Authored By: Hysen Maloku
Hysen Maloku
Introduction
The world has benefited from the use of artificial intelligence in ways that would have been unimaginable until recently‚ such as the creation of generative AI that can produce very realistic images‚ videos‚ and audio. The term deepfake has been one of the most contested outcomes of the rise of artificial intelligence technology‚ referring to manipulated or artificially generated digital content created with machine learning algorithms that emulate the speech‚ emotions‚ or behaviors of real people.
Although they have been used for entertainment‚ education‚ or artistic purposes‚ the technology is increasingly being used for more nefarious applications‚ especially the generation of non consensual pornographic content.
Given the rapid advances in deepfake AI technologies‚ it is difficult for the modern world legal system to tackle deepfake porn and how it affects human dignity‚ personality rights‚ privacy and state of mental health.
Since the victims of deepfake pornography have never acted in pornography professionally‚ deepfake pornography subjects are humiliated‚ harassed‚ and suffer long-term consequences while the deepfake pornography is accessible via social and digital platforms such as Reddit‚ where it can easily be spread and accessed internationally. Victims are often unable to identify perpetrators‚ have content removed or obtain effective legal remedy.
The controversy caused by accusations against Christian Ulmen from his ex Wife Collien Fernandes a German television actress resulted in more discussions about digital sexual violence and laws related to pornographic use of artificial intelligence- generated content in Germany. No matter how accurate the media was, the case highlighted more general questions related to the legal and social issues connected to the exploitation of AI to generate and distribute sexually violent material.
One may ask if German legal system is capable of addressing these problems and protecting people from such kind of cyber violence. In this Article, it is examined whether there are sufficient means of protecting victims of AI-driven pornographic deepfakes under German legislation. It is argued that the legal instruments provided at present can only offer fragmentary and incomplete means of protection.
This paper also highlights some legal holes present in the legal system of Germany regarding the issue, placing the matter into an international perspective and proposing possible changes to improve victim protection.
Understanding Deepfake Pornography as Digital Sexual Violence
A deepfake is created using machine learning algorithms. AI can map the person’s face, clone his or her voice and generate images. It is possible to create a video showing something that did not happen to the person. The development of such technologies made them publicly available, reducing the technological barriers significantly and making deepfake technology available not only to the specialists but to anyone who has access to the internet. One of the most dangerous types of deepfakes are pornographic videos. In this case, the face of someone else is put into a sex scene without any permission. Women are mostly affected by such deepfakes; celebrities, journalists, influencers, and exes are the most common targets. It is why the issue is being referred to as digital sexual violence instead of the abuse of AI technology. The harm done to the victims of deepfakes is much more complicated than just violation of their privacy rights. Victims have suffered defamation, violation of their sexual autonomy, and their dignity. The reality is that it might be hard to prove that what happened did not actually happen as it would look very real. This may result in loss of career opportunities, destruction of interpersonal relations, and defamation in the public sphere. It causes emotional stress and even depression and anxiety in many cases.
It must be noted that the issue is amplified by online availability and speed of sharing
this kind of information. Although it is technically possible to take down the video from one platform, there will always be copies somewhere else. As for the servers hosting such content, they may be located in a different country and it would make it hard to deal with such situations legally.
III. The Fernandes/Ulmen Controversy and the Public Debate in Germany
The affair involving the media couple Christian Ulmen and Collien Fernandes came to
light following serious allegations made by Fernandes against her former partner. She claims that Ulmen created fake online profiles in her name over an extended period of time and used these accounts to engage in sexual conversations with unknown men, while also sending intimate or pornographic content. The material allegedly included images that closely resembled Fernandes or were generated using artificial intelligence or deepfake technology.
Fernandes reportedly described the incidents as forms of digital sexual violence and even as “virtual rape” committed by her husband. As a consequence, investigations were initiated both in Spain and Germany. Authorities are examining several potential criminal offenses related to digital abuse and possible domestic misconduct. Through his legal defense, Ulmen has denied the allegations. As of now, no final legally binding judgment has been issued.
Current Legal Protection under German Law
At the moment, German law provides a regulation of deepfake pornography based on constitutional law, civil law, criminal law, and data-protection regulation. It should be noted, however, that none of the above-mentioned elements have been developed for the purpose of regulating the use of AI for pornography purposes. Thus, there are numerous issues and legal challenges in this regard.
