Artificial Intelligence Lacks Personhood To Become The Author Of An Intellectual Property

Rajiv Sharma and Ninad Mittal

22 Sep 2023 6:03 AM GMT

  • Artificial Intelligence Lacks Personhood To Become The Author Of An Intellectual Property
    Listen to this Article

    Traditionally, humans are supposed to have intelligence and create innovative or novel work capable of being recognized to be registered as intellectual property and the same belief has been reflected in the legislators of the IP laws.

    What if there is a machine or software that can comprehend the information fed into it to create an intelligence far more superior, accurate, and readily available for everybody. Why should such intelligence be treated differently and not be provided the rights available to intellectual property developed by human beings. Why should the creator of such a machine not get authorship of the work done by such a machine or why should such a machine not get authorship of the work developed by it autonomously without intervention of humans.

    Legitimacy Challenge to IP generated by AI

    Indian Copyright Act, 1957

    Section 2 (d) (vi) defines author to include, “in relation to any literary, dramatic, musical or artistic work that is computer-generated, the person who causes the work to be created”.

    Patents Act, 1970, wherein AI systems have been excluded from the ambit of applicability of the Act, thereby restricting the rights of the ‘person’. As evident in Section 2 (1) (p), Section 2 (1) (t), where the patentee is referred to as a person, and Section 6 (1) (a), which states that an application for a patent can be filed by ‘any person’, and also in Section 2 (1) (ja), which defines ‘inventive step’, an essential requirement for an invention to be patentable, that such invention must not be ‘obvious to a person skilled in the art’.

    The United Kingdom’s Copyright, Designs and Patents Act, 1988 (CDPA)

    Section 9 (3), states, “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.”

    Section 178 of the CDPA, has defined computer-generated work, as “generated by computer in circumstances such that there is no human author of the work”.

    Irish Copyright and Related Rights Act, 2000

    Section 21 defines ‘author’ as, the person who creates a work and includes: (f) in the case of a work that is computer-generated, the person by whom the arrangements necessary for the creation of the work are undertaken;”.

    ‘computer-generated’ is defined under Section 2 (1) as, “in relation to a work, means that the work is generated by a computer in circumstances where the author of the work is not an individual”.

    The IP laws across various jurisdictions only afford rights to a ‘person’. However, imparting rights to a work autonomously developed by the AI system without any human input or with minimal human contribution is still a challenge with no support from legislation or judicial verdicts.

    Decisions of copyright authorities or courts limit authorship to humans

    The Next Rembrandt

    An example of Generative AI created artwork, through the utilization of ML process, is “The Next Rembrandt”, a 3-D printed painting, created after analysing the works of 17th -century Dutch artist Rembrandt Harmenszoon van Rijn. This artwork was created with the collaboration of various persons and entities, along with the AI system. This creation raises crucial questions for the IP laws across various jurisdictions, as there will be more such cases in the future where IP laws will need to address such issues for the proper determination of rights of the AI systems and the natural person.

    DABUS

    Interestingly, across various jurisdictions, such as Australia, the United Kingdom, the United States, New Zealand, and the European Patent Office, applications filed by Stephen Thaler to patent the inventions by his AI system Device for the Autonomous Bootstrapping of Unified Sentience (“DABUS”), were rejected in the abovementioned jurisdictions on the ground that their patent laws do not allow the grant of patent to an AI system as the inventor, since a common classification in their patent laws only considers a natural person to be named as the inventor in a patent application. The case of DABUS is a landmark in the field of patents and AI systems, as South African Companies and Intellectual Property Commission (“CIPC”) accepted Stephen’s Patent Cooperation Treaty, on 24th June 2021, for a patent in respect of inventions made by DABUS. In July of 2021, CIPC released a notice of issuance for the patent, and DABUS became the first AI system to be regarded as an inventor and granted a patent for its inventions.

    RAGHAV

    Like DABUS, the copyright office in India, albeit mistakenly, has recognised an AI system RAGHAV as the co-author of an artistic work and registered the application for copyright protection. However, in the first instance, when Ankit Sahni, the creator of the AI system RAGHAV, filed the application listing the AI system as the sole author of such work, had that application rejected by the copyright office. Later, the copyright office issued a notice for withdrawal of such registration, as it had mistakenly granted the registration and asked the human co-author Mr. Sahni to deliberate upon the legal status of the AI system RAGHAV. The application status is still displayed as ‘registered’ on the website of the copyright office, but the court is yet to decide on this issue.

    The decision of the court holds strong significance for the issues that surround AI systems and IP laws regarding copyright protection in India, as it will act as a precedent for similar cases in the coming time, given the sophistication and capabilities of the AI systems keep advancing with the passing of time.

    Creativity Machine

    In a landmark judgment by the US Court neither the owner of the machine (since it’s the machine that created work) nor the machine itself (since it is not a person) got registration as an author under copyright laws.

    The United States District Court for the District of Columbia, vide an order dated 18th of August 2023, in Stephen Thaler v. Shira Perlmutter, Register of Copyrights and Director of the United States Copyright Office, et al.[1], the principal issue of whether work autonomously generated by an AI system is copyrightable in the United States was discussed and decided upon. It was concluded by the honourable court after deliberating on the issue that, the Copyright Office acted properly in denying copyright registration for a work created without any human involvement.

