Publication date: November 17, 2026
A single uploaded photograph is now enough to make a face blink, smile and speak. Since 2 August 2026, EU law for the first time defines a “deepfake” and requires such content to be labelled. For the person whose face is animated, however, the most effective protection still comes from well-established Polish rules on image rights, copyright, data protection, personal rights and criminal law. This article explains how the old and new layers fit together, where the 2024 analysis needed correcting, and when animating a photo remains lawful.
Animating a still image typically combines AI models that map facial features with generative models that produce motion, facial expression and lip movement synchronised with supplied text or audio. Services such as D-ID allow a user to upload a photo, type a script or upload a voice recording, and receive a short video in which the person in the photo appears to speak; a synthetic face or a stock image can be used instead of a real photograph.[1] Comparable functions are now built into general-purpose image-to-video tools and social media apps.
Not every animation is a deepfake. The EU Artificial Intelligence Act (AI Act) now defines a deepfake as AI-generated or manipulated image, audio or video content that resembles existing persons, objects, places, entities or events and would falsely appear to a person to be authentic or truthful.[2] An animated face invented by the software, or an obviously stylised cartoon, will usually fall outside that definition. The legal questions start when the animation is linked to an identifiable real person or a real event, and they become acute when it is used to deceive, for example in election disinformation, investment scams using a celebrity’s face, or sexualised fakes.[3]
Image right. Under Article 81(1) of the Polish Copyright Act, disseminating a person’s image requires that person’s permission; unless expressly reserved otherwise, permission is presumed where the person was paid for posing.[4] Permission is not required for the image of a commonly known person where the image was made in connection with that person’s public functions, nor for a person who is merely a detail of a larger whole such as a gathering, landscape or public event (Article 81(2)). The right is infringed by dissemination: animating a photograph for oneself is not dissemination, whereas posting the result on social media is. In our view, the public-figure exception does not cover an animation that puts invented words into a politician’s or celebrity’s mouth, because the exception concerns authentic images of public activity, not fabricated statements.
A clarification of the 2024 version of this article is needed here. The image right does not “arise” when a photograph is taken; it is a personal right of the person depicted, existing independently of any photograph. Copyright in the photograph is a separate right belonging to the photographer, and it exists only where the photograph meets the originality threshold of Article 1(1) of the Copyright Act. That threshold is low, but not every automatic snapshot will meet it.
Copyright in the photograph. Animating a protected photograph will usually produce an adaptation (opracowanie) within the meaning of Article 2 of the Copyright Act. Creating an adaptation is free, but disposing of and using it requires the consent of the author of the original work (Article 2(2)). The permitted personal use exception in Article 23 allows free use of a disseminated work within a circle of persons connected by personal relationships, such as family or friends; publishing the animation online goes beyond that circle.
Deceased persons. Animating photographs of deceased relatives is one of the most popular uses of these tools. Under Article 83 of the Copyright Act, claims for unauthorised dissemination of an image may be pursued for up to 20 years after the death of the person depicted. Close relatives may also rely on their own personal right to cherish the memory of the deceased (kult pamięci osoby zmarłej) under the Civil Code.
The image of an identifiable person is personal data under Article 4(1) GDPR.[5] Here too the 2024 analysis needs refining: a facial image is not automatically biometric data. Under Article 4(14) and Recital 51 GDPR, photographs are biometric data only when processed through specific technical means allowing the unique identification or authentication of a person. Animation tools map facial features in order to generate movement rather than to identify, so whether Article 9 applies is debatable and depends on the technology used; where it does, explicit consent will in practice be required.
Outside Article 9, processing needs a legal basis under Article 6, which for animating someone’s face will usually be consent (Articles 6(1)(a) and 7). Two limits on the GDPR’s reach are important. First, it does not apply to processing by a natural person in the course of a purely personal or household activity (Article 2(2)(c)); however, the Court of Justice has held that publishing data on the internet so that it is accessible to an indefinite number of people falls outside that exemption.[6] Second, the GDPR does not apply to deceased persons (Recital 27), so relatives must rely on the copyright and civil-law routes described above.
