AI Battle: Should the Use of Deepfakes in Advertising and Politics Be Restricted?

June 19, 2026 · 7 min read

Question: Should the use of deepfakes in advertising and politics be restricted?

Perplexity

The use of deepfakes in politics and advertising should be restricted, especially in election campaigns, as their uncontrolled application undermines the integrity of elections and misleads the audience [1][2]. In politics, deepfakes can create convincing fake news, where, for example, a real candidate can be shown saying something they never uttered (as in the well-known case of the fake Barack Obama video), which directly influences voter opinion [1][4]. Without legislative barriers and mandatory labeling of synthetic content, such technologies become a tool of bias, destroying trust in democratic processes [1][2].

The situation in advertising is more complex: a complete ban on the technology is unjustified, as it has positive commercial applications, for instance, in localizing advertisements featuring models (the Zalando campaign with Cara Delevingne) or in creating content with celebrities without their physical presence [2][5]. However, the use of deepfakes in advertising must be strictly regulated: legal permission must be obtained for the depiction of a person (Article 152.1 of the Civil Code of the Russian Federation), licensing agreements for AI applications must be observed, and content must be clearly labeled as synthetic to avoid violating advertising accuracy laws [3][6].

The optimal solution is a selective ban: to introduce a complete ban on deepfakes in political advertising and election campaigns, but to permit them in commercial advertising, provided that mandatory labeling and rights to the image are in place [1][2]. Amendments are already being discussed in Russia that would introduce criminal liability for the creation and distribution of deepfakes if they are misleading, as well as mandatory labeling of synthetic content [3][4]. This approach allows for the mitigation of risks in politics while preserving technological capabilities for marketing.

Sources:

Gemini

The use of deepfakes in advertising and politics raises serious ethical and legal disputes, making their restriction necessary. In advertising, deepfakes can be used to create false impressions about a product or service, misleading consumers. For example, using the image of a celebrity without their consent to promote a product may violate image rights and reputation. In politics, deepfakes pose an even greater threat, as they can be used to misinform voters, discredit political opponents, or even interfere in elections. The creation of convincing fakes that are difficult to distinguish from reality undermines trust in information and democratic institutions.

Existing legislative measures in various countries aim to regulate the use of deepfakes. In Russia, for instance, the creation of deepfakes is not directly prohibited, but their use without the consent of the depicted person may violate articles of the Civil Code of the Russian Federation on the protection of image. In the US, draft laws are being discussed aimed at protecting victims from deepfakes and establishing the right to control digital doubles. The European Union has introduced mandatory labeling of synthetic content as part of the AI Act. The introduction of restrictions, such as mandatory labeling of deepfakes or a complete ban on their use in political advertising, is an important step in minimizing risks. This will help maintain trust in information and protect society from manipulation.

Sources:

Company Lawyer. Deepfake: Legal Regulation and Content Authorship.
AVB Fort. Deepfakes Outlawed: How Artificial Content is Regulated and What it Means for Business.
SelSup. Deepfakes and the Law: Can Celebrity Photos Be Generated with AI in Russia?
"Center for Media Rights Protection" Foundation. The Dangers of Deepfakes: Examining Legal Risks.
LEGAS Company Legal Services. Rules for the Use of Deepfake Technologies in US and PRC Law: Adapting Foreign Legal Regulation Experience.
CyberLeninka. Potential Threats of Unauthorized Use of Political Deepfakes During Political Elections: International Experience.
The Reality Crisis: How Deepfakes Invisibly Change Politics.
Analysis by the research unit of the American company Jigsaw. Political Deepfakes Have Become the Most Popular Way to Abuse AI.
Real Fake: The Benefits and Harms of Deepfakes for Business.
Russian Journal of Economics and Law. Make America Fake Again?: A Ban on the Use of Deepfakes of Federal Candidates in Political Advertising Does Not Violate the First Amendment to the US Constitution.
Runetlex. The Illusion of Reality: Deepfakes and Law – Legal Services for Digital Business.
Cybersecurity. Deepfakes and Their Impact on Media and Society: A Real Threat or an Exaggerated Panic?
Legal Academy. Deepfakes and Liability for Their Distribution. What is Important to Know.

ChatGPT

The restriction on the use of deepfakes in advertising and politics is an important and timely measure, considering the potential risks associated with manipulating public opinion and disinformation. Deepfakes, by creating convincing video and audio fakes depicting real people, can distort information to such an extent that the viewer loses the ability to distinguish truth from falsification. This is particularly dangerous in a political context: substituting politicians' statements or demonstrating events that did not occur can lead to the destabilization of public trust and even cause social conflicts. For example, during elections, cases of deepfake use to discredit candidates have already been recorded, undermining the fairness and transparency of the electoral process.

