Äger jag mitt ansikte? En analys av individens rättsliga skydd mot deepfakes
(2025) LAGF03 20252Department of Law
Faculty of Law
- Abstract
- Recent developments in artificial intelligence have enabled the creation of AI-generated media, such as images and videos. Over time these have been known as deepfakes, which can convincingly depict real individuals in multiple different scenarios. This technological development raises questions regarding how well-equipped Swedish law is to protect individuals personal integrity against this new form of media. The purpose of this thesis is to ana-lyse the current Swedish legal system in regard to how it would treat deep-fakes. In addition, it will also highlight the legal possibilities and limitations in the current legal framework.
The thesis examines personality rights, copyright law, legislation concerning name and image, the Swedish... (More) - Recent developments in artificial intelligence have enabled the creation of AI-generated media, such as images and videos. Over time these have been known as deepfakes, which can convincingly depict real individuals in multiple different scenarios. This technological development raises questions regarding how well-equipped Swedish law is to protect individuals personal integrity against this new form of media. The purpose of this thesis is to ana-lyse the current Swedish legal system in regard to how it would treat deep-fakes. In addition, it will also highlight the legal possibilities and limitations in the current legal framework.
The thesis examines personality rights, copyright law, legislation concerning name and image, the Swedish Instrument of Government and the European Convention on Human Rights, as well as the General Data Protection Regulation. All these are presented and discussed in relation to their potential applicability to deepfakes. The analysis shows that no comprehensive protection against deepfakes exists in the current Swedish law. Protection may arise in certain legal areas, however it is often highly dependent on the context in which the AI-generated material is used. Overall, the legal protection against deepfakes appears inconsistent and gives rise to more challenges than effec-tive solutions. (Less) - Abstract (Swedish)
- Den senaste tidens utveckling av artificiell intelligens har möjliggjort framställning av AI-genererad media som bilder och videor. Med tiden har de blivit kallade deepfakes som på ett mycket trovärdigt sätt kan föreställa verkliga personer. Till följd av detta uppkommer frågor om hur gällande svensk rätt förmår skydda individens personliga integritet mot nya tekniska utvecklingar. Syftet med uppsatsen är att analysera vilket skydd svensk rätt ger mot deepfakes samt att belysa de rättsliga möjligheter och begränsningar med dagens lagstiftning.
Uppsatsen behandlar personlighetsrätten, upphovsrätten, lagstiftning rörande namn och bild, regeringsformen och Europakonventionen samt dataskyddsförordningen. Samtliga presenteras och diskuteras... (More) - Den senaste tidens utveckling av artificiell intelligens har möjliggjort framställning av AI-genererad media som bilder och videor. Med tiden har de blivit kallade deepfakes som på ett mycket trovärdigt sätt kan föreställa verkliga personer. Till följd av detta uppkommer frågor om hur gällande svensk rätt förmår skydda individens personliga integritet mot nya tekniska utvecklingar. Syftet med uppsatsen är att analysera vilket skydd svensk rätt ger mot deepfakes samt att belysa de rättsliga möjligheter och begränsningar med dagens lagstiftning.
Uppsatsen behandlar personlighetsrätten, upphovsrätten, lagstiftning rörande namn och bild, regeringsformen och Europakonventionen samt dataskyddsförordningen. Samtliga presenteras och diskuteras utifrån en eventuell tilllämpning på deepfakes. Analysen visar att något heltäckande skydd mot deepfakes inte existerar i svensk rätt. Skydd kan däremot uppstå inom ett visst område men är ofta fragmenterat och beror helt på i vilket sammanhang det AI-genererade materialet används. Sammantaget framstår skyddet mot deepfakes som inkonsekvent och bidrar med fler utmaningar än möjligheter i slutändan. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9216971
- author
- Stenström, Arvid LU
- supervisor
- organization
- course
- LAGF03 20252
- year
- 2025
- type
- M2 - Bachelor Degree
- subject
- keywords
- Immaterialrätt, Personlighetsrätt, AI
- language
- Swedish
- id
- 9216971
- date added to LUP
- 2026-04-14 10:27:34
- date last changed
- 2026-04-14 10:27:34
@misc{9216971,
abstract = {{Recent developments in artificial intelligence have enabled the creation of AI-generated media, such as images and videos. Over time these have been known as deepfakes, which can convincingly depict real individuals in multiple different scenarios. This technological development raises questions regarding how well-equipped Swedish law is to protect individuals personal integrity against this new form of media. The purpose of this thesis is to ana-lyse the current Swedish legal system in regard to how it would treat deep-fakes. In addition, it will also highlight the legal possibilities and limitations in the current legal framework.
The thesis examines personality rights, copyright law, legislation concerning name and image, the Swedish Instrument of Government and the European Convention on Human Rights, as well as the General Data Protection Regulation. All these are presented and discussed in relation to their potential applicability to deepfakes. The analysis shows that no comprehensive protection against deepfakes exists in the current Swedish law. Protection may arise in certain legal areas, however it is often highly dependent on the context in which the AI-generated material is used. Overall, the legal protection against deepfakes appears inconsistent and gives rise to more challenges than effec-tive solutions.}},
author = {{Stenström, Arvid}},
language = {{swe}},
note = {{Student Paper}},
title = {{Äger jag mitt ansikte? En analys av individens rättsliga skydd mot deepfakes}},
year = {{2025}},
}