President Donald Trump on Tuesday publicly embraced blaming artificial intelligence for a viral video showing someone tossing an object out of an upper-story White House window, saying he might fault AI if "something happens that's really bad." His remark followed a contradictory signal from his own press team, which had told reporters the footage was authentic.
The exchange underscored a growing political strategy in which officials and public figures point to AI as an explanation for embarrassing or damaging material. Experts and analysts say the approach exploits what has become known as the "liar's dividend" — the advantage gained by those who cast doubt on real evidence by invoking the possibility of synthetic manipulation.
"If something happens that's really bad, maybe I'll have to just blame AI," Trump told reporters, framing AI as a ready defense. The comment came as social media circulated the video and reporters sought clarification from the White House. Administration aides initially indicated the footage was genuine, a detail that highlights the tension between public statements and subsequent explanations invoking AI.
The tactic is not unique to the United States. On the same day, Venezuela's communications minister, Freddy Ñáñez, publicly questioned the veracity of a video circulated by the Trump administration, illustrating how claims about synthetic media can be deployed internationally to challenge or deflect political narratives.
Analysts say the liar's dividend arises from a simple dynamic: as generative tools become more capable of producing convincing audio, images and video, it becomes easier for those accused of wrongdoing to claim that incriminating evidence is a forgery. Unlike people, AI cannot be deposed or held legally accountable, nor can it be cross-examined to establish intent or provenance. That creates a legal and credibilty gray area that critics say undermines efforts to hold officials to account.
Journalists and technologists caution that legitimate concerns about manipulated media should not become a universal reflex for dismissing inconvenient facts. AI systems do make mistakes and can produce realistic fabrications, complicating efforts to verify material quickly. At the same time, experts warn that reflexive denial can erode trust in institutions and hamper public understanding of real events.
The spread of deepfakes and other synthetic content has prompted major technology companies and news organizations to develop detection tools and authentication practices. Lawmakers and regulators in several countries are also debating measures to address the harms of disinformation and to mandate clearer labeling of AI-generated content. Still, detection remains imperfect and lagging behind the pace of generative advances.
The political utility of invoking AI is likely to persist as a rhetorical shield, analysts say, especially when rapid public reaction to viral material can shape narratives before verification is complete. The growing use of that shield raises questions about how officials, journalists and the public can maintain standards of accountability when the plausibility of both real and fabricated evidence is increasing.
For now, the episode at the White House serves as a case study in how leaders may deploy emerging technology as a preemptive excuse. Observers say the longer-term implications will depend in part on whether institutions can improve verification methods and on whether audiences demand clearer standards for when claims of manipulation are accepted or rejected.