Author: Saif Gilani
The notion of technology complementing human capabilities is not new. Society has already utilised portable and wearable technology. With Artificial Intelligence (AI) developing at an alarming rate, the human body is already being lined up as the next computer interface, creating enhanced humans. This poses an array of legal issues, particularly in the area of Intellectual Property (IP). In light of this development, I will argue that IP law may be characterised by a new conflict between the IP owner’s rights that are vested in an enhancement device and the personal autonomy of the beneficiary of the device – between the programmer and the enhanced human. Firstly, I will analyse the preliminary point about how the two legal regimes currently interact. Thereafter, I will examine the specific areas where legal disruption may arise: the acquisition and exercise of IP rights.
Due to the emerging nature of this topic, it is important to provide clear terminology. I will use the following definition of an ‘enhanced human’: a human who has undergone unconventional enhancement. In contrast to ‘conventional’ forms of enhancement, such as education or biochemistry, ‘unconventional enhancement’ will refer to the invasive (or partially invasive) utilisation of AI. Consequently, humans could benefit from the advanced information processing capabilities of AI, and its machine learning potential, in order to augment one’s cognition beyond normal capacity. For example, one can envisage enhanced perception, enhanced communication, or even enhanced memory. Therefore, this definition excludes radical enhancement, such as brain emulation and the integration of Artificial General Intelligence (augmenting every single aspect of oneself), which is typically associated with the term ‘cyborg.’ While the following discussion is reliant upon the continued development and widespread adoption of enhancement technologies, this is a foreseeable and attractive future.
- Legal Disruption?
Indeed, IP and personal autonomy have existed in isolation of one another. IP rights are intangible rights that reward and protect creations of the mind. Patents and copyright are the most applicable forms of IP for enhancement devices, with the former relevant to the hardware of the device and the latter relevant to the software. An inventor may apply for patent protection, which is a twenty-year, exclusive right that is granted for a product or process. Copyright protection exists for a person who has created an original literary or artistic work, with protection arising automatically.
Personal autonomy has been defined by the European Court of Human Rights (ECtHR) as the ‘ability to conduct one’s life in the manner of one’s own choosing,’ which can be extended to making decisions about one’s body.  Therefore, this can protect one’s personal choice and bodily integrity. Significantly, personal autonomy is not an absolute right; rather it is limited. This is where the interface with IP law arises. It may be legitimate for an external party to have rights in an implanted enhancement device, such that the IP owner is rewarded and protected for his or her creation, notwithstanding the fact that this may limit the enhanced human’s personal autonomy. However, it is unclear the extent to which IP rights embedded within AI enhancement technology will restrict one’s personal autonomy.
- Legal Challenges
- Transfer of IP Rights
Should the enhanced human have IP ownership, such concerns would be alleviated as there would be no external IP controller. However, this would necessitate the transfer of IP rights to the enhanced human – a process that is not currently facilitated by the law. This obstacle arises as IP rights are separate proprietary interests to tangible personal property rights, so the transfer of the latter does not automatically entail the transfer of IP rights. Consequently, irrespective of who owns the tangible property, the current regime would allow the programmer to retain IP rights (totally or partially) in a device that is physically and biologically intertwined with an enhanced human’s body. To overcome this, the IP rights could be licensed or explicitly assigned to the enhanced human. There are issues in each of these modes of transfer, however.
For enhanced humans, licensing would be a defective mode of transfer as it would not entail the conveyance of IP ownership – restrictions could still be imposed by the programmer against the enhanced human. One could argue that this would not be disruptive. Instead, it may seem that the law already legitimises such control. In Howard Florey Institute/Relaxin, considering whether the patenting of human genes amounted to modern slavery, the European Patent Office held that, ‘No woman is affected in any way by the present patent – she is free to live her life as she wishes and has exactly the same right to self-determination as she had before the patent was granted.’ Thus, the patentability of body parts was not considered overly restrictive to one’s personal autonomy. Applying this to enhancement devices would legitimise the restrictions that the IP owner may impose, rendering it non-disruptive.
