September 7th, 2023: Starting this November, Google will require political advertisers to openly say if their ads were made using artificial intelligence (AI).
The new rule says that any ad about elections needs to be clear if it has “synthetic content.” This means content that looks real but is made by a computer.
The rule includes ads that change how a person looks or sounds to make it seem like they’re doing or saying something they didn’t.
For example, if an ad uses AI to make a person look like they’re saying something they never said, it will need a label. This label must be easy to see and read. It should say things like “This audio was computer generated” or “This image is not from a real event.”
Google said that small changes, like making an image brighter or fixing red eyes, won’t need a label. Allie Bodack, a Google spokesperson, stated that because AI tools for making synthetic content are becoming more common, Google is updating its rules.
Some political groups are already using AI in their ads. For instance, the Republican National Committee has used AI to create images targeting President Joe Biden. Florida’s Governor Ron DeSantis also used AI to create images in his ads.
This decision by Google comes after some lawmakers said they are worried about these ads.
Representative Yvette Clarke (D-NY) even made a bill that asks for labels on political ads with AI content. The Federal Election Commission is also considering limiting these kinds of ads.
Google’s new rule will apply to all kinds of content: images, videos, and audio. The company said that if the ad has something fake that doesn’t matter to what the ad is saying, it doesn’t need a label.
