“Fake News” is a buzz term that could encompass the 2016 election. Spawned during the onslaught of legitimately made-up stories designed to go viral on social media and impact public opinion, fake news continues to run rampant across social media platforms. It’s an area that outlets like Twitter and Facebook have verbally committed to tamping down, but as we begin the run up to the 2020 election, something just as insidious is already rearing its ugly head: deepfakes.

If you’re unfamiliar with what a deepfake is, there’s a chance you may have already fallen for one. Deepfakes are altered videos that utilize artificial intelligence to make them appear genuine—and many of them are incredibly successful in that regard. Last week an edited video of House Speaker Nancy Pelosi began circulating on social media sites, edited and spliced in such a way that it made her appear to slur her words.

While not a top-tier deepfake, it served its purpose as a tool of disinformation—and made many headlines, as well. When legitimate news outlets are put into a position of letting their viewers know that something is false, it means that the narrative has already seeped into the cultural groundwater.

The fear going forward, according to The Washington Post columnist Brian Klaas, is that this is only the beginning, and these fake-out videos will get much better in quality. Videos edited to show people saying things that they never did; videos designed to destroy someone’s reputation—it’s a frightening but very real possibility in our social media driven world.

If it was difficult enough to snuff out fake news stories in text, well, just think of how quickly falsehoods will spread when they’re on video. So as always—and especially now—be sure to vet and verify everything that you see on social media. Fake articles are pervasive enough, but fake videos are especially terrifying.