Skip to content
Submit incident
AII-1001-58011
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.Supporting sources are pending verification.

AI-generated evidence with watermark submitted in Chinese rental dispute

A plaintiff submitted AI-generated photographs with a visible watermark as evidence in a rental dispute case in a Chinese court. The court detected the AI generation tool's watermark, identified contradictions in the plaintiff's testimony, and found that fabricated evidence had been intentionally submitted. The plaintiff admitted the falsification and was formally admonished by the court.

27 December 2025 Β· Dong Β· Misinformation and synthetic media
27 December 2025
AII-1001-58011
Quick facts
Category
Misinformation and synthetic media
Country
China
Date
27 December 2025
Drafted by
Raidu Incident Detection

What happened

In December 2025, during a housing rental contract dispute case in Hubei province's Dawu Court, the plaintiff submitted photographs as evidence showing unpaid utility bills that contained a "Doubao AI generated" watermark in the lower right corner.

The defendant's advocate, Dong, had initially claimed the utility meter was used exclusively by the tenant and promised to provide photos of the meter before and after the rental period. When the court examined the photos further, judges discovered contradictions: Dong first stated the utility meter was for single use, then admitted the building was "one staircase, two households" sharing a single meter.

Reported impact

The court identified that the AI-generated photos constituted fabricated important evidence and obstructed the litigation process.

Governance controls

The court issued a formal admonishment to Dong for using AI-generated photographs as legal evidence, explaining that such conduct violates the Civil Procedure Law and related judicial interpretations, and can result in admonishment, fines, detention, or potentially criminal liability.

Because Dong voluntarily confessed and corrected her error by providing authentic photographs, the court applied the principle of "combining punishment with education" and processed the matter as an admonishment. The fabricated evidence was rejected and the case was to be decided based on authentic evidence and verified facts.

Classification

Category
Misinformation and synthetic media
Industry
Government, security and defence
Country
China
Harm status
Actual harm
Intent
Intentional
Timing
Unknown
Harm types
Economic or property
Failure modes
Deepfake image, Manipulation or dark pattern
Consequences
Litigation
AI systems
θ±†εŒ…AI (Doubao AI)

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident