepping stone on the road to security.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