However, such conclusion cannot be made in the context of enhanced humans. Licensing restrictions imposed on the enhanced human through the enhancement device could compromise the enhanced human’s self-determination and ability to live one’s life unperturbed. For example, the programmer may impose Terms & Conditions to restrict how the enhanced human can utilise the enhancement device. Additionally, the enhanced human may have to pay a royalty fee to the IP owner to use the device freely. This would restrict one’s bodily integrity as introducing such financial burden may limit one’s ability to enhance one’s body to the fullest. Moreover, it is questionable how enforcement would be achieved – currently, an IP owner of software content can ‘brick’ the programme, rendering it non-functional, or can withdraw technical support. Each outcome would produce unsafe and perverse results as enhanced humans will be physically and biologically dependent upon the enhancement device’s capabilities. For example, could a device that ought to enhance one’s senses but is rendered non-functional actually compromise the human’s normal capability? To overcome this issue, it has been suggested that IP adherence should be integrated into the enhancement device, so questions of enforcement become non-existent. That is problematic as it would limit the freedom upon which the enhanced human can utilise her enhancement to the full extent. Therefore, the licensing model exacerbates the tension between personal autonomy and the control that a programmer of an enhancement device may have.
It could be argued that forcing the programmer to assign some or all of the IP rights may resolve this issue. Such a conveyance would allow the enhanced human to retain full control, thus, protecting one’s personal autonomy. There are issues with this argument. Not only is it rare for consumers to hold IP rights, but introducing an obligatory assignation would create a more fundamental issue. Programmers need an incentive to develop and continuously update such technology. However, the more rights that a programmer is forced to assign, the less incentive there is to maintain the device. The lack of incentive could threaten the entire development of enhancement technology as it would be economically questionable why a company should invest in the technology. In addition, the lack of incentive to continuously update and improve such technology may mean that the devices are more likely to be compromised. Considering the invasive nature of these devices, this could have physical safety implications for the enhanced human. It may also expose the enhanced human to a cybersecurity risk, with the AI device generating vast quantities of the enhanced human’s cognitive data. Consequently, to safeguard one’s bodily control against potential physical or digital intrusions, forced assignation should not be considered.
Disruptively, the current methods of transferring IP rights in an enhancement device are unsuitable as new issues in the law are exposed. The licensing framework may limit one’s self-control in so far as it may restrict an enhanced human’s ability to conduct one’s life in a manner of her own choosing. Assignation, on the other hand, can overcome these issues, but forced assignation may hinder the development of enhancement devices.
- Exercise of IP Rights
Without ownership of the IP rights to an enhancement device, the enhanced human will be considered a user of the device. As a user, subjected to the rights of the IP owner, it is foreseeable that the exercise (or lack thereof) of IP rights may limit the bodily control of the enhanced human.
Regarding the patentability of the enhancement hardware, a patent owner may prevent the enhanced person making, using or selling the invention in the country for which the patent was granted. However, such features are less relevant in non-commercial contexts, so this has limited applicability to the everyday life of an enhanced human. Additionally, protection could be afforded under the principle of exhaustion – once the product in question is marketed with the consent of the IP owner, her right to control the distribution of the product will be extinguished. Thus, these rights have little practical relevance and are not material threats to personal autonomy.
Of greater significance is copyright protection. The exhaustion principle applies only to distributional control, so the IP owner may still have certain rights. The IP owner is protected against unauthorised copying of the software and has the moral right to oppose changes to the work that may harm his reputation. These rights could restrict the personal autonomy of the enhanced human as they limit the enhanced human’s freedom to access the software of his enhancement device. Access to the software may be useful to the enhanced human for the purposes of biohacking (self-improvement), a means to further improve his capability. Alternatively, access may be required to exploit the personal data of the enhanced human. As applied to software, the law has developed this area, offering protection against copyright infringement when the source code of a programme is studied to create another programme of the same functionality. But it is doubtful whether this protection should extend to an enhanced human who updates and improves the existing programme on an enhancement device (as opposed to using the information to create a similar programme). Furthermore, due to the safety issues inherent in biohacking, it may be in society’s interest to limit an enhanced human’s access to the device. This would ensure that only certified actors are able to update the enhancement technology, rather than allowing for an enhanced person who (absent an AGI device) may not have the competence to self-update. Therefore, while the exercise of copyright powers can limit the personal autonomy of the enhanced human, it can be justified in this limited instance.
However, tension could arise in the scenario where an enhanced human is not rewarded for a work or invention produced by virtue of an enhancement device that has machine learning capabilities. It is foreseeable that enhancement devices may self-update or may even self-replicate to produce duplicates of such software for the enhanced human. Extending beyond a mere tool that enables the creation, the enhancement device may autonomously produce the work, removing human intervention – whether that be from the programmer or the enhanced human.
