ZRON documents show AI packaging stolen files as police intel

Internal documents from a Chinese hacker-for-hire firm show artificial intelligence being used to turn purportedly stolen foreign-government material into finished intelligence products for police or security customers, the Wall Street Journal reported, though the paper says it could not authenticate every document, prove how the information was obtained, or confirm that the firm carried out the alleged intrusions.

The trove, attributed to Zhengzhou Zhirong Network Technology Co., known as ZRON, includes "company chat logs, intelligence reports and a slide deck apparently intended for prospective clients," per the Journal. The reporting describes purported Russian diplomatic correspondence, preparations for foreign leaders' visits to the Philippines, and a document that "contained minutes and decisions from a 2023 meeting at the prime minister's office" of Pakistan.

The AI layer is what separates this from earlier hack-and-dump material. According to the reporting, ZRON's system was designed to process large datasets, create timelines and generate automated reports. One employee suggested AI could conduct risk assessments while human staff supplied the underlying information. The stated aim, per the Journal, was to make information "more digestible for police or security customers."

Alongside the intelligence product, the documents describe surveillance tooling: screen activity and keystroke capture on Windows PCs, contact and location data from Apple devices, and methods for extracting information from webmail accounts and internal email systems.

The Journal is deliberate about what the records prove. Reviewing them "does not authenticate every purported government document, establish how any information was obtained or prove that ZRON carried out the alleged intrusions," and the paper could not confirm whether any of the material was actually delivered to customers. It slots into a wider stream of coverage we track on China AI and cybersecurity, where AI is now routinely described inside offensive operations rather than only alongside them.