    Mr. Stephen Thaler, the owner of a computer system called the ‘Creativity Machine’ capable of producing artworks independently, filed for a copyright to register the work created by the computer machine, listing it as the author and requesting that the copyright be granted to him as the owner of the computer machine. The application was denied by the copyright office on the grounds that the work did not have human authorship necessary to support a copyright claim[2]. The Plaintiff then filed for a reconsideration of his application contesting that AI systems should be acknowledged as an author where it otherwise meets the criteria for authorship, but this application was also rejected by the Copyright Office with the rationale that, “because copyright law is limited to ‘original intellectual conceptions of the author’, the Office will refuse to register a claim if it determines that a human being did not create the work”[3], the Plaintiff then made another request for reconsideration with the same reasoning as the previous reconsideration letter, and it was again refused by the Copyright Office Review Board on the grounds that ‘copyright protection does not extend to the creations of non-human entities’. The decision of the Copyright Office Review Board was then challenged by Mr. Stephen resulting in a lawsuit against the Register of the United States Copyright Office.

    The honourable United States District Judge, Beryl A. Howell, in this case discussed the issue, in the light of the Copyright Act of 1976, which provides copyright protection to “original works of authorship fixed in any tangible medium of expression, no known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device”. The “fixing” of the work in the tangible medium must be done “by or under the authority of the author”. So in order for the work to be eligible for copyright a work must have an “author”. The word “author” is not defined in the Copyright Act of 1976, definition of the term as taken from the dictionaries define it as, “one that is the source of some form of intellectual or creative work” and “the creator of an artistic work; a painter, photographer, filmmaker, etc”

    The literal interpretation of the statute was concluded by the Judge as, “By its plain text, the 1976 Act thus requires a copyrightable work to have an originator with the capacity for intellectual, creative, or artistic labor. Must that originator be a human being to claim copyright protection? The answer is yes”.

    Further, the intent of the Congress and the founding idea behind the Copyright Act of 1909, the predecessor to the current Copyright Act of 1976, provided that only a “person” could secure copyright for his work” was unambiguously limited to the work of human creators. The understanding that ‘authorship’ is synonymous with human creation and how to best encourage individuals to engage in that creation to promote science and useful arts, was central to the American copyright laws from their inception, as non-human actors need no incentivisation with the promise of exclusive rights under United States law, and copyright was not designed to reach them. This ideology was reflected in the relevant congressional report that in enacting the Copyright Act of 1976, Congress intended to incorporate the “original work of authorship” standard “without change” from the previous 1909 Act.

    Citing several judgements of the Supreme Court of the United States, where the court has also consistently recognized the human authorship requirement in its interpretation of the copyright law, and it was stated that the Plaintiff, Mr. Stephen can point to no case in which the court has recognised copyright in a work originating with a nonhuman. In Sarony, “the Court’s recognition of the copyrightability of a photograph rested on the fact that the human creator, not the camera, conceived of and designed the image and then used the camera to capture the image”, And in Mazer v. Stein, the prerequisite for copyrightability of a work it “must be original, that is, the author’s tangible expression of ideas”[4], Also in Goldstein v. California, “author” was defined as “an ‘originator’, ‘he to whom anything owes its origin.’”[5]. As evident, authorship is centred on the acts of human creativity, and the courts have uniformly declined to recognize copyright in the works created without any human involvement.

    In concluding the discussion in this case, the court held that, “the work at issue did not give rise to a valid copyright upon its creation’, and ‘the image autonomously generated by the plaintiff’s computer system was never eligible for copyright’, upholding the decision of the Register in denying copyright registration to a ‘work created absent any human involvement’”[6].

    The law makers across various jurisdictions did not frame IP laws, keeping AI and related technologies in mind. Back then the AI may have not be significant to attract attentions of law makers, however, currently it is an industry of the size of thousands of billions of dollars.

    In today’s time where considerable investment is being made in the development of AI technologies and it is a huge industry with global market size of USD 454.12 billion in 2022 and is expected to hit around USD 2,575.16 billion by 2032, progressing with a compound annual growth rate (CAGR) of 19% from 2023 to 2032.

    It is imperative that the issues surrounding AI generated IP are discussed by the legislators for framing laws appropriate to the present and future concerns regarding AI, and the otherwise orphan AI be legitimatised by enacting or amending existing IP laws.

    Author: Mr. Rajiv Sharma (Partner) and Mr. Ninad Mittal (Associate) at Singhania & Co. LLP. Views are personal.

    1. Civil Ac on No. 22-1564 (BAH).

    2. Id, Ex. D, Copyright Office Refusal Le er Dated August 12, 2019 at 1, ECF No. 13-4.

    3. Id, Ex. F, Copyright Office Refusal Le er Dated March 30,2020 at 1, ECF No. 13-6.

    4. Mazer v. Stein 347 U.S. 201, 214 (1954).

    5. Goldstein v. California 412 U.S. at 561 (quo ng Sarony, 111 U.S. at 58).

    6. Stephen Thaler v. Shira Perlmu er, Register of Copyrights and Director of the United States Copyright Office, et al. Civil Ac on No. 22-1564 (BAH).

    Next Story