The primary legal basis that must be mentioned is the general personality right provided by Articles 1 and 2 of the German Constitution. The German constitutional law pays considerable attention to the protection of human dignity and the personal rights of individuals. Victims of deepfake pornography have the possibility to bring legal actions regarding injunctions, deletion, and compensations if a person’s personal integrity is violated or his/her reputation is harmed due to manipulated content.
One should highlight the right to one’s own image under Section 22 of the Kunsturhebergesetz (KunstUrhG). Under this section, any distribution or public presentation of a person’s image without his/her consent is prohibited. On paper, victims of pornographic deepfake could sue under this articleto protect themselves against publication of manipulated material. However, a number of complications arise if the image itself is not an actual photograph but rather an entirely artificial creation made via AI. One wonders if it falls under the category of image protection that was developed long before the existence of generative AI technologies.
There is a number of relevant criminal offenses in German criminal law. § 201a of the German Criminal Code penalizes violations of a highly personal sphere via image recording, especially unauthorized image recordings in an intimate situation. Again, it is still unknown whether artificial creation will be qualified as „image recording.“ In addition to this, there is also the crime provided under § 184k of the German Criminal Code violation of intimate sphere by image materials. Also, there are crimes of insult § 185 StGB), defamation §186 StGB) and malicious gossip (§187 StGB). Such violations might occur if victims of deepfake pornography suffer damage to their honor and reputation.
Namely, victims of deepfake pornography may have grounds to file a claim about unlawful use of biometric data in accordance with General Data Protection Regulations. However, the procedure of protecting the right to such data is quite complicated, lengthy, and ineffective in terms of rapid dissemination. One of the main problems is platform regulation. Explicit deepfakes tend to be shared through international sites or anonymous users. The identification of culprits becomes a difficult task along with demanding timely removal from platforms. In case the content has been removed, it may be reposted elsewhere. Taken into account altogether, the current legislation in Germany offers piecemeal protections but fails to provide a comprehensive solution tailored to the problem of AI-generated pornographic deepfakes.
Legal Gaps and Regulatory Challenges
The development of artificial intelligence poses several legal challenges to the existing legislation. Most importantly, there exists a lack of a dedicated framework for regulating deepfake pornography. Existing regulations operate indirectly, using previous definitions of crimes to describe modern realities. This creates a problem of interpretation that must be addressed. Another question relates to the difference between authentic and synthetic material. Many criminal regulations were based on real materials. However, AI-generated images do not need any alteration of the original private photograph. This calls into question the adequacy of current crimes in dealing with deepfakes producing similar psychological and reputational harm.
There are also enforcement problems. Offenders act anonymously, communicating through encryption, foreign servers, or using pseudonyms online. Thus, jurisdictional differences hinder criminal and civil proceedings against them. In addition, the speed of distribution makes the process of seeking legal remedy ineffective because the damage has already been done.
Moreover, there exist additional concerns regarding evidentiary matters. As technology develops at high speed, deepfakes become harder to detect. Therefore, victims face difficulties in proving manipulation. In addition, the credibility of authentic content as evidence will be undermined.
Lastly, many current measures place a heavy burden on victims. Affected persons are forced to go through lengthy civil lawsuits, interact with many sites, and confront traumatic images again during the trial. It is argued that the existing system is unable to accommodate the psychological and societal impacts of digital sexual assault.
Comparative and International Perspectives
A number of jurisdictions have begun developing more sophisticated approaches to dealing with deepfake pornography. As a result, comparative law analysis provides useful lessons for possible future reforms in Germany.
European Union
On the European Union level, there is an increased focus on regulating both artificial intelligence and digital platforms. With respect to the former, the recently developed EU AI Act is aimed at establishing a risk-based regulatory framework for artificial intelligence, which includes mandatory provisions concerning transparency for some AI-generated content. Although this measure does not apply specifically to the problem of deepfake pornography, it demonstrates a growing concern about AI risks. In turn, the new Digital Services Act is going to increase platform responsibilities related to illegal content moderation.
United Kingdom & USA
The United Kingdom has also introduced additional legal regulations in the context of the recently adopted Online Safety Act. This Act covers certain cases of online abuse and expands the responsibility of digital platforms. Several states in the USA have adopted criminal statutes making the production or dissemination of non- consensual deepfake pornography illegal.
