TheBoard.World: The Liar’s Dividend: When Seeing Is No Longer Believing in Wartime

AI deepfakes are hyper-realistic audio, video, or images generated by artificial intelligence to mimic real people or events, with the intent to deceive. During the Iran war, a surge of deepfake content on X (formerly Twitter) has overwhelmed both the public and verification teams, eroding the credibility of all digital evidence. This creates an environment where even real atrocities can be dismissed as fabrications, fundamentally undermining trust in digital information.

Key Findings

• During the peak of the Iran war, individual AI-generated fakes on X reached over 30 million impressions each, with three new high-profile deepfakes appearing every hour (aicerts.ai, 2026).
• The volume and realism of deepfakes have outpaced current detection and verification capabilities, as reported by the New York Times and aicerts.ai.
• The principal risk is not just misinformation, but the collapse of all digital evidence credibility—a “liar’s dividend” where real events are dismissed as fakes.
• Regulatory, legal, and media systems face urgent pressure to adapt verification protocols or risk systemic mistrust and operational paralysis