AI’s Minimal Impact on the 2024 US Election: Old Misinformation Tactics Prevail
The 2024 US presidential election was anticipated to be the first significantly influenced by the widespread availability of artificial intelligence (AI) tools for generating synthetic media. Fears of AI-generated deepfakes manipulating voters and disrupting the democratic process were rampant. Early actions, such as the FCC’s ban on AI-generated robocalls following a deceptive call mimicking President Biden’s voice, and the implementation of AI-related legislation in sixteen states, seemed to presage a wave of AI-driven election interference. However, the anticipated deluge of AI-generated misinformation failed to materialize. Instead, traditional forms of misinformation dominated the landscape.
While concerns about AI’s potential to create deepfakes and spread misinformation were valid, the actual impact of AI on the 2024 election was surprisingly limited. Experts agree that the election was not significantly shaped by AI-generated content. Rather, traditional misinformation tactics, including text-based social media posts and deceptively edited videos and images, remained the primary vectors for spreading false narratives. While some AI-generated content appeared, it did not achieve widespread influence or significantly alter the course of the election.
The AI-generated content that did gain traction primarily reinforced existing narratives, rather than introducing entirely new and fabricated claims. One example is the spread of AI-generated images and memes depicting animal abuse following false claims by Donald Trump and his running mate, JD Vance, about Haitians harming animals in Ohio. These visuals amplified an existing, false narrative rather than creating a wholly new one. This suggests that the most effective disinformation campaigns still rely on exploiting existing societal biases and anxieties.
Several factors contributed to the muted role of AI in the election. The proactive efforts by technology companies and policymakers played a crucial role in mitigating potential harm. Platforms like Meta and TikTok implemented policies requiring disclosure of AI use in political ads and labeling AI-generated content. OpenAI prohibited the use of its tools for political campaigns and blocked the creation of images of real people. These measures likely discouraged widespread use of AI for malicious purposes.
Additionally, the effectiveness of traditional disinformation methods may have reduced the incentive for malicious actors to invest in developing and deploying AI-generated content. Prominent figures with large followings on social media platforms could readily spread misinformation without relying on AI-generated media. The ease and effectiveness of these traditional methods possibly overshadowed the potential benefits of utilizing AI for disinformation campaigns.
Furthermore, the focus on identifying and debunking AI-generated content also likely contributed to its limited impact. Experts in digital media forensics were prepared to analyze and expose deepfakes and other forms of AI-generated manipulation. This vigilance, combined with public awareness campaigns, may have deterred the widespread use of such content.
While AI did not play a major role in spreading misinformation during the 2024 election, it did surface in specific instances. The AI-generated robocall imitating President Biden’s voice serves as a prominent example of AI’s potential for abuse. However, this incident remained isolated and did not trigger a wave of similar attacks. It did, however, highlight the relative ease with which such content can be created and the potential for future misuse.
Another notable use of AI was in creating content designed to stoke partisan animosity. Deepfakes, while not necessarily aimed at spreading outright false information, were used to reinforce existing negative perceptions of candidates. This use of AI, while not as direct as spreading fabricated claims, highlights the potential for AI to be used to manipulate public opinion and deepen political divisions.
Foreign interference operations, a significant concern in recent elections, also did not rely heavily on AI-generated content. While intelligence agencies identified foreign influence campaigns, these efforts primarily employed actors in staged videos rather than AI-generated media. This suggests that, at least for now, human-driven disinformation campaigns remain more effective and easier to deploy than those relying on sophisticated AI technology.
The proactive measures taken by technology platforms, coupled with state legislative efforts, likely played a crucial role in curbing the worst potential abuses of AI in the 2024 election. While some loopholes and limitations in these safeguards were identified, the overall effect was positive. The vigilance of experts and the public, combined with the efforts of platforms and policymakers, contributed to a relatively AI-free election landscape.
Despite the limited impact in 2024, the potential for AI to be used for malicious purposes in future elections remains a significant concern. As AI technology continues to evolve and become more accessible, the threat of sophisticated deepfakes and other AI-generated misinformation will likely increase. Continuous efforts to develop detection technologies, educate the public, and implement robust platform policies will be essential to safeguarding the integrity of future elections. The 2024 election serves as a valuable learning experience, demonstrating the importance of proactive measures to counter the evolving threat of AI-powered disinformation.
