Our world is passing through an era of explosive technological innovation. New technologies are affecting unexpected areas of our lives, even that of artistic practice. Amidst this unprecedented technological advance, creatives have taken the opportunity to use artificial intelligence to assemble musical pieces, paintings and even write poetry.
Start-ups are often at the forefront of this AI-driven practice, as they place a unique emphasis on boundary-breaking and innovative solution finding. The increase in AI-driven innovation that we have seen already has sparked great interest within the world of IP. Most recently, the United Kingdom Intellectual Property Office opened a consultation on Artificial Intelligence and IP, to consider how to effectively cater to this unprecedented trailblazing.
This article will begin by explaining why and how AI presents unique opportunities for small and medium-sized enterprises (SMEs) in the arena of IP. The key issues that creatives may face when navigating copyright protection of AI-made works will be outlined. We will take a comparative jurisdictional approach, focusing on the different responses that have been implemented so far, across the UK, EU, US and Singapore.
SMEs and copyright
Start-ups often rely on innovation to stand out in a competitive market environment. The reliance of a small or medium enterprise on innovation is likely to be much heavier than that of an established entity, as it enters markets which are already saturated with traditional ways of offering products and services.
As recognised in WIPO’s guide to Intellectual Property for Start-ups, during its early growth, a start up will both consume and generate IP. Whenever an IP-protected work is used, a start-up will have to acquire relevant rights – such as licences or assignments – to use the work without offending against the owner’s rights. Consumption of IP will likely be greatest during the research and development phase of the company’s cycle. Following the completion of this phase, most start-ups will begin to develop their own novel solutions and thereby generate IP. As a start-up begins to innovate in its own right, it should look to the law to protect its novel solutions and inventions against copyists and free-riders. It is important to steer one’s journey within IP at both phases of the company’s growth. In the former, IP must be managed to avoid infringing on others’ rights, whilst in the protection-seeking phase acquiring IP is a fundamental tool to monetise your own assets.
To aid this journey, it is important to firstly understand your own technology, in particular grasping how close a product or service may be to commercialisation – also known as the product’s technology readiness level (TRL). Knowing this will help you navigate the existing framework of IP law effectively. It is important to obtain relevant IP rights at the right time, as all IP rights have a lifecycle of their own and face expiration, subject to possible renewal in the case of trade marks and designs.
The increasing use of AI creates unique opportunities and challenges for SMEs as they seek copyright protection.
AI and copyright
Novel practices enable the use of AI which produce AI artworks through robotic will. One illustration is Obvious which is a collective consisting of “researchers, artists, and friends, working with the latest models of deep learning to explore the creative potential of artificial intelligence”. The group has used AI to generate the first AI-made artwork to be sold in a major auction house; Portrait of Edmond Belamy.
Since then, larger companies have followed the trend. For example, The Next Rembrandt, is a collaboration between ING, Microsoft, Tu Delf and Mauritshuis in which data scientists, developers, AI and 3D printing experts have produced software that collates old paintings of Rembrandt, to produce new works that imitate the style of the Old Master’s renowned brush work.
This practice of using an AI system to generate an artwork challenges the existing copyright law. It may make it harder for innovators to navigate and secure their legal standing. Many jurisdictions have responded to this differently.
Divergent jurisdictional approaches
The solution provided to this legal conundrum by UK law is straightforward in comparison to the response given by other jurisdictions. Section 9(3) of the Copyright, Designs and Patents Act 1988 (CDPA 1988) legislates that a “In the case of a literary, dramatic, musical or artistic work which is computer-generated, the author shall be taken to be the person by whom the arrangements necessary for the creation of the work are undertaken.”
The effect of this provision is that a work which has been produced through automation by a computer programme, will generate copyright that is automatically granted to the producer of the relevant software. In practice, this will be the computer programmer or software engineer, often resulting in joint ownership between more than one person.
However, such a work is granted only 50 years of protection from the end of the calendar year in which the work was made; section 12(7) CDPA 1998 in comparison with other works which are granted protection until 70 years after the author’s death under section 12(2) CDPA 1988.
EU legislation does not address copyright protection of computer-generated works explicitly. However, its dicta on the interpretation of ‘originality’ which is a legal standard required for copyright to arise in an artistic work arguably closes the door for copyright protection of AI-made works.
A pivotal decision is that of Infopaq (C-5/08 ), handed down by the CJEU. It stated that only if a literary, dramatic, musical or artistic work is found to meet the requisite standard of originality will it be granted copyright protection. This CJEU decision v could potentially be used to exclude AI-made artistic works from protection by copyright.
Singapore has taken a directly opposite legal stance to the UK’s. Whilst it has not legislated on this matter through explicit statutory provision, the Singaporean Court of Appeal has stated that “not infrequently, in cases involving a high degree of automation, there will be no original work produced for the simple reason that there are no identifiable human authors”; Asia Pacific Publishing Pte Ltd v Pioneers & Leaders (Publishers) Pte Ltd  SGCA 37. This hints at, if not outright demonstrates, clear reluctance to copyright protection of works that result from AI automation.
The US has also yet to legislate explicitly on this issue. However, clear guidance has been given by the relevant US legal bodies, which suggests that copyright protection would be refused to AI-made works.
The Supreme Court of the United States has stated that copyright law only protects “the fruits of intellectual labor” that “are founded in the creative powers of the mind”; Feist Publications, Inc., v. Rural Telephone Service Co., 499 U.S. 340 (1991). This dictum also points strongly in the direction of an ingrained reluctance to grant copyright to AI-made artworks.
The diverse trajectories on copyright protection of AI-made literary, dramatic, musical and artistic works show us a lack of a clear jurisdictional consensus. The divergent approaches of different jurisdictions are only telling of the complex nature of this matter.
As SMEs innovate and push boundaries within their own industries, they have to stay sharply aware of the changing legal landscape which continues to evolve to better support our society’s changing demands.
You can read more about the involvement of AI in patenting inventions in our blog Inventive AI: can machines innovate?
Download our IP Guide for the Forward-Looking SME to learn more about why IP should be a core asset for an SME.
Sign up to our newsletter: Forward - news, insights and features
We have an easily-accessible office in central London, as well as a number of regional offices throughout the UK and an office in Munich, Germany. We’d love to hear from you, so please get in touch.