When compared to the aforementioned developments, it can be seen that Germany lacks special measures aimed at preventing non-consensual deepfake pornography. While the existing legislation provides partial protection against deepfakes, there is no legal measure that explicitly refers to this problem.
Proposed Reforms
Based on the analysis of existing problems with current legal mechanisms in Germany, several reforms are to be considered to improve protection against AI- generated sexual deepfakes in Germany.
First, there is a need to consider establishing a separate criminal offense that covers cases involving the production and distribution of non-consensual deepfake pornography. This step would make the law clearer in terms of applicability to synthetic media and recognize the special nature of this kind of crime.
Second, there is a need to increase the effectiveness of content removal in case of harm from the use of generative AI systems, especially considering that victims usually need urgent protection as online dissemination usually happens relatively quickly.
Third, the existing legal regulations should pay more attention to victim protection, especially through the implementation of simplified mechanisms for seeking injunctions, expanding psychological counseling, and facilitating compensation processes.
Fourth, there is a need for introducing additional transparency measures regarding AI- generated content. For instance, requiring labeling or watermarking might help in identifying manipulated images or videos. This solution will not completely prevent deepfakes but can add more accountability in this sphere. Last, international cooperation is extremely important since the online abuse can be committed across national borders.
VII. Conclusion
AI-generated pornographic deepfakes represent one of the most serious modern challenges at the intersection of technology, privacy, and human dignity. The emergence of generative artificial intelligence has fundamentally transformed the scale, accessibility, and realism of digital manipulation. As demonstrated by the public debate surrounding allegations involving Christian Ulmen and Collien Fernandes, deepfake pornography has become a matter of significant legal and societal concern within Germany.
Current German law provides certain mechanisms for protecting victims through personality rights, image protection, criminal law, and data protection rules. However, these protections remain fragmented and insufficiently adapted to the realities of generative AI. Existing legal provisions were largely developed before the emergence of modern deepfake technologies and therefore struggle to address synthetic sexual content effectively.
The challenges associated with deepfake pornography extend beyond individual reputational harm. They concern broader issues of digital sexual violence, technological accountability, evidentiary reliability, and the protection of human dignity in increasingly virtual environments. Without more specialized regulation, victims may continue to face substantial barriers in obtaining effective protection and legal remedies.
Germany therefore faces an important regulatory choice. While technological innovation should not be unnecessarily restricted, legal systems must evolve to protect individuals from emerging forms of AI enabled abuse. A coherent and victim oriented framework addressing deepfake pornography is necessary to ensure that fundamental rights remain meaningful in the digital age.
Academic literature
Deeptrace (2019) – The State of Deepfakes
https://regmedia.co.uk/2019/10/08/deepfake_report.pdf
European Institute for Gender Equality (EIGE) – Cyber Violence against Women and Girls
https://eige.europa.eu/publications/cyber-violence-against-women-and-girls
Henry, N., Flynn, A., Powell, A. (2020) – Technology-Facilitated Sexual Violence – Violence Against Women Journal
https://journals.sagepub.com/doi/10.1177/1077801219875820
German Basic Law (Grundgesetz), Arts. 1–2
https://www.gesetze-im-internet.de/englisch_gg/
BGHZ 20, 345 („Herrenreiter“)
Kunsturhebergesetz (KunstUrhG), §§ 22–23
https://www.gesetze-im-internet.de/kunsturhg/
German Criminal Code (StGB) – §§ 185–187, 201a, 184k
https://www.gesetze-im-internet.de/englisch_stgb/
Fischer, T. – StGB Commentary
Dreier, H. (ed.) – Grundgesetz Commentary
GDPR (Regulation (EU) 2016/679)
https://eur-lex.europa.eu/eli/reg/2016/679/oj
Kühling / Buchner (eds.) – GDPR / BDSG Commentary
EU Digital Services Act (DAS)
https://digital-strategy.ec.europa.eu/en/policies/digital-services-act-package
European Commission – AI Policy
https://digital-strategy.ec.europa.eu/
European Parliamentary Research Service (EPRS, 2021) – Tackling Deepfakes in the EUhttps://www.europarl.europa.eu/thinktank/
t-online (2026) – Collien Fernandes gegen Christian Ulmen: Die Chronologie des Falls
https://www.t-online.de/unterhaltung/stars/id_101178826/collien-fernandes-gegen- christian-ulmen-die-chronologie-des-falls.html
