Exclusive: OpenAI endorses bill banning deceptive AI in election ads
Illustration: Natalie Peeples/Axios
OpenAI is endorsing a bill that would ban deceptive AI material in political advertising, the first time the AI giant is throwing its support behind election-related legislation, per an announcement shared first with Axios.
The big picture: The company is backing Sen. Amy Klobuchar's bipartisan Protect Elections from Deceptive AI Act ahead of its Wednesday markup in the Senate Rules Committee.
- The bill would ban AI-generated audio, images or video related to federal candidates in political advertisements that is purposely misleading, with certain exceptions for parody and satire.
- Klobuchar spoke with Axios this week about her goal to get election-related AI bills passed before the fall.
Between the lines: OpenAI, backed by Microsoft, has been the dominant company in the conversation around generative AI and has been an early supporter of legislative efforts on the technology.
- Those sorts of endorsements garner OpenAI goodwill among lawmakers, and also ensure the company is part of any conversations about how the bills will work.
What they're saying: "We do not want our technology — or any AI technology — to be used to deceive voters and we believe this legislation represents an important step to addressing this challenge in the context of political advertising," said OpenAI spokesperson Liz Bourgeois.
- OpenAI also recently announced as part of its election transparency work it would allow researchers to access a tool to help identify images created with the company's image generator, DALL·E 3.
