Following numerous reports on how decidedly creepy content aimed at children is getting past YouTube’s filters, the video platform has once again vowed to take action.

The videos in question depict real children’s cartoon characters, such as Peppa Pig, involved in vulgar or violent acts. The videos masquerade as content suitable for kids and find themselves appearing next to actual children’s content. Some of these videos have even made it onto YouTube Kids, a safe-surfing channel for young children.

YouTube has already taken steps to make YouTube Kids safe by giving parents more options to customize the channel, and later also added more protection features so that inappropriate videos anywhere on the platform could more easily be flagged. YouTube has admitted that “no system is perfect” but called such content “the extreme needle in the haystack.” Such comments were largely seen as feeble.

YouTube has now promised to ramp up its efforts to either delete vulgar content that seems to be aimed at kids, or better compartmentalize such content. YouTube announced more updates on Wednesday.

Johanna Wright, vice president of product management at YouTube, said that since June it had removed ads from 3 million videos that were violent, offensive or inappropriate and that masqueraded as family content. “We’ve further strengthened the application of that policy to remove ads from another 500K violative videos,” Wright said.

Vulgar comments on children’s videos have also been a problem, and YouTube has taken steps to weed them out. “Comments of this nature are abhorrent and we work with NCMEC to report illegal behavior to law enforcement,” said Wright. From now on, comments will be turned off on videos where offensive language has been used.

Just in case creators don’t know what suitable content for kids is, YouTube has also written up a set of guidelines. At the same time, the platform admitted that some content is nuanced, and while aimed at adults could mistakenly find itself among children’s videos. YouTube said it has now doubled its number of “Trusted Flaggers” and is mulling over ways to better treat such content.

“We’re wholly committed to addressing these issues and will continue to invest the engineering and human resources needed to get it right,” said Wright. “As a parent and as a leader in this organization, I’m determined that we do.”

The updates will starting taking shape over the next weeks and months.

Image: Philippe Put via Flickr