Earlier this year, a deepfake scam in a New Hampshire primary election (the Associated Press initially reported in January) included a robocall with what appeared to be President Joe Biden’s voice urging voters to “save your vote for the November election.”.
The Ohio House proposed a bill in February focused on prohibiting non-consensual deepfakes. It would create the “nonconsensual dissemination of fabricated sexual images” offense, a third degree felony.
The bill says a written or spoken statement must be included with a recording, video or image indicating it’s a manipulation.
“The cause of action created by this division is in addition to any other cause of action available under statutory or common law,” the bill continued.
The Ohio Attorney General’s Office issued a statement in April, about VERIFY, a team of journalists and researchers that work with newsrooms to fact-check supposed news stories, where they described deepfakes as: “a deepfake video is made using artificial intelligence technologies, like programs that can be used to replace or synthesize faces, speech or expressions of emotions.”
The attorney general’s office cited examples involving Jennifer Aniston, Taylor Swift and Selena Gomez, where these celebrities appear to endorse a product in a video, which were all revealed to be fake due and created by artificial intelligence software.
Bloomberg reported on June 20 more than 10 states such California, Washington, Texas, Georgia and others have proposed some type of regulation against deepfakes.
Other deepfake and AI bills have been similarly proposed.
VERIFY by the attorney general’s office recommended the following “litmus test” to help people determine whether a celebrity social media endorsement is real or part of an AI-generated scam:
- Movement: How does the celebrity move in the video clip? Do you see any unusual facial expressions and/or body language?
- Background: Is the video background distorted or out of place?
- Source: Who published the video, and what method was used to share it? Does the video have any logos or watermarks?
- Context: Is enough context to explain the actions happening in the clip? Does the celebrity and/or product background make sense with what is shown in the video?
VERIFY also reported that the Massachusetts Institute of Technology Media Lab has very specific tips to help spot deepfake videos, and noted that consumers can do a reverse image search to check as well.
It also recommended that people, before clicking on any suspicious links, closely verify where a link would take them by hovering over the link with their computer mouse.
Individuals who suspect a scam or an unfair business practice should contact the Ohio Attorney General’s Office at 800-282-0515.
.
About the Author