Artificial intelligence (AI) has revolutionized the threat of election disinformation worldwide. With the availability of free and low-cost generative AI services, creating convincing deepfake content has become easier than ever before. Previously, phony photos, videos, and audio clips required extensive resources and technical skills, but now anyone with a smartphone and a devious imagination can create high-quality deepfakes using simple text prompts.
As more than 50 countries prepare for upcoming elections, there is a growing concern about the influence of AI deepfakes. Henry Ajder, an expert in generative AI, believes that the question is not whether AI deepfakes will affect elections but rather how influential they will be. These deepfakes can smear or soften a candidate’s image, steer voters toward or away from candidates, and even discourage people from voting altogether. However, the most significant threat that experts identify is the erosion of public trust in what they see and hear.
Recent examples of AI deepfakes in elections include a video of Moldova’s pro-Western president endorsing a Russian-friendly party, audio clips of Slovakia’s liberal party leader discussing vote rigging and raising the price of beer, and a video of an opposition lawmaker in Bangladesh, wearing a bikini. The novelty and sophistication of the technology make it challenging to track the originators of these deepfakes.
Governments and companies are struggling to address this issue effectively. The technology is evolving rapidly, making it increasingly difficult to differentiate between genuine and fake content. As the technology improves, definitive answers about the authenticity of this content become harder to come by. This challenge raises concerns over the impact of AI deepfakes on elections and democracy as a whole.
AI deepfakes aim to sow doubt in candidates’ allegiances and erode trust in electoral processes and institutions. In Moldova, for example, deepfakes have targeted the pro-Western President Maia Sandu, undermining trust in the electoral process. Similar concerns have been raised about China’s use of generative AI for political purposes in Taiwan. The Chinese Communist Party has been accused of meddling in Taiwanese politics by spreading deepfakes on platforms like TikTok.
Audio-only deepfakes are particularly difficult to detect as they lack visual cues of manipulation. In countries like Slovakia, where Russian influence looms, audio clips resembling the voice of the liberal party leader were shared widely, alleging vote rigging and beer price manipulation. Humans are more accustomed to judging with their eyes rather than their ears, making audio deepfakes more deceptive.
Experts worry that upcoming elections in India, where social media platforms are breeding grounds for disinformation, will face the threat of AI deepfakes. In poorer countries with lower media literacy, even low-quality AI deepfakes can have a significant impact on public opinion and trust.
Addressing this challenge requires global efforts. The European Union already mandates social media platforms to reduce the spread of disinformation and manipulate elections, including special labeling of AI deepfakes. While major tech companies have voluntarily committed to preventing AI disruptions in elections, much more needs to be done to combat the spread of deepfakes on platforms like Telegram.
FAQ
What are AI deepfakes?
AI deepfakes are manipulated audio, video, or images created using artificial intelligence algorithms. They can convincingly mimic real people, making it challenging to distinguish between genuine and fake content.
How do AI deepfakes impact elections?
AI deepfakes can be used to smear or soften a candidate’s image, steer voters towards or away from candidates, and erode public trust in electoral processes. They can influence public opinion and sway the outcome of elections.
Can AI deepfakes be detected?
AI deepfakes are becoming increasingly sophisticated, making detection more challenging. Audio-only deepfakes are particularly difficult to verify as they lack visual cues. However, researchers and technology developers are continually working on improving detection methods.
What measures are being taken to address the threat of AI deepfakes?
Governments and tech companies are implementing measures to combat the spread of AI deepfakes. The European Union mandates special labeling of deepfakes on social media platforms, while major tech companies have committed to preventing disruptions caused by AI tools in elections.
Are AI deepfakes a global concern?
Yes, AI deepfakes pose a global concern. With the proliferation of social media platforms and the ease of creating and sharing deepfake content, no country is immune to the potential impact of deepfakes on elections and public trust.
Sources:
What are AI deepfakes?
AI deepfakes are manipulated audio, video, or images created using artificial intelligence algorithms. They can convincingly mimic real people, making it challenging to distinguish between genuine and fake content.
How do AI deepfakes impact elections?
AI deepfakes can be used to smear or soften a candidate’s image, steer voters towards or away from candidates, and erode public trust in electoral processes. They can influence public opinion and sway the outcome of elections.
Can AI deepfakes be detected?
AI deepfakes are becoming increasingly sophisticated, making detection more challenging. Audio-only deepfakes are particularly difficult to verify as they lack visual cues. However, researchers and technology developers are continually working on improving detection methods.
What measures are being taken to address the threat of AI deepfakes?
Governments and tech companies are implementing measures to combat the spread of AI deepfakes. The European Union mandates special labeling of deepfakes on social media platforms, while major tech companies have committed to preventing disruptions caused by AI tools in elections.
Are AI deepfakes a global concern?
Yes, AI deepfakes pose a global concern. With the proliferation of social media platforms and the ease of creating and sharing deepfake content, no country is immune to the potential impact of deepfakes on elections and public trust.
Sources:
Definitions:
– Deepfake: Manipulated audio, video, or images created using artificial intelligence algorithms to convincingly mimic real people.
– Generative AI: Technology that uses artificial intelligence algorithms to generate content, such as audio, video, or images, based on given prompts or data.
– Disinformation: False or misleading information spread with the intention to deceive or manipulate.
– Erosion of public trust: The gradual decline and loss of confidence or belief in institutions, processes, or information by the general public.
– Social media platforms: Online platforms where users can create and share content, interact with others, and engage in social networking, such as Facebook, Twitter, and TikTok.
Related Links:
- Deepfakes and Democracy: The Problem of Perception
- 4 Ways to Protect Your Elections from Deepfakes
- Are deepfakes a threat to the 2022 Indian elections?
The source of the article is from the blog guambia.com.uy