Turning to the question of who ought to have IP ownership in this scenario, Igor Dzordov states that an enhanced human cannot have IP rights. But even if the enhanced human has neither personal property rights nor IP rights in the enhancement device, these comments are descriptively inaccurate and normatively problematic.
Descriptively, the law is unclear about the notion of authorship. This term implies that there must be a human element, so rewarding the AI device would be unsuitable. Instead, UK copyright law has been updated such that the ‘author’ of copyright material is viewed in an entrepreneurial manner – it is the person who makes the arrangement to create the work. However, it is unclear how proximate such person must be to the work. Rather than the programmer, the enhanced human would, physically, be the most proximate to the device as he would be directly involved in its operation. Due to this oversight, placing greater emphasis on the programmer, or even the AI device itself, would jeopardise the autonomy of the enhanced person as it would arbitrarily preclude the enhanced human from having IP control over his thoughts.
The normative dilemma stems from comments made by philosopher, Le Chapelier, stating that:
Writers deserve rights in the fruits of their thoughts, require rights for reasons of autonomy and personhood or in recognition of their natural law rights in their body and/or will be harmed emotionally or financially by being denied property or other rights.
Indeed, the data from an enhancement device could be considered as part of the extended mind of the enhanced human, and, thus, part of one’s ‘thoughts.’ This is justified as the device, and its data, is within the cognitive and biological domain of the enhanced human. Failing to protect the enhanced human in this instance can jeopardise the control that one has over his body. Consequently, denying property rights in one’s (extended) thoughts could deter persons from enhancing as it would restrict the benefits of undergoing enhancement. This shows the systemic issues that would arise from inadequately rewarding the boosted efforts of the enhanced human. Hence, the enhanced human ought to be affixed with the extended ‘mind’ (and the ‘fruits’) of the enhancement device.
This section assumed that the greater the control that an enhanced human has over her thoughts, the greater her autonomy. However, from a policy perspective, the enhanced human could benefit from limitations. As per the utilitarian perspective, IP is about innovation and sharing innovation. One may argue that AI-enhancement devices further this goal as enhancement devices may increase the spread of people who can create their own inventions. However, this could lead to a more fundamental threat. It has been argued that the ability for more people to create inventions may engender a ‘patent thicket,’ where the market becomes overcrowded with patents – this can equally be extended to an ‘IP thicket.’ For a business, it would not make economic sense to invest in the development of enhancement technology if the enhanced human – rather than the business – will be more greatly rewarded. Essentially, the greater the bodily control of the enhanced human, the less incentive there is for a business to develop such technology. Therefore, limitations may need to be placed upon the enhanced human’s control over creations of her extended mind to preserve the functionality of the system. This is in society’s interest as, otherwise, there could be no opportunity for people to enhance in the first place.
Overall, there is a strong argument for enhanced humans to have some control over products created by their enhancement device. The law is currently ambiguous on this point – in the remit of copyright, greater clarity is needed of the term ‘authorship.’ Regardless, it can be argued that the enhanced human – more than any other agent – should be rewarded for the fruits of his ‘extended mind.’ But perhaps, due to the policy concerns, the enhanced human should not be the only party that has protection. Failing to resolve these issues could be disruptive as it would mean that an enhanced human loses complete control of the fruits of his thoughts.
Enhanced humans will be disruptive to varying degrees. They may alter our current conceptions of legal principles that relate to the transfer of IP ownership and the exercise of IP rights.
Transferal of IP rights from the programmer to the enhanced human is unlikely as forced assignation would make the development of such technology economically questionable. Licensing is an inadequate alternative as it may overly restrict the utilisation of enhancement technology, and the enforcement of such conditions could be unsafe.
Considering the mechanisms of control in more detail, the essay analysed how the exercise of an IP holder’s rights pose risks. While limited problems arise regarding patent and copyright control, of greater concern is the ownership of products deriving from IP-protected enhancement devices. Currently, the law lacks clarity in this area. Since the enhancement device can be considered as an extension of one’s mind, a continued failure to clarify this issue could materially interfere with the enhanced human’s right to bear the fruits of his body.
