A high school athletic director in Maryland has been accused of using AI to produce a fake audio clip containing racist and antisemitic remarks, ostensibly spoken by a principal. This instance, reportedly one of the first of its kind in the United States, has not only exposed the increasing ease of creating such deceptions but also the dire need for new legal measures to combat the misuse of AI technology.
Dazhon Darien allegedly crafted an audio recording mimicking the voice of Pikesville High School’s principal. This incendiary clip, which reportedly spread rapidly on social media, led to severe repercussions. The principal was placed on administrative leave, and the resultant social media backlash and flood of hateful messages and calls to the school have created a tangible atmosphere of fear and disruption within the educational community.
Experts have warned that the AI, which has enabled the creation of such deepfakes, has become “increasingly powerful and yet very easy to use,” according to Siwei Lyu, director of a media forensics lab at the University at Buffalo. The technology has become so accessible that, according to Lyu, “You can basically upload any subject’s voice up to this platform. And then you can give it text and you can start creating voices of that person.”
Baltimore County State’s Attorney Scott Shellenberger has identified an urgent need for legislative action, stating that while charges like disrupting school activities carry a “6-month sentence,” there is a greater necessity to “take a broader look at how this technology can be used and abused to harm other people.” The case has spurred discussions about the potential for AI to inflict harm and the lag in detection methods, which are struggling to keep pace with the rapid advancements in AI technology. As Lyu points out, it’s a “perpetual cat-and-mouse game,” where detection is often a step behind due to comparatively fewer resources and lesser attention than the generative aspect of AI.
The recorded voice of Principal Eric Eiswert was artificially generated to include disparaging remarks about Black students and Jewish individuals. Despite the eventual exposure of the fraudulence through expert analysis, including evidence of “AI-generated content with human editing after the fact,” as per court documents, the episode has left an indelible mark on the community and the individuals involved.
The incident has heightened awareness of the vulnerability of anyone to AI-manipulated attacks. Hany Farid, a digital forensics professor at the University of California, Berkeley, emphasized, “Everybody is vulnerable to attack, and anyone can do the attacking.”
Relevant articles:
– Athletic director used AI to frame principal with racist remarks in fake audio clip, police say, apnews.com, 05/02/2024
– Maryland Athletic Director Used AI to Frame Principal With Racist Remarks in Fake Audio Clip, Police Say, Ponca City Now, 04/30/2024
– Deepfake of principal’s voice is the latest case of AI being used for harm, WSB Atlanta, 04/29/2024