Sign up for our daily and weekly newsletters to stay updated with the latest and exclusive content on leading AI developments.
Concerns about the misuse of AI in the 2024 U.S. elections are on the rise, but this hasn’t deterred AI voice cloning startups from delving into the political arena. For instance, Instreamatic, an AI audio/video advertising platform based in Boca Raton, Florida, is expanding its services into political advertising. This platform raised $6.1 million in a Series A funding round in 2021. Their technology allows political campaigns to create highly-targeted, AI-driven contextual ads in video and audio formats. These ads feature voiceovers instead of talking head videos and can adapt to various events or locations in real-time.
A demonstration video by Instreamatic showed how a candidate, like Barack Obama, can change any audio or video political ad by cloning their voice without needing to re-record in a studio. Since last March, Instreamatic has been offering its generative voice AI product to brands and agencies, boasting that from a single ad, their AI can create unlimited ad versions tailored to different audiences based on factors like location, time of day, the platform where the ad is seen, or the nearest store.
However, experts warn that AI’s role in the 2024 U.S. election campaigns could create a disinformation minefield. For example, an ABC News report highlighted summer efforts by the campaign of Florida Governor Ron DeSantis, which included AI-generated images and audio of former President Donald Trump. Nathan Lambert, a machine learning researcher at the Allen Institute for AI, noted that generative AI could turn the 2024 elections into a chaotic affair.
Stas Tushinskiy, CEO and co-founder of Instreamatic, assured that the company has measures in place to prevent its product from being used for election disinformation. He explained that clients must confirm they have permission to use any voice in their campaigns, and the political advertising feature is not available to everyone. Tushinskiy emphasized that the company would closely manage campaign creation to avoid misuse and would swiftly remove any problematic political ads and issue public safety statements if necessary.
Tushinskiy also pointed out that Instreamatic is not reinventing political ads but automating an existing manual process. He described how traditionally, candidates or voice talents spend hours in studios recording ads, which then undergo various stages of checking and uploading. Instreamatic’s technology compresses this six to eight-week process into mere minutes. Additionally, it reduces the back-and-forth between agencies and clients regarding script changes and new recordings.
Concerns about AI in political ads remain high, with experts noting the lack of federal regulations for AI-generated content in campaigns. Russell Wald, policy director at Stanford University’s Institute for Human Centered AI, remarked on the absence of clear rules, indicating that campaigns are essentially setting their own guidelines as they proceed. Tushinskiy reiterated that Instreamatic seeks to enhance campaign effectiveness, and the company is committed to stopping and exposing any unethical use of its technology.
Stay informed by subscribing to our daily newsletter for the latest news.