Skip to content 
The countless intellectual property issues arising from the innovation and use of deepfakes recently came to light when Apple revealed that they had acquired a United States patent for a procedure for the creation of deepfakes. The patent showed that deepfake technology is ageing and that Apple is legally able to shield its technology in the US.
It is fairly simple to create deepfakes by using artificial intelligence. There are also a variety of apps that specifically generate deepfake imagery and allow the user to input their faces onto anyone they wish.
From the viewpoint of intellectual property, UK established law can accommodate for the effects of the production and usage of deepfakes. This is mostly because their creation is merely digital imagery which could be replicated manually by a human or a printer.
Copyright and deepfakes
Copyright would be the relevant right in protecting deepfake imagery since it is created digitally which would class it as an artistic work under section 4(1) of the Copyright, Designs and Patents Act 1988. Section 9(3) of the Act confirms that the author of a computer-generated work is the person who made the necessary arrangements for the creation of the work. Assuming the requirements of copyright are fulfilled, the copyright in deepfake imagery will subsist. On the other hand, copyright will not subsist if the deepfake is an infringing copy of a previous work.
It is common practice now for luminaries to register their photographic image as a trade mark to avoid their image being used by a third party in the course of trade. The right to privacy remains crucial in relation to deepfake imagery. A claim for breach of privacy may be valid if the deepfake is on the basis of private representation. Nevertheless, most deepfakes are publicly available which may limit such claims.
In the near future, the Courts will have to understand the consequences of the creation and reproduction of such deepfakes to protect individuals from exposure, defamation and many other situations.
By Zohaib Tahir, an LLM student at Solent University
