Political ads made with AI break into the midterms campaign
More than 200 commercials produced using the technology have been identified; its use in politics is not regulated at the federal level. Its electoral and social impact is yet to be seen
Political ads created using artificial intelligence (AI) have begun to appear in the midterms campaign, according to recent reports. One example is a video featuring Democratic candidate James Talarico, who is running for the U.S. Senate from Texas. The ad depicts Talarico rewriting the Book of Genesis, claiming that God was nonbinary and proposed six genders.
However, the ad is clearly fake, as it misrepresents Talarico's past statements and includes unrealistic elements like a pen and crayons drawing a crowned face onto a Bible. A notice at the bottom of the screen reveals that certain video content in the ad was generated using AI. According to a count by the Wesleyan Media Project, there are now over 200 political ads generated with AI circulating in the United States during the midterm election cycle.
Travis Ridout, one of WMP's directors, estimates that the total number of AI-generated ads surpassed 200 in early September and the spending amounted to over $100 million. The most aggressive advertising period is still ahead, and the overall campaign ad spending is estimated to be more than $10 billion, according to AdImpact. Despite the relatively small amount of money spent on AI-generated ads compared to overall campaign ad spending, experts are uncertain whether this will truly level the playing field in political contests.
It is already evident that the use of AI-generated ads is strikingly varied, ranging from entire videos and deepfakes to subtle image or sound editing processes. The count of AI-generated ads by WMP is only an approximation since identification is done manually and campaigns do not always explicitly flag them. Furthermore, it is difficult to measure material produced by individuals unaffiliated with campaigns that circulates on social media and sometimes goes viral.
The spread of AI-generated political ads raises concerns about the potential effects on voters. While a majority of Americans hold a negative view of AI's rapid advancement, many believe that AI-generated ads should carry disclosures. Small experiments conducted by Ridout's team suggest that people can easily identify ads with deepfakes, but there is still uncertainty about whether AI-made ads affect trust in political advertising overall.
The primary concern is whether seeing these ads, even knowing they are fake, still impacts how voters think and behave afterward. Currently, there is no federal legislation addressing this issue, but 31 states have enacted laws regulating AI-generated political advertising in various ways, including mandatory disclosures and bans on deepfakes.
Controversies have arisen in some states, such as Tennessee, where a law requiring AI-generated ads to include a notice at the beginning and the end of the ad has been challenged by campaigns in the 6th District and gubernatorial races.
Written by urgent.news from El Pais English's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
