AI-generated images and videos could wreak havoc as the legal field struggles to keep up with emerging technology.
By Brian Kabateck
Doctored images are nothing new. The first known manipulated photo dates back to the mid-1800s, only a few years after the invention of photography itself.
Back in the day, manipulating photos meant cutting, pasting, layering, and piecing together photographs and retouching by hand with ink or paint.
Digital manipulation of images became possible in the early 1990s. Almost as soon as digital cameras were available to consumers, you could edit your photos at home.
But today, there’s a brave new world: deepfake digital images, video, and audio.
What is Deepfake?
Deepfake is synthetic media created through Artificial Intelligence (AI). As defined by Tech Target, “Deepfake AI is a type of artificial intelligence used to create compelling images, audio, and video hoaxes. The term describes both the technology and the resulting bogus content and is a portmanteau of deep learning and fake.”
Today you don’t even need an actual, original photograph to manipulate. A few words typed into an AI generator will prompt it to create almost any image you desire—it can produce a high-quality “photo” of a person who doesn’t even exist. Deepfake videos use “deep learning” (machine learning based on neural networks and numerous layers of processing that extract progressively more detailed data) and sophisticated face-swapping technology to create realistic videos of people saying and doing things that never happened.
Even experts need help to identify the forgeries. Deepfake technology is advancing at breakneck speed—far outpacing the technology that can identify counterfeit media.
The legal system is bracing itself for the storm that a proliferation of phony images and footage will almost inevitably bring.
Visual and audio evidence that substantiates claims (or appears to) is enormously impactful in court proceedings. It’s disturbing to consider how forged images and videos could complicate and potentially compromise everything from parking violations to headline-making criminal cases, even resulting in wrongful convictions.
However, as Riana Pfefferkorn, a Stanford Internet Observatory research scholar, points out. At the same time, deepfake evidence is a legitimate threat, and the court has a long history of being robust and resilient against tampering. Public distrust is more problematic for the legal system when you can’t believe your eyes and ears. Just the knowledge that AI can create such persuasive images calls into question everything we thought we knew about what we believe. It enables people to leverage this doubt in court—with accusations that objective evidence is fake.
This scenario has already happened in some high-profile cases. When the family of a man killed in a Tesla accident sued the company, Elon Musk’s lawyers argued that a 2016 video recording of Musk championing the car’s self-driving feature could be a deepfake. Two men charged with involvement in the Capitol riot of January 6, 2021, claimed that footage of the events may have been AI-generated.
As Hany Farid, professor at UC Berkeley School of Information and a digital forensics expert, recently told NPR, “That’s exactly what we were concerned about: That when we entered this age of deepfakes, anybody can deny reality.”
The implications for the legal process could be far-reaching—juries may demand more proof of the authenticity of evidence; getting evidence admitted will become more time-consuming and laborious, prolonging trials and adding expense; parties with fewer resources will face the most significant challenge. Trust in the judicial system could erode.
While the court catches up to technology, Pfefferkorn says, ethics and professional norms—due diligence such as lawyers not making frivolous arguments—will play an essential role in minimizing the damage AI could do. She emphasizes that lawyers must educate themselves to spot deepfakes, put forensic experts in their budget, and do due diligence with evidence the client brings.
Kabateck LLP is a nationally renowned plaintiff’s firm that litigates complex cases with an impressive record of success. We have recovered over $2 Billion in verdicts and settlements for our clients. These notable victories have significantly impacted the legal community and resulted in better consumer protections in personal injury, insurance bad faith, pharmaceutical litigation, wrongful death, class action, mass torts, and disaster litigation.