Where a person’s image has been processed unlawfully, they may demand erasure under Article 17(1)(d) GDPR, or under Article 17(1)(b) after withdrawing consent, and may complain to the President of the Personal Data Protection Office (UODO). In serious cases, unlawful processing is also a criminal offence under Article 107 of the Polish Personal Data Protection Act.[7] The UODO President has publicly called for dedicated rules on deepfakes.[8]
The most significant change since 2024 is that the AI Act’s transparency obligations in Article 50 have applied since 2 August 2026. They were not postponed by the Digital Omnibus on AI (Regulation (EU) 2026/1744), which entered into force on 27 July 2026 and deferred mainly the high-risk rules.[9] Two obligations matter most for photo animation:
A key nuance: the AI Act’s definition of a deployer excludes use of an AI system in the course of a personal, non-professional activity (Article 3(4)). A private individual animating a grandmother’s photograph for the family is therefore not bound by Article 50(4), although the app provider remains bound by Article 50(2). Businesses, advertisers, media, influencers acting commercially and political campaigns are bound. Breaches of Article 50 may attract fines of up to EUR 15 million or 3% of total worldwide annual turnover, whichever is higher (Article 99(4)).
Practical guidance is provided by the Commission’s Code of Practice on marking and labelling of AI-generated content, published on 10 June 2026, and by the Commission’s guidelines on the Article 50 transparency obligations.[10] The Code is voluntary, but the Commission has confirmed it as an adequate means of demonstrating compliance. The Omnibus also added to the list of prohibited AI practices systems designed to generate non-consensual intimate imagery or child sexual abuse material, with a transitional period ending on 2 December 2026.
In Poland, the Act of 3 July 2026 on Artificial Intelligence Systems entered into force on 11 August 2026.[11] It establishes the Commission for the Development and Security of Artificial Intelligence (KRiBSI) as the market surveillance authority, which accepts complaints and imposes administrative fines; its provisions on inspections, proceedings and penalties have applied since 28 October 2026.
The Digital Services Act (DSA) has applied in full since 17 February 2024.[12] Every hosting provider must operate a notice-and-action mechanism (Article 16) and give reasons for its moderation decisions (Article 17), which gives victims a direct route to have a deepfake reviewed and removed. Very large online platforms must also assess and mitigate systemic risks; among the mitigation measures, Article 35(1)(k) expressly refers to ensuring that generated or manipulated images, audio or video that appreciably resemble existing persons or events are distinguishable through prominent markings.
Poland’s national implementation has been slow. After the President vetoed the first implementing act on 9 January 2026, the Government split the reform into two bills. The first, the Act of 4 September 2026 amending the Act on the Provision of Electronic Services, which designates the President of the Office of Electronic Communications (UKE) as Digital Services Coordinator, was signed in September 2026. The second, introducing a national procedure for orders to disable access to illegal content (including content relating to identity theft and fraud), has been passed by Parliament and sent to the President for signature.[13] The DSA itself applies directly regardless of these national acts.
Polish law still contains no stand-alone deepfake offence, but several existing provisions apply.[14] (The 2024 version of this article referred to Article 212 §6; the correct provision on defamation through the mass media is Article 212 §2.)
EU law will soon require further change. Directive (EU) 2024/1385 on combating violence against women and domestic violence obliges Member States to criminalise producing or manipulating, and then making accessible to the public, material that makes it appear that a person is engaged in sexually explicit activities, without that person’s consent, where likely to cause serious harm. The transposition deadline is 14 June 2027.[16]
Articles 23 and 24 of the Civil Code remain the most flexible tool for victims.[17] The catalogue of personal rights is open: it expressly includes image, name, pseudonym and honour, and voice is widely recognised as protected as well, which matters for voice-cloning. A person whose personal rights are threatened or infringed may demand that the conduct cease, that its effects be removed (for example by deleting the material and publishing a statement or apology), and monetary compensation for non-material harm or payment to a social cause (Articles 24 §1 and 448), as well as damages for any financial loss. Unlawfulness is presumed, so it is for the creator or publisher of the animation to prove a justification such as valid consent.