In advertising, deepfakes can also be used harmfully, causing consumers to make decisions based on false information or a fake brand reputation. Imagine a brand that uses a deepfake to "revive" a celebrity to promote a product without their consent. This can mislead the buyer and negatively affect the reputation of both the star and the company. On the other hand, deepfakes can have positive applications, such as creating high-quality content or retouching old materials, but this requires transparency and clear labeling that the audience is viewing a synthetic product.

Thus, regulating the use of deepfakes in advertising and politics is necessary to protect human rights and ensure fairness. Restrictions should include obligations for labeling such materials, penalties for abuse, and preventive measures to improve society's media literacy. Without such measures, deepfakes risk turning from a tool of creativity into an instrument of manipulation and dangerous disinformation.

Sources:

The Guardian — article on the risks of deepfakes in elections
Vedomosti — materials on the application of deepfakes in advertising and legislative initiatives

Grok

Yes, the use of deepfakes in advertising and politics requires restrictions — primarily through mandatory labeling and a ban on the dissemination of knowingly deceptive content during key periods.

In politics, the risks are particularly high: deepfakes undermine voter trust and can directly influence election outcomes. Examples already exist: in Slovakia, two days before the 2023 vote, an audio deepfake of the opposition leader was circulated, allegedly discussing vote-buying — the party lost; in Indonesia, the Golkar party "resurrected" the deceased dictator Suharto for a pre-election address; in the US in 2024, New Hampshire voters received a fake call from Biden's voice urging them not to vote. By 2026, 30 US states had introduced rules for labeling or banning deepfakes in political advertising, recognizing the threat of disinformation.

In advertising, the problem is no less acute: deepfakes of famous personalities (Elon Musk, Tom Hanks, Taylor Swift) are used to promote dubious investments, goods, or services, misleading consumers and damaging brand reputations. Such fake endorsements blur the line between reality and fiction, reducing trust in any commercial communication. A complete ban is unrealistic due to considerations of freedom of speech and satire, but transparent requirements — to indicate that content is AI-generated — are already working in several jurisdictions and allow for creativity without harming society.

Sources:

National Conference of State Legislatures (NCSL) — Artificial Intelligence (AI) in Elections and Campaigns.
Reuters — AI deepfakes blur reality in 2026 US midterm campaigns (March 2026).
Public Citizen — Tracker: State Legislation on Deepfakes in Elections.
Multistate.us — How AI-Generated Content Laws Are Changing Across the Country (February 2026).
Wired and other reports on specific incidents in Slovakia and Indonesia.

Claude

I will search for current information on deepfake regulation and ongoing debates on this issue.## Deepfakes Need to Be Restricted — But Selectively

An outright ban on deepfakes would be a mistake, but strict regulation of their use in political advertising and for manipulative purposes is quite justified. Deepfakes can not only damage the reputation of anyone — whether a public figure or an average person — but also threaten national institutions or even national security. Practice confirms this threat: in the 2019 elections, British politician Boris Johnson urged voters to support his rival Jeremy Corbyn, and Corbyn, in turn, urged the audience to support Johnson. Furthermore, during the Israel-Hamas war in 2023-2025, dozens of false videos of attacks, statements by politicians, and journalists spread on social media. The US example shows a constructive approach: in 2019, laws were passed in Texas and then in California prohibiting the use of audio, photo, and video fakes in political campaigning and elections.

However, a complete ban would be excessive. Well-known examples of good-faith use of deepfake technology include the rendering of actor Bruce Willis's image for Megafon's advertising and the character of Georges Miloslavsky from the Soviet comedy "Ivan Vasilievich Changes Profession" (actor Leonid Kuravlev) for Sber's advertising. Moreover, the technologies themselves that enable the creation of deepfakes are legally neutral — they are merely a data processing tool and do not constitute an offense by their mere existence or use outside of malicious content. The optimal solution is to develop a mechanism for mandatory labeling of materials created using deep synthesis technology, which will protect against manipulation without hindering the innovative potential of the technology.

Sources:

Lenta.ru
Pravo.ru
Legal Academy (Sfera)
RBC Trends
MSU Journalism
GARANT.RU
Journal KUI of the Ministry of Internal Affairs of Russia

Sources:

Blog