Content uploaded by Anthony Wong
Author content
All content in this area was uploaded by Anthony Wong on Jun 24, 2021
Content may be subject to copyright.
This is the preprint manuscript accepted for publication by Springer in Artificial Intel-
ligence for Knowledge Management, Volume 614 of the IFIP Advances in Information
and Communication Technology series in 2021.
Ethics and Regulation of Artificial Intelligence
Anthony Wong1,2,3
1Managing Director, AGW Legal & Advisory, Sydney, Australia
2Vice-President, IFIP
3Past President, Australian Computer Society (ACS)
anthonywong@agwconsult.com
Abstract. Over the last few years, the world has deliberated and developed nu-
merous ethical principles and frameworks. It is the general opinion that the time
has arrived to move from principles and to operationalize on the ethical practice
of AI. It is now recognized that principles and standards can play a universal
harmonizing role for the development of AI-related legal norms across the globe.
However, how do we translate and embrace these articulated values, principles
and actions to guide Nation States around the world to formulate their regulatory
systems, policies or other legal instruments regarding AI? Our regulatory systems
have attempted to keep abreast of new technologies by recalibrating and adapting
our regulatory frameworks to provide for new opportunities and risks, to confer
rights and duties, safety and liability frameworks, and to ensure legal certainty
for businesses. These past adaptations have been reactive and sometimes piece-
meal, often with artificial delineation on rights and responsibilities and with un-
intended flow-on consequences. Previously, technologies have been deployed
more like tools, but as autonomy and self-learning capabilities increase, robots
and intelligent AI systems will feel less and less like machines and tools. There
is now a significant difference, because machine learning AI systems have the
ability ‘to learn’, adapt their performances and ‘make decisions’ from data and
‘life experiences’. This paper presented at the International Joint Conference on
Artificial Intelligence - Pacific Rim International Conference on Artificial Intel-
ligence in 2021 provides brief insights on some selected topical developments in
ethical principles and frameworks, our regulatory systems and the current debates
on some of the risks and challenges from the use and actions of AI, autonomous
and intelligent systems. [1]
Keywords: AI, Robots, Automation, Regulation, Ethics, Law, Liability, Trans-
parency, Explainability, Data Protection, Privacy, Legal Personhood, Job Tran-
sition, Employment.
1 Introduction
AI and algorithmic decision-making will over time bring significant benefits to many
areas of human endeavour. The proliferation of AI systems imbued with increasingly
2
complex mathematical and data modelling, and machine learning algorithms, are being
integrated in virtually every sector of the economy and society, to support and in many
cases undertake more autonomous decisions and actions.
Previously, technologies have often been deployed more like tools, as a pen or paint-
brush, but as autonomy and self-learning capabilities increase, robots and intelligent AI
systems feel less and less like machines or tools. AI will equip robots and systems with
the ability to learn using machine-learning and deep-learning algorithms. They will
have the ability to interact and work alongside us or to augment our work. They will
increasingly be able to take over functions and roles and, perhaps more significantly,
the ability ‘to make decisions’.
How much autonomy should AI and robots have to make decisions on our behalf
and about us in our life, work and play? How do we ensure they can be trusted, and that
they are transparent, reliable, accountable and well designed?
While technological advances hold tremendous promise for mankind, they also pose
and raise difficult questions in disparate areas including ethics and morality, bias and
discrimination, human rights and dignity, privacy and data protection, data ownership,
intellectual property, safety, liability, consumer protection, accountability and transpar-
ency, competition law, employment and the future of work and, legal personhood. In a
world, that is increasingly connected and where machinebased algorithms use available
data to make decisions that affect our lives, how do we ensure these automated deci-
sions are not opaque, appropriate and transparent? And what recourse do we have when
these decisions intrude on our rights, freedoms, safety and legitimate interests?
