Top 4th of July Sales Best 4K Projectors 7 Early Prime Day Deals Wi-Fi Range Extenders My Favorite Summer Gadgets Cheap Car Insurance Target's 4th of July Sale Best Running Earbuds, Headphones

Ahead of Iowa caucuses, YouTube pledges to remove deepfakes, voting disinformation

The company spells out its policies on manipulated media as the 2020 election season kicks off in earnest.

YouTube headquarters in San Bruno, California
James Martin/CNET

As the Iowa caucuses kick off Monday evening, YouTube and its parent Google spelled out its policies regarding deepfake videos and other disinformation. 

The video service said it would take down "technically manipulated or doctored" videos, as well as content that tries to mislead people about voting and census issues, like when and where to vote. 

"As the 2020 election season kicks into high gear in the United States, people will visit YouTube to learn about the candidates and watch the election season unfold," Leslie Miller, vice president of government affairs and public policy at YouTube, wrote in a blog post published Monday. "Over the last few years, we've increased our efforts to make YouTube a more reliable source for news and information, as well as an open platform for healthy political discourse."

With the 2020 nomination process kicking off in earnest with the Iowa caucuses, Silicon Valley companies are still reeling about their role in the 2016 US election. Russian agents exploited YouTube, Facebook and Twitter to spread disinformation and sow discord among voters. The reputations of the companies still haven't recovered. Last week, CNET traveled through Iowa to interview locals across the state, who said Facebook is the most distrusted of all the social platforms. 

YouTube's policy regarding deepfakes isn't new, but a company spokeswoman said the announcement marks the most in-depth look at how it handles both technical and political manipulation. 

"Content that has been technically manipulated or doctored in a way that misleads users (beyond clips taken out of context) and may pose a serious risk of egregious harm," Miller wrote, regarding deepfakes. "For example, a video that has been technically manipulated to make it appear that a government official is dead."

YouTube has been quick to take down deepfakes of politicians. In May, the platform removed a video of House Speaker Nancy Pelosi that had been slowed down to make it appear as if she was drunk and slurring her words. Facebook decided to keep the video up.

The point about clips being taken out of context came to the forefront last month, when the Sen. Bernie Sanders campaign criticized former Vice President Joe Biden for his positions on social security. The Sanders campaign sent out a video of a 2018 Biden speech and said the vice president "lauded Paul Ryan for proposing cuts to Social Security and Medicare." Biden's campaign pushed back, calling the video "doctored." The video didn't appear to be altered, but it omitted details of Biden's larger argument.

When it comes to voter suppression, YouTube isn't the only one worried about false news. Sen. Elizabeth Warren's campaign  last week released a plan to deal with disinformation, calling for civil and criminal penalties for knowingly spreading false information online when it comes to when and how to vote in US elections.