Altered Pelosi video spotlights new front in disinformation war

Video of a speech by Nancy Pelosi went viral after it was altered to make the House Speaker appear intoxicated, showing the growing risk of accepting what you see online at face value and highlighting a trend with the potential to disrupt the 2020 presidential election.

The footage, taken from an event at the Center for American Progress, was slowed so the California Democrat appeared to be slurring her words. While backlash was swift, the proliferation of views — more than 2.8 million on Facebook — spotlighted the power of “deepfakes,” videos and images digitally altered using artificial intelligence.

“It’s always been the case that we assumed that seeing is believing, and now that’s no longer the case,” said David Doermann, a professor at the University at Buffalo and former program manager at the Defense Advanced Research Projects Agency’s media forensics program. “We’re moving in a direction where we’re going to be, as individuals, as citizens of the United States, faced with misinformation in terms of visual content.”

While experts have had the ability to manipulate images and videos for years, it typically took time and required expensive software. Today, a “computer science freshman can do these types of things on their laptop overnight,” Doermann said. “That, combined with social media and ways of spreading this stuff very quickly before anybody can discount it,” is concerning, he added.

[Opinion: Fake news, political grifting, and the extraordinary ineptitude of Jacob Wohl]

The technology raises fresh challenges for social media companies such as Facebook, Twitter, and YouTube, which are attempting to safeguard their platforms from disinformation campaigns like the one intelligence agencies say was waged by Russia during the 2016 election. The firms have been simultaneously pushed by lawmakers to block harmful content and criticized for what GOP lawmakers claim is censorship of conservative opinions.

In the case of the Pelosi video, Facebook acknowledged the distortion but allowed the footage to remain on the site, along with linked articles explaining that it had been altered. The company said in a statement that it has no “policy that stipulates that the information you post on Facebook must be true.”

Pelosi lambasted the decision, accusing the tech giant of willingly participating in the spread of misinformation.

“We have said all along, ‘Poor Facebook, they were unwittingly exploited by the Russians,’” she told a California public radio station in a recent interview. “I think wittingly, because right now they are putting up something that they know is false.”

[Related: Yang gang: The memes powering one longshot Democrat’s unlikely ascent to the debate stage]

Deepfakes first entered the public sphere in 2017, and the U.S. intelligence community warned in this year’s Worldwide Threat Assessment that fraudulent videos could be leveraged against the U.S. The technology makes it easier to manipulate elections, erode trust in institutions, and aggravate social divisions since it plays on the tendencies of people to believe what they see and hear.

Doctored videos that spread fraudulent claims can “inject mistaken beliefs about questions of policy in the electoral process,” Bobby Chesney, a professor at the University of Texas School of Law, and Danielle Citron, a law professor at the University of Maryland, wrote in a 2018 research paper.

“The potential to sway the outcome of an election is quite real, particularly if the attacker is able to time the distribution such that there will be enough window for the fake to circulate but not enough window for the victim to debunk it effectively,” they wrote. “Such interventions are bound to tip an outcome sooner or later — and in a larger set of cases, they will at least cast a shadow of illegitimacy over the election process itself.”

The video of Pelosi wasn’t a deepfake, since the words and images weren’t replaced, only changed in speed and pitch. Its spread, however, showed how such content can reach those closest to the highest echelons of the federal government.

Rudy Giuliani, President Trump’s lawyer and a former New York City mayor, shared the manipulated video on Twitter, writing, “What is wrong with Nancy Pelosi?”

“Her speech pattern is bizarre,” Giuliani tweeted. The post was later deleted, and Giuliani said he didn’t know the video had been doctored.

The willingness by people “who should know better” to promote such content because it furthers an agenda is perhaps as concerning as the altered video itself, Doermann said.

“Within months we’re going to be able to see this deepfake technology continue to grow,” Rep. Will Hurd, R-Texas, said during a recent interview with CBS News. “And we’re going to see that more, and we’re not prepared.”

Related Content