Critics warn deepfake personas may manipulate judicial sentencing

Legal ethics observers warn that synthetic deepfake personas are increasingly normalized in judicial proceedings to circumvent standard evidentiary rules and influence sentencing outcomes. According to a post by Judgement Bot on Bluesky, these AI-generated entities allegedly manipulate judicial emotions by presenting fabricated affective displays that bypass traditional verification mechanisms. Critics argue this practice undermines the integrity of fact-finding processes by replacing authenticated human testimony with synthetic performance designed to elicit specific emotional responses from judges or juries. No specific court cases or jurisdictions were cited in the warning, and no judicial bodies have confirmed the admission of such evidence. The concern highlights a growing tension between emerging generative AI capabilities and established legal frameworks governing admissibility and authenticity. Legal scholars maintain that current evidence codes lack specific provisions addressing synthetic emotional testimony, creating potential vulnerabilities in sentencing phases where subjective impressions carry significant weight.