The Rising Threat of Deepfakes in Democratic Elections 

The last years, deepfakes have imposed new threats to democratic processes worldwide. This kind of technology uses AI or other technological tools, to manipulate images, audio, or videos, creating highly convincing fake content. As deepfakes become more and more accessible, over time, concerns over election integrity and public trust in media have risen significantly. Singapore, known for its forward-thinking, has recently passed a “deepfake ban” legislation; a bold step towards eliminating manipulated content that could mislead voters, damage reputations, and influence electoral outcomes. This pivotal moment raises important questions about how the European Union will react, as it faces similar threats but lacks, for now, a comprehensive legislation specifically targeting deepfakes on the electoral process.

Singapore’s Legislative Approach

On 15^th of October 2024, the “Elections (Integrity of Online Advertising) (Amendment) Bill” was tabled by the Ministry of Digital Development and Information of Singapore. The bill imported provisions designed to limit the use of synthetic media in ways that threaten election integrity. Essentially, the legislation criminalizes the deliberate creation or even dissemination of manipulated content that portrays a political candidate saying or doing something that they did not, in fact, say or do. The bill includes strict penalties for violators, provisions for immediate content takedown, and mechanisms allowing authorities to issue correction notices to clarify false information. Citizens can report suspected deepfakes directly to authorities, initiating an official review process to verify whether content violates the law. Once flagged, platforms and websites hosting the content are legally obligated to remove it within a specified timeframe, to minimize potential public misperception or harm to electoral integrity. The legislation also empowers the Singaporean government to issue correction notices, which clarify that a piece of media is fake, providing accurate information to counteract the manipulated content. 

Public response to this bill, has been generally supportive, with many citizens valuing the government’s commitment to tackling potential threats to the democratic process. On the other hand, critics express concerns over the potential for censorship or unfair enforcement of the law, as it can be really challenging to distinguish harmful deepfakes from satirical or parody content. However, the government’s correction mechanisms are intended to find a balance between protecting the democratic process and respecting individual freedoms.

Balancing Free Speech and Election Security: The EU’s Unique Challenge

In Europe, concerns about the risks posed by deepfakes are similarly growing. The European Union has taken steps to address misinformation through frameworks like the Digital Services Act (DSA) and the Code of Practice on Disinformation, but there is nothing directly addressing deepfakes’ use. Some European countries, like France and Germany, have strict laws against hate speech and defamation that can be applied to disinformation; however, these do not explicitly cover the manipulation capabilities of deepfake technology.

 A unified stance of Europe, like Singapore’s “deepfake ban”, could close regulatory gaps, enabling the EU to act cohesively in the face of evolving disinformation tactics, while restoring public trust in digital information and fostering a more transparent electoral environment. However, the EU’s struggle to maintain a unanimous approach highlights the difficulty of applying regulatory standards across diverse political, and cultural landscapes while respecting democratic principles. EU members have a strong commitment to freedom of speech and privacy rights and have different tolerances for government intervention in media. Having said that, concerns around government overreach might arise, as the power to regulate digital content could be perceived as a threat to civil liberties, particularly in countries with histories of state censorship. Even with today’s technology it is not that easy to distinguish harmful deepfakes against innocuous content. Consequently, while Singapore’s model is promising, adapting it to the EU context would require substantial adjustments to accommodate European values, diverse legal frameworks, and political and cultural landscapes; possibly rendering the adoption of a similar “deepfake ban” ineffective. 

Prospects for European Policy 

Overall, Singapore’s deepfake ban sets a valuable example for proactive legislation, although its applicability to the European Union remains complex. Europe must balance the need for robust election security with the preservation of fundamental rights. While a ban on synthetic media in elections could contribute to protecting the integrity of the democratic process, it will be critical for EU policymakers to address potential risks to free speech and ensure fair enforcement. Future EU legislation could consider a hybrid approach, combining elements of Singapore’s model with Europe’s existing disinformation frameworks. For example, establishing clearer definitions and exceptions for deepfakes in political contexts could mitigate misuse while maintaining freedom of expression. By refining its approach, the EU could address the growing risks of deepfakes without compromising democratic values, setting a standard for both member states and other nations grappling with similar concerns.