(Credit: Boonchai Wedmakawand via Getty Images)
How do you stop criminals from using generative AI to exploit unsuspecting users? The FBI suggests you may want to cool it with the public social media posts.
In a Tuesday alert, the agency warned that criminals are abusing generative AI technologies in numerous ways, often to trick victims into handing over their cash. It lays out 17 techniques that criminals favor for creating deepfakes—or AI-generated images, video, and voice clones—which help them pose as the victims, their family members, or celebrities.
“Criminals generate short audio clips containing a loved one's voice to impersonate a close relative in a crisis situation, asking for immediate financial assistance or demanding a ransom,” the FBI said in one example—a tactic that grabbed headlines last year. In some cases, the culprits are even using AI to power real-time video calls that can deepfake a company executive or a law enforcement official.
(Credit: Metaphysic)In response, the FBI urges the public to consider some rather drastic measures to deny criminals access to personal content that could fuel their schemes. That’s because scammers often steal publicly available photos, videos, and audio to convincingly mimic individuals.
“If possible, limit online content of your image or voice, make social media accounts private, and limit followers to people you know to minimize fraudsters' capabilities to use generative AI software to create fraudulent identities for social engineering,” the FBI wrote in the alert.
The agency also says users should consider creating “a secret word or phrase with your family” to help users weed out voice-cloning scams that could occur on the phone or online. “Listen closely to the tone and word choice to distinguish between a legitimate phone call from a loved one and an AI-generated vocal cloning,” the FBI adds.
The alert underscores how generative AI, for its all benefits, is also unleashing a new wave of more advanced cybercriminal schemes. The FBI’s warning suggests that the public may need to adopt stricter security practices and rethink how much personal information they share online.