Members of YouTube’s LGBTQ community have raised concerns about their videos being demonetized seemingly because they fall under queer categories, but CEO Susan Wojcicki says that’s not the case.

Demonetization and poor recommendation of LGBTQ content are two areas that Wojcicki spoke about at length with popular vlogger Alfie Deyes in a recent interview. Wojcicki said that YouTube does “not automatically demonetize LGBTQ content” and that YouTube’s monetization system, which determines whether a video is appropriate for ads, and its recommendation systems, which promote videos across the platform, operate independently. It’s set up that way to ensure that the “systems are fair,” she said.

“There’s no policies that say ‘If you put certain words in a title that will be demonetized.’” Wojcicki told Deyes. “We work incredibly hard to make sure that when our machines learn something — because a lot of our decisions are made algorithmically — that our machines are fair. There shouldn’t be [any automatic demonetization].”

Questions about the treatment of LGBTQ creators’ videos on the platform have been around for years. Creators like Rowan Ellis, Tyler Oakley, and Stevie Boebi vocalized their concerns and frustrations with YouTube seemingly hiding their videos and demonetizing their content in 2017. Then in June 2018, trans creator Chase Ross accused YouTube of age-gating and demonetizing his videos because he used the term “transgender” in his titles. Just a couple of months later, YouTube found itself in another controversy after anti-LGBTQ ads started appearing on videos from LGBTQ creators.

To say that it’s been difficult for the queer YouTube community to believe YouTube has their best interests at heart would be an understatement. Earlier this year, Vox host Carlos Maza called out YouTube for allowing conservative pundit Steven Crowder to remain on YouTube after Crowder made repeated homophobic remarks. Wojcicki, who apologized to the LGBTQ creator community after the incident with Maza, told Deyes “we want to support the LGBTQ community,” adding that “we have specifically created systems and processes to make sure our [machine learning] systems are fair.” 

“We have a machine-learning fairness initiative,” Wojcicki said, “to make sure that our algorithms and the ways our machines work are fair. We want to have all different voices expressing different points of views.”

That doesn’t mean everything is working as intended, as Wojcicki acknowledged. Sometimes YouTube’s machine learning tools get things wrong, as the company previously admitted.

“I want to be sensitive to people who have those concerns,” Wojcicki said. “It always comes down to the specific video and what happened in the video.”