In light of these concerns, the following statement will hold truer as the technology develops: ‘two areas that were once strangers [IP and personal autonomy] are now becoming increasingly intimate bedfellows.’ The advent of enhanced humans will add greater tension to the IP regime whilst, more significantly, exposing new legal irregularities in IP law. This may be exacerbated in the future: once the AI becomes so advanced that human involvement is completely eradicated, rather than an external human interfering with the autonomy of an enhanced human, it is possible that the AI device, itself, may interfere in such a way. Thus, policymakers must act to orientate the law to the world of enhanced humans in such a way that robot technology can complement the capabilities of humans in a legally satisfactory manner.
Books and Journals
- Bainbridge D, Intellectual Property (9th edn, Pearson Education Ltd 2012)
- Bently L and Sherman B, Intellectual Property Law (4th edn, OUP 2014)
- Blok P, The inventor’s new tool: artificial intelligence – how does it fit in the European patent system?  EIPR Vol 2 (No 2) 69
- Bostrom N & Sandberg A, Cognitive Enhancement: Methods, Ethics, Regulatory Challenges (2009) Sci Eng Ethics 311
- Carvalko J, Law and Policy in an Era of Cyborg-Assisted-Life: The Implications of Interfacing in-the-Body Technologies to the Outer World (2013) 2013 IEEE Int. Symp. Technology and Society (ISTAS) 204
- Goold I, The Legal Aspects of Cognitive Enhancement in Meulen R et al (eds.), Rethinking Cognitive Enhancement (Oxford, 2017)
- Hagger L and Hagger Johnson G, Super kids: Regulating the use of cognitive and psychological enhancement in children (2011) Law, Innovation and Technology 3(1):137-66
- Helfer L, Human Rights and Intellectual Property: Conflict or Coexistence? 5 Minn. Intell. Prop. Rev.47 (2003)
- Lambert P, Computer-Generated Works and Copyright: Selfies, Traps, Robots, AI and Machine Learning (2017) EIPR 12
- Laurie et al, Foresighting Futures: Law, New Technologies, and Challenges of Regulating for Uncertainty (2012) 4(1) Law, Innovation and Technology 1-33
- Naam R, More Than Human: Embracing the Promise of Biological Enhancement (Broadway Books, 2005)
- Pila J, Intellectual Property Rights and Detached Human Body Parts, Journal of Medical Ethics Jan 2014, 40(1) 27
- Quigley M, Ayihongbe S, Everyday Cyborgs: On Integrated Persons and Integrated Goods (2018) Medical Law Review Vol 26 (Issue 2) 276
- Schafer B et al, A Fourth Law of Robotics?: Copyright and the law and ethics of machine co-production (2015) Artificial Intelligence and Law Vol 23 (Issue 3) 217
- Sotala K & Yampolskiy R.V, Responses to Catastrophic AGI Risk: A Survey  Scr. 90 069501
- Commissioner for Human Rights, Human Rights and Disability: Equal Rights for all (2008) CommDH/Issue Paper 2
- Committee on Legal Affairs, Report with recommendation to the Commission on Civil Law Rules on Robotics (2017) (2015/2103 (INL))
- Standing Committee on the Law of Patents, Exceptions and Limitation to Patent Rights: Private and/or Non-Commercial Use (2013) SCP/20/3
- Eurasian Women’s Community, About Robots, Cyborgs, and their Intellectual Property Rights, <http://eawfpress.ru/en/about-robots-cyborgs-and-their-intellectual-property-rights/> accessed 8/4/18
- Koffeman N.R, (The right to) personal autonomy in the case law of the ECtHR (2010) <https://openaccess.leidenuniv.nl/bitstream/handle/1887/15890/N.R.+Koffeman+-+(The+right)+to+personal+autonomy+in+the+case+law+of+the+ECtHR+(2010).pdf> accessed 30/4/18.
- Newcastle Law School, The Everyday Cyborg: Mapping Legal, Ethical, & <https://www.ncl.ac.uk/media/wwwnclacuk/newcastleuniversitylawschool/files/Workshop%20Report%20-%20June%202017.pdf> accessed 8/4/18
- Neuralink <https://www.neuralink.com/> accessed 28/4/18
- Taylor Wessing LLP, Exhaustion of Rights and Digital Content (2013) <https://www.taylorwessing.com/download/article_exhaustion_of_rights.html> accessed 28/4/18
- Wittes B & Chong J, Our Future: Law and Policy Implications (2014) <https://www.brookings.edu/research/our-cyborg-future-law-and-policy-implications/> accessed 20/4/18.
