Political campaigns nationwide spent over $137 million on computer-generated images of events. Some ads were comical, others were polarizing.
More campaigns seem to be using AI-generated images or deepfakes — or am I just just noticing it? Yes, AI-generated content, often referred to as 'AI slop,' is increasingly common in political campaigns. The primary concern extends beyond simply fake images to the strategic use of AI for 'propaganda-type targeting.' This involves creating highly personalized content based on an individual's psychology and data, streamlining a process that previously required multiple steps, as seen with organizations like Cambridge Analytica. This immediate, tailored content delivery poses a significant ethical challenge in elections. Are you saying that’s because the algorithm automatically delivers to an individual voter an ad that’s going to be tailored to them? Exactly. AI systems can now gather and access extensive data about individuals, including their online behavior and even credit card records. This data is then used to generate and deliver content specifically designed to influence what a voter sees or does not see, leveraging their unique psychological profile. This real-time, personalized targeting raises substantial privacy and ethical questions, especially due to the lack of transparency regarding whether content is AI-generated, prompting calls for mandatory watermarking. What do you mean by watermark and why? Watermarking refers to embedding a persistent label or mark, such as 'AI generated,' directly onto an image or piece of content. This mechanism serves to disclose the origin of the content, indicating that it was created by artificial intelligence rather than a human. This is crucial for transparency, as much of the AI-generated political content observed on social media platforms like TikTok and Instagram before elections was not flagged, making it difficult for the public to discern its authenticity and source. The Steve Hilton AI ads often had a line saying they were AI-generated. While some political campaigns, like that of gubernatorial candidate Steve Hilton, did include disclosures indicating their ads were AI-generated, this practice is not universal. The broader and more critical issue remains that citizens need clear and consistent information to determine whether the content they encounter online is 'real' or human-created. As digital platforms increasingly shape public perception and information consumption, the absence of standardized AI disclosure mechanisms can lead to widespread confusion and distrust in political messaging. Are voters being fooled by deepfakes? Are they confused? In many instances, it is practically impossible for voters to distinguish between genuine, human-created content and sophisticated AI-generated deepfakes. This inability to differentiate is not a failure on the part of the voters but rather a testament to the advanced capabilities of AI. While some AI content, termed 'AI slop,' is intentionally surreal and not meant to deceive, other deepfakes are designed for believability, eroding public trust and creating a state of confusion regarding the authenticity of political information. What ends up being effective though? Does AI convince people to vote for a candidate? Initial research indicates that there's no significant evidence people can reliably tell the difference between highly realistic AI-generated content and human-created content. Therefore, any content, whether real or AI-generated, that influences how or if people vote, can be effective. This poses a fundamental challenge to the functioning of a democracy. Micro-targeting, exacerbated by AI, fragments public perception, leading to vastly different 'realities' for different voters. A shared basis of truth and reality is essential for collective governance and agreement, which AI-driven micro-targeting undermines. Are people on both sides of the political spectrum doubting everything they see online? Yes, skepticism about online content is widespread across the political spectrum. The expert highlights that technology's true purpose should be to augment human capabilities, not replace them. The current trajectory of AI in politics, particularly its use for propaganda and micro-targeting, raises concerns about diminishing human agency in democratic processes. The danger lies in machines dictating thoughts and behaviors, rather than serving as tools for exploration and understanding, thus jeopardizing foundational democratic principles. What role might AI play in the 2028 elections? The critical question for the 2028 elections revolves around leadership that can effectively navigate the persistent presence of AI technologies. The challenge is to leverage AI for productive purposes, such as expanding understanding and exposure to the wider world, as tools rather than replacements for human intellect. This necessitates significant effort to regulate AI platforms and companies, ensuring they adhere to fundamental standards of transparency, such as watermarking AI-generated content and fairly compensating individuals for their data, thereby safeguarding democratic processes.
More campaigns seem to be using AI-generated images or deepfakes — or am I just just noticing it?
Yes, AI-generated content, often referred to as 'AI slop,' is increasingly common in political campaigns. The primary concern extends beyond simply fake images to the strategic use of AI for 'propaganda-type targeting.' This involves creating highly personalized content based on an individual's psychology and data, streamlining a process that previously required multiple steps, as seen with organizations like Cambridge Analytica. This immediate, tailored content delivery poses a significant ethical challenge in elections.
Are you saying that’s because the algorithm automatically delivers to an individual voter an ad that’s going to be tailored to them?
Exactly. AI systems can now gather and access extensive data about individuals, including their online behavior and even credit card records. This data is then used to generate and deliver content specifically designed to influence what a voter sees or does not see, leveraging their unique psychological profile. This real-time, personalized targeting raises substantial privacy and ethical questions, especially due to the lack of transparency regarding whether content is AI-generated, prompting calls for mandatory watermarking.
What do you mean by watermark and why?
Watermarking refers to embedding a persistent label or mark, such as 'AI generated,' directly onto an image or piece of content. This mechanism serves to disclose the origin of the content, indicating that it was created by artificial intelligence rather than a human. This is crucial for transparency, as much of the AI-generated political content observed on social media platforms like TikTok and Instagram before elections was not flagged, making it difficult for the public to discern its authenticity and source.
The Steve Hilton AI ads often had a line saying they were AI-generated.
While some political campaigns, like that of gubernatorial candidate Steve Hilton, did include disclosures indicating their ads were AI-generated, this practice is not universal. The broader and more critical issue remains that citizens need clear and consistent information to determine whether the content they encounter online is 'real' or human-created. As digital platforms increasingly shape public perception and information consumption, the absence of standardized AI disclosure mechanisms can lead to widespread confusion and distrust in political messaging.
Are voters being fooled by deepfakes? Are they confused?
In many instances, it is practically impossible for voters to distinguish between genuine, human-created content and sophisticated AI-generated deepfakes. This inability to differentiate is not a failure on the part of the voters but rather a testament to the advanced capabilities of AI. While some AI content, termed 'AI slop,' is intentionally surreal and not meant to deceive, other deepfakes are designed for believability, eroding public trust and creating a state of confusion regarding the authenticity of political information.
What ends up being effective though? Does AI convince people to vote for a candidate?
Initial research indicates that there's no significant evidence people can reliably tell the difference between highly realistic AI-generated content and human-created content. Therefore, any content, whether real or AI-generated, that influences how or if people vote, can be effective. This poses a fundamental challenge to the functioning of a democracy. Micro-targeting, exacerbated by AI, fragments public perception, leading to vastly different 'realities' for different voters. A shared basis of truth and reality is essential for collective governance and agreement, which AI-driven micro-targeting undermines.
Are people on both sides of the political spectrum doubting everything they see online?
Yes, skepticism about online content is widespread across the political spectrum. The expert highlights that technology's true purpose should be to augment human capabilities, not replace them. The current trajectory of AI in politics, particularly its use for propaganda and micro-targeting, raises concerns about diminishing human agency in democratic processes. The danger lies in machines dictating thoughts and behaviors, rather than serving as tools for exploration and understanding, thus jeopardizing foundational democratic principles.
What role might AI play in the 2028 elections?
The critical question for the 2028 elections revolves around leadership that can effectively navigate the persistent presence of AI technologies. The challenge is to leverage AI for productive purposes, such as expanding understanding and exposure to the wider world, as tools rather than replacements for human intellect. This necessitates significant effort to regulate AI platforms and companies, ensuring they adhere to fundamental standards of transparency, such as watermarking AI-generated content and fairly compensating individuals for their data, thereby safeguarding democratic processes.