Updated
Updated · The Washington Post · Sep 5
Jeremy Moss Audio Clip Roils Michigan Race as 25-Second Recording Defies Authentication
Updated
Updated · The Washington Post · Sep 5

Jeremy Moss Audio Clip Roils Michigan Race as 25-Second Recording Defies Authentication

1 articles · Updated · The Washington Post · Sep 5

Summary

  • Digital-forensics experts found no evidence the 25-second clip attributed to Jeremy Moss was AI-generated or spliced, but they could not verify the speaker, source or missing context.
  • The recording surfaced on Reddit in November and spread in July after Moss’s Democratic primary rival Aisha Farooqi posted it; a separate anti-AIPAC group then added misleading captions and drew at least 1.5 million X views.
  • Moss has called the audio manipulated, unsourced disinformation and linked the fallout to antisemitic tropes and a toxic climate that included a bomb threat at his home last year.
  • Hasan Piker amplified the clip on Aug. 20 during a stream about Jews and Israel, triggering bipartisan backlash and forcing Michigan Democratic Senate nominee Abdul El-Sayed to distance himself.
  • The episode underscores the 'liar’s dividend' in AI-era politics: synthetic media can deceive voters, while real but murky material can be dismissed as fake, deepening distrust ahead of elections.

Insights

If experts cannot authenticate a viral recording, how can ordinary citizens separate digital truth from fiction?
Does the mere existence of AI technology provide a perfect shield for anyone looking to deny reality?