LONGWORTH: YOU’VE PROBABLY HEARD ONE OF THESE…

SONGS THAT SOUND LIKE A POPULAR ARTIST, BUT THEY’RE ACTUALLY AI-GENERATED.

WELL, NOW YOUTUBE IS CRACKING DOWN ON THEM…ANNOUNCING IT’LL GIVE MUSIC LABELS AND DISTRIBUTORS THE ABILITY TO REQUEST THAT SYNTHETIC CONTENT MIMICKING AN ARTIST’S UNIQUE SINGING OR RAPPING VOICE BE REMOVED. THE PLATFORM WILL CONSIDER THE CONTENT’S PURPOSE–WHETHER IT’S PART OF NEWS REPORTING, ANALYSIS OR CRITIQUE.

THIS MOVE IS PART OF WHAT YOUTUBE’S CALLING ITS “APPROACH TO RESPONSIBLE AI INNOVATION.”

IN THE COMING MONTHS, THERE’LL BE SEVERAL CHANGES.

THE PLATFORM WILL:

REQUIRE CREATORS TO DISCLOSE AI-GENERATED CONTENT

REQUIRE CREATORS TO DISCLOSE WHEN THEY’RE UPLOADING SYNTHETIC OR ALTERED CONTENT THAT LOOKS REALISTIC.

LABEL AI-GENERATED CONTENT

YOUTUBE WILL LABEL THAT CONTENT TO VARIOUS EXTENTS.

REMOVE SYNTHETIC CONTENT VIOLATING COMMUNITY GUIDELINES

AND IF SYNTHETIC CONTENT VIOLATES COMMUNITY GUIDELINES–LET’S SAY IT SHOWS REALISTIC VIOLENCE–IT’LL BE REMOVED

WE ARE SEEING OTHER POLICIES FROM BIG TECH COMPANIES, ADDRESSING CONCERNS ABOUT AI-GENERATED CONTENT MISLEADING USERS.

JUST LAST WEEK, META ANNOUNCED IT’S BANNING ADVERTISERS FROM USING ITS OWN AI PRODUCT TO CREATE ADS ABOUT POLITICS, SOCIAL ISSUES, AND MORE. THE COMPANY IS REQUIRING DISCLOSURE FOR SUCH ADS GENERATED WITH THIRD-PARTY AI TOOLS.