Source: AP
The advancement of artificial intelligence is exponentially amplifying the threat of election disinformation globally, according to the Associated Press. Now, even individuals armed with just a smartphone and a devious imagination can easily produce convincing fake content aimed at manipulating voters.
This represents a significant leap from just a few years ago when creating deceptive photos, videos, or audio clips required extensive resources in terms of time, technical expertise, and money. However, with the accessibility of free or low-cost generative AI services provided by companies like Google and OpenAI, anyone can now generate high-quality “deepfakes” with minimal effort, merely using a simple text prompt.
The proliferation of AI deepfakes has been particularly notable in elections across Europe and Asia, serving as a stark warning for the more than 50 countries scheduled for elections this year. According to Henry Ajder, a prominent expert in generative AI based in Cambridge, England, the concern is no longer whether AI deepfakes could impact elections but rather how influential they will become. Ajder, who operates a consulting firm named Latent Space Advisory, emphasizes that the technology’s sophistication makes it increasingly difficult to discern real content from fabricated material.
As the U.S. presidential race intensifies, FBI Director Christopher Wray has recently issued a cautionary statement regarding the escalating threat, highlighting how generative AI facilitates malicious interference by foreign adversaries. AI deepfakes can be employed to tarnish or enhance a candidate’s image, manipulate voter sentiment, or even discourage participation in elections altogether. However, experts argue that the most significant danger to democracy lies in the erosion of public trust in visual and auditory information due to a surge in AI deepfakes.
Examples of recent AI deepfakes include a video depicting Moldova’s pro-Western president endorsing a Russian-friendly political party, audio clips purportedly featuring Slovakia’s liberal party leader discussing vote manipulation and increasing beer prices, and a video showing an opposition lawmaker in Bangladesh, a conservative Muslim-majority nation, wearing a bikini. The novelty and sophistication of the technology make it challenging to trace the originators of AI deepfakes, with governments and companies struggling to keep pace with the onslaught or develop effective countermeasures.
Moreover, the difficulty in verifying audio-only deepfakes exacerbates the problem, as demonstrated by instances in Slovakia where manipulated audio clips were widely disseminated before parliamentary elections. In the United States, robocalls impersonating President Joe Biden aimed to dissuade voters from participating in primary elections, underscoring the potential impact of AI deepfakes on electoral processes.
In response to the growing threat, authorities worldwide are scrambling to establish regulatory frameworks. While the European Union mandates special labeling of AI deepfakes and tech giants voluntarily commit to mitigating their spread, challenges persist, especially on platforms like Telegram, which operate outside of these agreements. Concerns also arise regarding unintended consequences, such as potential limitations on legitimate political discourse or the dissemination of false information through AI-generated content.
Ultimately, the escalating use of AI in spreading disinformation poses significant challenges to democratic processes globally, emphasizing the urgent need for comprehensive strategies to address this complex issue.