- Howard Florey Institute/Relaxin  EPOR 541
- London County and Westminster Bank Limited v Tompkins  1 KB 515
- Moore v Regents of the University of California (1990) 793 P 2d 479 (Cal 1990)
- Pretty v UK (2002) 35 EHRR1
- R v Bentham  1 WLR 1057
- R v Brown  1 AC 212
- Riley v California (2014) 134 S.Ct.2743
- Santley v Wilde  2 Ch 474
- C-406/10 SAS Institute v World Programming  ECR-I 0000
- Yearworth v North Bristol NHS Trust  3 WLR 118
- Consolidated Version of the Treaty on the Functioning of the European Union  OJ C115/47
- Copyright, Designs and Patents Act 1988 c.48
- Patents Act 1977 c.37
- Scotland Act 1998 c.46
 The ‘IP owner’ will refer to the programmer of the device as it is realistic that the programmer would retain such rights.
 See N Bostrom & A Sandberg, ‘Cognitive enhancement: methods, ethics, regulatory challenges’ (2009) Sci Eng Ethics 311.
 B Wittes & J Chong, Our future: law and policy implications, available at https://www.brookings.edu/research/our-cyborg-future-law-and-policy-implications/.
 Patents Act 1977, (1977 Act) s 25(1).
 Ibid, s 7(1).
 Copyright, Designs and Patents Act 1988 (1988 Act) s 1(1).
 Pretty v UK (2002) 35 EHRR 1 paras 61 and 66.
 Commissioner for Human Rights, ‘Human rights and disability: equal rights for all’ (2008) CommDH/Issue Paper 2 para 5.2.
 D Bainbridge, Intellectual Property, 9th edn (2012) 12.
 1977 Act s 30; 1988 Act s 90. It is also possible to mortgage IP rights, but that requires an underlying debt.
 L Bently and B Sherman, Intellectual Property Law, 4th edn (2014) 644.
  EPOR 541.
 Ibid, 551.
 N Koffeman, (The right to) personal autonomy in the case law of the ECtHR, available at https://openaccess.leidenuniv.nl/bitstream/handle/1887/15890/N.R.+Koffeman+-+(The+right)+to+personal+autonomy+in+the+case+law+of+the+ECtHR+(2010).pdf.
 Ibid, 301.
 M Quigley & S Ayihongbe, ‘Everyday cyborgs: on integrated persons and integrated goods’ (2018) Medical Law Review Vol 26 (Issue 2) 276.
 B Schafer et al, ‘A fourth law of robotics?: copyright and the law and ethics of machine co-production’ (2015) Artificial Intelligence and Law Vol 23 (Issue 3) 217.
 J Carvalko, ‘Law and policy in an era of cyborg-assisted-life: the implications of interfacing in-the-body technologies to the outer world’ (2013) Proc. 2013 IEEE Int. Symp. Technology and Society (ISTAS) 204.
 1977 Act s 60.
 Standing Committee on the Law of Patents, ‘Exceptions and limitation to patent rights: private and/or non-commercial use’ (2013) SCP/20/3.
 Consolidated Version of the Treaty on the Functioning of the European Union  OJ C115/47 Arts 34-36.
 Taylor Wessing LLP, Exhaustion of rights and digital content, available at https://www.taylorwessing.com/download/article_exhaustion_of_rights.html.
 1988 Act ss 17 and 80.
 Case C-406/10 SAS Institute v World Programming  ECR-I 0000.
 P Lambert, ‘Computer-generated works and copyright: selfies, traps, robots, AI and machine learning’ (2017) EIPR 12.
 Eurasian Women’s Community, ‘About robots, cyborgs, and their intellectual property rights’, available at http://eawfpress.ru/en/about-robots-cyborgs-and-their-intellectual-property-rights/ .
 1988 Act s 9(3).
 J Pila, ‘Intellectual property rights and detached human body parts’ (2014) Journal of Medical Ethics 40(1) 27.
 P Blok, ‘The inventor’s new tool: artificial intelligence – how does it fit in the European patent system?’  EIPR Vol 2 (No 2) 69.
 Ibid, 73.
 Admittedly, by taking this approach, questions could be raised about whether the enhanced human should also be liable for the wrongs committed by the autonomous enhancement device.
 L Helfer, ‘Human rights and intellectual property: conflict or coexistence?’ (2003) 5 Minn. Intell. Prop. Rev.47.