How deep is your fake: The use of deepfakes in elections

Deepfakes have emerged as a potent tool with the capacity to distort reality in unprecedented ways. The term deepfake is a combination of “deep learning” and “fake media”.

This technology, which employs advanced artificial intelligence to create hyper-realistic fake videos and audio, has profound implications for many sectors, but none more critical than the democratic process itself. As we move closer to upcoming elections worldwide, the potential misuse of deepfakes poses a significant threat to the integrity of our electoral systems.

The evolution of disinformation

Disinformation is not a new phenomenon. For decades, political campaigns have utilized various forms of media manipulation to sway public opinion.

However, the advent of deepfake technology marks a dangerous escalation. Unlike traditional forms of disinformation, which can often be identified through critical analysis and fact-checking, deepfakes blur the lines between truth and fiction to a degree that makes detection exceedingly difficult.

Deepfakes can create convincingly realistic videos of political figures saying or doing things they never actually said or did. This not only confuses the electorate but can also be used to undermine candidates, spread disinformation, and erode trust in public institutions. The implications are clear: in the hands of malicious actors, deepfakes can significantly disrupt democratic processes by misleading voters and manipulating election outcomes.

The power of visual persuasion

Humans are inherently visual creatures. Studies have shown that we are more likely to believe and be influenced by visual information than text-based data.

This psychological predisposition makes deepfakes particularly potent. A well-crafted deepfake can exploit our trust in visual evidence, making it a powerful tool for spreading false narratives.

During elections, this can manifest in several ways. Imagine a deepfake video of a candidate making racist remarks, or another falsely showing a politician accepting a bribe. Such content, if disseminated widely through social media and other platforms, could irreparably damage a candidate's reputation and alter the course of an election. The stakes are high, and the consequences for democracy are dire.

Real-world examples

In Indonesia, then-President-elect Prabowo Subianto's campaign demonstrated the influential power of digital manipulation. His team utilized TikTok and AI-generated art to sway voters, effectively reaching and engaging younger demographics. These tools allowed the campaign to craft a compelling narrative and visually appealing content that resonated with a broad audience, showcasing the strategic use of AI in modern electoral politics.

Similarly, the infamous Cambridge Analytica scandal highlighted the impact of micro-targeting in elections. By harvesting data from millions of Facebook users without their consent, Cambridge Analytica was able to create highly personalized political advertisements. This strategy played a significant role in influencing voter behavior during the 2016 US elections, underscoring the profound effects of targeted digital manipulation.

Combating the deepfake menace

Addressing the threat posed by deepfakes requires a multi-faceted approach.

Firstly, technological solutions are paramount. Advances in deepfake detection technology are being made, with AI systems increasingly capable of identifying manipulated media. However, this is a constant arms race; as detection methods improve, so too do the techniques used to create deepfakes.

Secondly, legislative measures must be implemented. Governments worldwide need to recognize the threat and enact laws that specifically address the creation and dissemination of deepfakes, particularly in the context of elections. Penalties for those found guilty of using deepfakes to influence political outcomes should be severe enough to act as a deterrent.

Thirdly, public awareness and education are crucial. Voters must be made aware of the existence and potential impact of deepfakes. Media literacy programs should be promoted to help the public critically evaluate the content they consume and recognize signs of manipulation.

The role of social media platforms

Social media companies have a significant role to play in combating the spread of deepfakes. These platforms are often the primary means through which deepfakes are disseminated.

As such, they must invest in robust detection mechanisms and implement strict policies to remove deepfake content promptly. Moreover, transparency is key; social media platforms should work closely with independent fact-checkers and provide users with clear information about why certain content has been flagged or removed.

A call to vigilance

The rise of deepfake technology presents a formidable challenge to the integrity of democratic elections.

As we navigate this new digital landscape, vigilance is essential. Technological advancements, legislative action, public education, and responsible practices by social media platforms are all critical components in the fight against the misuse of deepfakes.

Democracy relies on informed and rational decision-making by the electorate. To preserve the sanctity of our elections, we must collectively address the deepfake menace with urgency and determination. Only then can we ensure that the democratic process remains fair, transparent, and resilient in the face of evolving technological threats.

Dominic Ligot is the founder, CEO and CTO of CirroLytix, a social impact AI company. He also serves as the head of AI and Research at the IT and BPM Association of the Philippines (IBPAP), and the Philippines' representative to the Expert Advisory Panel on the International Scientific Report on Advanced AI Safety.