The legislative picture is still moving. Besides the criminal-law bill and the Directive mentioned above, the Polish Ministry of Culture has indicated that it is considering deepfake-related provisions in the planned amendment of the Copyright Act.[18] Elsewhere in the EU, a Danish bill introduced on 8 October 2026 would give individuals copyright-style control over realistic digital imitations of their face and voice, an approach that may influence the debate in Poland.
Not every animation is unlawful. In summary:
Legal sources
[1]D-ID, Creative Reality Studio, https://www.d-id.com/creative-reality-studio/ (accessed 5 August 2024).
[2]Article 3(60) of Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act), OJ L, 2024/1689, 12.7.2024.
[3]See e.g. Communication from the Commission – Commission Guidelines for providers of Very Large Online Platforms and Very Large Online Search Engines on the mitigation of systemic risks for electoral processes pursuant to Article 35(3) of Regulation (EU) 2022/2065, C/2024/3014, OJ C, 26.4.2024.
[4]Act of 4 February 1994 on Copyright and Related Rights (consolidated text: Journal of Laws (Dz.U.) 2025, item 24, as amended), hereinafter the “Copyright Act”.
[5]Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation), OJ L 119, 4.5.2016, p. 1.
[6]Judgment of the Court of Justice of 6 November 2003, Lindqvist, C-101/01, EU:C:2003:596.
[7]Act of 10 May 2018 on the Protection of Personal Data (consolidated text: Journal of Laws (Dz.U.) 2019, item 1781, as amended).
[8]“Deepfake a prawo – prezes UODO apeluje o nowe regulacje”, Prawo.pl, https://www.prawo.pl/biznes/deepfake-a-prawo-prezes-uodo-apeluje-o-nowe-regulacje,535237.html.
[9]Regulation (EU) 2026/1744 (Digital Omnibus on AI), adopted by the European Parliament on 16 June 2026 and by the Council on 29 June 2026; see e.g. Goodwin, “Not Delayed, Not Deferred: EU AI Act Transparency Obligations Are Now in Force”, August 2026.
[10]European Commission, Code of Practice on marking and labelling of AI-generated content, 10 June 2026 (press release IP/26/1328), https://digital-strategy.ec.europa.eu/en/policies/code-practice-ai-generated-content.
[11]Act of 3 July 2026 on Artificial Intelligence Systems (Journal of Laws (Dz.U.) 2026, item 1003).
[12]Regulation (EU) 2022/2065 of the European Parliament and of the Council of 19 October 2022 on a Single Market For Digital Services and amending Directive 2000/31/EC (Digital Services Act), OJ L 277, 27.10.2022, p. 1.
[13]See “Druga ustawa wdrażająca DSA w Polsce czeka na podpis prezydenta”, Tabletowo.pl, 2026; Ombudsman (RPO), letter to the President of 20 September 2026 on the DSA implementing act.
[14]Act of 6 June 1997 – Criminal Code (consolidated text: Journal of Laws (Dz.U.) 2025, item 383, as amended).
[15]Sejm print no. 3268 (10th term), deputies’ bill amending the Criminal Code, 7 October 2026.
[16]Article 5(1)(b) of Directive (EU) 2024/1385 of the European Parliament and of the Council of 14 May 2024 on combating violence against women and domestic violence, OJ L, 2024/1385, 24.5.2024.
[17]Act of 23 April 1964 – Civil Code (consolidated text: Journal of Laws (Dz.U.) 2026, item 795).
[18]Reply of the Minister of Culture and National Heritage to parliamentary interpellation no. 14310, 2026, https://api.sejm.gov.pl/sejm/term10/interpellations/attachment/ATTDR2K79/i14310-o3.pdf.
Services
KIELTYKA GLADKOWSKI is a Polish business law firm headquartered in Warsaw and Krakow, with a New York office under the iSTART1 brand. Building on a practice dating back to 1981, we advise clients in the following areas.
Select a practice area to learn more.
Service directory: kg-legal.eu/services
KG LEGAL KIELTYKA GLADKOWSKI offers legal support concerning AI-generated content in Poland. Our assistance covers image and personality rights, copyright licensing, data protection, consent documentation and AI Act transparency requirements.
We help businesses review proposed uses of synthetic media and assess responses to unauthorised use of faces or voices, including platform notices and civil remedies.
Contact: office@kg-legal.pl.