OpenAI and Meta warn of AI deepfakes threatening legal evidence integrity

On October 20, 2026, Reuters reported that OpenAI and Meta representatives warned at a U.S. judicial conference on evidence rules that AI-generated deepfakes pose a significant threat to the reliability of evidence in legal proceedings. The companies emphasized that distinguishing between real media and content produced by advanced AI models is increasingly difficult. They noted that the cost and time required to create such media are minimal. Experts highlighted that courts are struggling to detect AI-generated false evidence, despite efforts by AI firms to add visible or invisible watermarks. However, deepfakes remain rare in legal practice, with only 15 out of 931 U.S. judges reporting encounters with deepfake-related evidence challenges. OpenAI and Meta stressed the growing difficulty for courts to counter AI-generated misinformation. While some companies have introduced watermarks to identify synthetic media, no universal detection methods exist. The companies called for stronger safeguards to protect the integrity of legal evidence. The report underscores the urgent need for updated legal frameworks to address the risks posed by deepfake technology in the courtroom.