Introduction

Intellectual properties are creations of the mind including inventions, literary and artistic works and symbols, names and images used in commerce1. Artificial Intelligence (AI) refers to computer systems capable of performing tasks such as visual perception, speech recognition, decision-making and translation that otherwise would require human intelligence. Weak AI systems cannot think for themselves but can respond to specific situations. On the other hand, strong AI systems are conceptualised to be able to learn from previous experiences and to think and act like humans.

AI systems can be useful to reduce human casualties in conflict situations and hazardous workspaces or during accidents and natural calamities. They might also prove useful for extending their help in routine tasks like cleaning, shopping and transportation. AI systems are rapidly evolving by incorporating sophisticated technologies and it is expected that in near future they will produce 'inventions' without human intervention. This brings us to an interesting scenario where questions on the intellectual property related to such AI creations arise. The expanding scope of intellectual property in the context of increased role of AI in today's world is discussed here.

Intellectual property laws and AI

In common parlance, AI is the ability of a computer system to take decisions by itself. A relevant question with respect to the application of IP laws to AI is whether the results provided by the system are the products of its own intelligence or should be attributed to the underlying algorithms and commands.

AI and Copyrights

Copyright is a legal right granted to the creator of an original work, which allows him/her exclusive rights for its use and distribution. In order to obtain a copyright, the work should be tangible and original. Currently, AI can be employed for the creation of literary works and hence, the discussion on copyright in light of AI becomes significant2. Conventionally, the issue of copyright in computer generated works was not relevant since the programme was only a tool to support the creative process. However, with the advent of AI, the programme itself is capable of making many decisions involved in the creative process without human intervention.

In the case of Burrow Gilles Lithographic Co. v. Sarony, the question arose as to whether a copyright protection can be granted to a photograph. The Court held that purely mechanical labour per se the works are not creative, thereby narrowing the scope of its protection. In Bleistein v. Donaldson Lithographing, the Court described the uniqueness of human personality and specified the same to be a prerequisite to grant a copyright whereas in Alfred Bell & Co. v. Catalda Fine Arts, the Court lowered the standard for originality and held that it must not be copied from any other artistic work of similar character in order for it to be original. This judgment came out to be favourable for claiming copyrights for AI generated works, since such works were obviously not copied, but derived through programming and algorithms. However, the lack of a definitive perspective on these matters continues to affect the prospective right holders2.

Furthermore, the question of who gets the copyright still remains unanswered since the current law requires a legal personhood for the right holder, which an AI lacks2. Although the grant of copyright to AI systems has never been specifically prohibited, the laws of many countries do not have provisions for non-human copyright3. According to the United States Copyright Office, an original work of authorship will only be registered, provided that the work was created by a human being. In Feist Publications v. Rural Telephone Service Company, the Court ruled that the copyright law only protects the fruits of intellectual labour that are founded in the creative powers of the mind. In Acohs Pty Ltd v Ucorp Pty Ltd, an Australian Court declared that a work generated with the intervention of a computer could not be protected by copyright because it was not produced by a human. Similarly, the Court of Justice of the European Union (CJEU) in the case of Infopaq International A/S v Danske Dagbaldes Forening, reiterated that copyright only applies to original works, and that originality must reflect the author's own intellectual creation. This brings us to a conclusion that that a human author is necessary for a copyright work to exist3.The possible criminal liability of an AI is another area of concern. As per the current law, it will be the creator who is liable, although he may not own the mens rea or actus reus of a criminal act by the AI2.

AI and Patents

A patent refers to the exclusive right over an invention and the person granted with such a right is entitled by law to exclude others from making, selling, or even using the patented invention for a limited term2. For an invention to be granted a patent, it should pass the patentability criteria -novelty, inventive step, and industrial applicability. The biggest challenge faced by the AI systems in obtaining a patent for their 'inventions' is satisfying this three steps test2. The current laws need to be updated to permit grant of patents for inventions and lack of 'legal personality' status of AI systems is a major impediment in achieving this'2. However, the granting of patents to AI systems could raise the pertinent question of whether an AI system could be held liable for infringement of other inventors' patents, including those held by other AI systems. This leads us to a legal grey area over ownership rights that is expected to become harder to define in the future4.

Role of AI in Intellectual property

According to the World Intellectual Property Organization (WIPO), AI along with big data analytics and new technologies such as blockchain can be used to address the growing challenges the intellectual property offices (IPO) are facing. These include the examination and formalities checks for trademarks and patents, automatic classification of patents, search of patent prior art and figurative elements of trademarks, services for trademark applications, helpdesk services etc.5

Conclusion

As AI systems are becoming more advanced following which the number of 'inventions' by such systems is bound to increase in future. This offers wide scope for the framing of suitable legislations in order to provide adequate legal safeguards. The way forward would be, ensuring uniform international recognition for AIs, AI data protection through the passing of relevant acts and addressing the lacunae in fixing criminal liability for AIs' actions2. More importantly, there is a need to formulate clear and widely accepted guidelines with respect to the application of patent laws to AI.

Footnotes

1 https://www.wipo.int/edocs/pubdocs/en/intproperty/450/wipo_pub_450.pdf

2 https://cisse.info/pdf/2019/rr 01 2 https://cisse.info/pdf/2019/rr 01 artificial_intelligence.pdf

3 https://www.wipo.int/wipo_magazine/en/2017/05/article_0003.html

4 https://www.raconteur.net/risk-management/ai-ip-rights

5 https://www.wipo.int/about-ip/en/artificial_intelligence/

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.