The base tenets of our regulatory systems were created long before the advances and
confluence of emergent technologies including AI (artificial intelligence), IoT (Internet
of Things), blockchain, cloud, quantum computing, to name a few. With the rise of
these technologies we have taken many initiatives to address their consequences by
recalibrating and adapting our regulatory frameworks to provide for new opportunities
and risks, to confer rights and duties, safety and liability frameworks, and ensure legal
certainty for business.
Sector-specific regulation has also been adopted and adapted to address market fail-
ures and risks in critical and regulated domains. These changes have often been reactive
and piecemeal, with artificial delineation of rights and responsibilities. There have been
many unintended consequences. More recently we have begun to learn from past mis-
haps, and these regulatory adaptations are now more likely to be drafted in technologi-
cally neutral way avoiding strict technical definition, especially when the field is still
evolving rapidly.
Emerging technologies are rapidly transforming the regulatory landscape. They are
providing timely opportunities for fresh approaches in the redesign of our regulatory
systems to keep pace with technological changes, now and into the future. AI is cur-
rently advancing more rapidly than the process of regulatory recalibration. Unlike the
past, there is now a significant difference—we must now take into consideration, ma-
chine learning AI systems that have the ability to learn, adapt their performances and
‘make decisions’ from data and ‘life experiences’.
3
The UN Secretary-General in his June 2020 report, commented that, “The world is
at a critical inflection point for technology governance, made more urgent by the ongo-
ing pandemic” [2]. He further emphasized the need to redouble our efforts to better
harness digital technologies while mitigating the harm that they may cause.
This paper presented at AI4KM 2020 at the International Joint Conference on Arti-
ficial Intelligence - Pacific Rim International Conference on Artificial Intelligence
(IJCAI-PRICAI 2020), Yokohama, Japan 7th January 2021, provides brief insights on
some selected topical developments in our ethical and regulatory systems and, the cur-
rent debates to address some of the challenges and risks from the use and actions of AI,
autonomous and intelligent systems. [1] The paper is partly based on the keynotes,
presentations and engagements in Australia [3], Malaysia [4], Zimbabwe [5], Cambodia
[6], Sri Lanka [7], Switzerland [8] and Brazil [9]. It extends on the paper published as
“The Laws and Regulation of AI and Autonomous Systems” [10].
The paper is organized as follows. Section 2 briefly reviews the state of ethical
principles and frameworks. Section 3 looks at the responsibility and liability
challenges for damages caused by AI. Section 4 discuss transparency and explainability
of AI and section 5 on the debates on Legal personhoods for AI. Section 6 briefly looks
at AI and Implications of Employment. Section 7 concludes this paper.
2 Ethical Issues Arising from AI
It is perhaps apt at this juncture, that I pause to reflect on the journey that has brought
me to the cross-roads of ethics and regulation of AI. In 2016, I initiated a series of
articles on AI for The Australian. The columns commenced with a piece on “Ethics
must travel as AI’s associate” [11], which was followed by a series of closely related
topics including: “How far should AI replace human sense?” [12], “We need plans for
when robots are in driver’s seat” [13], “Complex algorithms can use a little of that
human touch” [14], “AI: Are Musk and Hawking right, or is our future in our hands?”
[15], “Do robots and artificial intelligence think about copyright?” [16], “Data frame-
works critical for AI success” [17] and, “Who is liable when robots and AI get it
wrong?” [18].
These columns led to a series of interviews, panels, and presentations looking at the
possible risks that AI poses and the notion of building ethics into machine intelligence.
These included an ABC TV News interview on “Neurotechnologies and AI, privacy,
agency and identity, and bias”, panels to explore the social and ethical concerns of AI
and, submission to the consultation on Artificial Intelligence, Australia’s Ethics Frame-
work [3].
The many topics canvassed included:
• What happens when AI and algorithmic decision-making leads to someone
being disadvantaged or discriminated against? There have been numerous
instances where this has happened [19], not necessarily due to the algorithm
itself, but because the underlying data reflects an inherent bias, statistical
distortion or pattern that becomes obvious when the algorithm is applied to
it.
4
• How do you think traditional business models will be disrupted in the future
by AI?
• How will AI disruption of traditional business models impact society?
