Harvard Astronomer Avi Loeb Targeted by AI Impersonation Scheme on YouTube
The rapid advancement of generative AI is creating new avenues for deception, and even prominent scientists are not immune. Harvard astronomer Avi Loeb, known for his theories regarding the interstellar object 3I/ATLAS, has become the target of a sophisticated impersonation scheme on YouTube, highlighting a growing threat to online authenticity.
The proliferation of AI-generated deepfakes and cloned voices has made it easier than ever for malicious actors to impersonate individuals for nefarious purposes, ranging from phishing scams to the spread of misinformation. This trend recently ensnared Loeb, with a YouTube channel falsely presenting itself as the astronomer through the use of AI-cloned likeness and voice.
“Indeed, these videos are fake, produced by AI,” Loeb confirmed in an email to Futurism. “I reported them to YouTube.”
The fraudulent channel distinguishes itself from Loeb’s actual research by presenting sensationalized claims about 3I/ATLAS, with titles like “3I/ATLAS Is a PROBE — New Data Leaves No Doubt.” This contrasts with Loeb’s own approach, which acknowledges the possibility of both natural and technological origins for the interstellar object while maintaining a rigorous scientific methodology.
Loeb detailed the broader implications of this impersonation in a recent blog post, questioning how the public can discern truth from fabrication in an age of readily available AI tools. “Imagine creating videos that feature avatars of scientists that look like them and speak in their voice, but spread counterfactual information about 3I/ATLAS,” he wrote. “How would the public know who to believe?”
He emphasized the urgency of the situation, stating, “This is not a plot from a science fiction novel but the reality we face today.” Over the past two weeks, Loeb reported receiving hundreds of emails from concerned fans who identified the fraudulent channel. He noted subtle clues indicating the videos were AI-generated, including “strange jerky movements” and a “clock that’s ‘frozen on a specific time’” in the background.
Loeb intends to pursue legal action against the creators of the fake videos, citing defamation and the dissemination of false content. Despite filing numerous reports with YouTube, he expressed frustration with the platform’s lack of response.
YouTube’s terms of service explicitly prohibit impersonation, stating that content “intended to impersonate a person or channel” is a violation that may result in account termination. However, as of press time, YouTube had not responded to inquiries from Futurism regarding the matter.
The motivation behind the impersonation appears to be twofold: financial gain and the potential for spreading misinformation. Loeb suggested the creators may be attempting to profit from advertising revenue, given his substantial online following – exceeding 5 million readers per month on Medium.com.
The offending YouTube channel, created in September, had already amassed over 1.4 million views as of late November. If monetized at a rate of one to three cents per view, the account could have generated between $14,000 and $42,000. Prior to impersonating Loeb, the channel featured content in Tagalog with a man in a lab coat offering health advice.
This incident serves as a stark warning about the ease with which AI can be used to deceive and manipulate. “We are living in a new reality where fake content can be easily generated by AI,” Loeb wrote. “This raises a serious problem of how to authenticate the validity of information on the internet.”
He concluded by asserting that “science focuses on figuring out the physical reality based on facts,” and that “the nemesis of science is fake content created by AI.”
