Meta, other tech companies to label political ads that use artificial intelligence
RICHMOND, Va. (WWBT) - With less than a year before the next US presidential election, there is growing concern about using artificial intelligence in campaign ads.
“I’ve just been monitoring developments with artificial intelligence, and especially the boom this year with generative AI and the challenges that it poses, as well as with our focus on campaigns and elections, realizing that this is going to be a major threat,” said Carah Ong Whaley, the Academic Program Officer at the University of Virginia’s Center for Politics.
Ong Whaley has a background in nuclear weapons technology and teaches a class on the impact of artificial intelligence on politics and elections at UVA.
“There’s actually quite a bit of parallels in terms of the development of the nuclear weapons industrial complex, and now the development of artificial intelligence,” said Ong Whaley.
The use of AI has increased significantly in recent years, allowing voices and images to be manipulated to the extent that it is often difficult to tell what is real and what is fake.
Florida Governor and Republican presidential candidate Ron DeSantis drew criticism over the summer for showing A.I.-generated images of President Donald Trump hugging Dr. Anthony Fauci.
“This practice of using artificial intelligence, most recently displayed by Governor DeSantis, I think, is inappropriate,” said Rich Anderson, Chair of the Republican Party of Virginia. “It does beg for regulation of some sort.”
Facebook and Instagram’s parent company, Meta, says it will start requiring political ads to disclose if they used artificial intelligence beginning sometime next year.
For those that do, a label will be attached to the video to alert viewers.
Google and Microsoft are also working to increase transparency on their platforms by adding labels and watermarks to AI-generated content.
There are also efforts within the federal government to try and regulate AI-generated deep fakes in political ads before the 2024 election.
The Federal Election Commission is looking into it. President Joe Biden issued an executive order to encourage responsible development of AI last week, and lawmakers on Capitol Hill are debating legislation.
“I think we could do a lot more to require all ads disclose who generated the audio, video, or imagery and and to make clear, for example, if it’s coming from a foreign actor or a domestic actor and not just a requirement of political campaigns,” said Ong Whaley.
“I don’t think that AI ought to play any role in politics by any Republican, Democrat, Independent or any other party because politics is real,” said Anderson.
While Meta and others plan to label political ads that use AI, it is unclear what criteria they will use to vet them.
Without federal regulation, other manipulated videos can circulate across social media platforms unchecked, potentially influencing voters and impacting future elections.
“We’ve got to get a grip on this, and at the front end,” added Anderson.
Copyright 2023 WWBT. All rights reserved.