• What options does the government have to constrain or enable artificial in-
telligence and what should be its focus?
• What ethical considerations must be taken into account when developing
artificial intelligence and what are the priorities?
The conclusions derived from algorithms are probabilistic in nature and may carry
inherent biases, which may be replicated, amplified and reinforced. Algorithms are not
infallible. As algorithmic complexity and autonomy increase, it becomes imperative to
build in checks and balance to protect the legitimate interests of stakeholders [11].
If ethical parameters are programmed into AI, whose ethical and social values are
these? This question was foremost in my mind, when I presented on the “Ethical
Dimensions of AI & Autonomous Systems” to an audience studying Buddhist ethics
[20]. Each society, tradition, cultural group, religion, system and country view ethics
and morality through the contextual lenses of their underlying philosophical beliefs.
The variations in ethical and social values that underpin our global landscape are
challenging and, changes with the passage of time.
In response to the challenges articulated above, a range of stakeholders have pro-
duced AI ethical principles and frameworks. When I reviewed AI ethical principles and
frameworks produced by public, private, and non-governmental organizations in 2019,
there were more than 70 in existence. The number continues to grow. In 2019, jurisdic-
tions including Australia [21] and the EU [22] published their frameworks, adding to
the lists of contributors including the OECD Principles on Artificial Intelligence [23],
the World Economic Forum AI Governance: A Holistic Approach To Implement Ethics
Into AI [24] and the Singapore Model AI Governance Framework [25], to name a few.
An analysis of 84 principles and guidelines by Jobin et al [26] reveals a convergence
emerging around five ethical principles (transparency, justice and fairness, non-malef-
icence, responsibility and privacy).
By 2020, a study by Fjeld et al. [27] of 36 principles and guidelines, revealed an
extended list around eight key themes: (1) Privacy, (in 97% of documents), (2) Ac-
countability (in 97% of documents), (3) Safety and Security (in 81% of documents), (4)
Transparency and Explainability (in 94% of documents), (5) Fairness and Non-discrim-
ination (in 100% of documents), (6) Human Control of Technology (in 69% of docu-
ments), (7) Professional Responsibility (in 78% of documents), (8) Promotion of Hu-
man Values (in 69% of documents).
The UN Secretary-General in his 2020 report commented that “there are currently
over 160 organizational, national and international sets of artificial intelligence ethics
and governance principles worldwide” [28] and calls for a common platform to bring
these separate initiatives together.
UNESCO was given the mandate by its Member States to develop an international
standard-setting instrument on the ethics of artificial intelligence, which is to be sub-
mitted to the General Conference in the later part of 2021.The first draft of UNESCO’s
Recommendation on the Ethics of Artificial Intelligence was released to Member States
5
in late 2020 [29]. The Recommendation has largely been considered as an inter-disci-
plinary and multi-stakeholder initiative in light of the proliferation of ethical principles
and frameworks.
The Recommendation includes many common or shared ethical concepts and values
with an extended list around ten key themes: (1) proportionality and do no harm, (2)
safety and security, (3) fairness and non-discrimination, (4) sustainability, (5) privacy,
(6) human oversight and determination, (7) transparency and explainability, (8) respon-
sibility and accountability, (9) awareness and literacy and (10) multi-stakeholder and
adaptive governance and collaboration.
The debates have matured significantly since 2017, beyond the ‘what’ of ethical
principles to more of the ‘how’, with detailed guidelines on how such principles can be
operationalised in the design and implementation to minimise risks and negative out-
comes. But the challenge has always been putting principles into practice and creating
accountability mechanisms.
There is a growing consensus that the time has arrived to move from principles and
to operationalize on the ethical practice on AI [30]. Many of the proponents for regula-
tory intervention have argued that abstract high-level AI principles lack the specificity
to be used in practice and require legal enforcement mechanisms that are more robust
to provide redress when things go wrong. With the growing lists of AI related incidents,
there is a general distrust that AI developers could self-regulate effectively.
There is also a growing awareness that principles can play a useful base from which
to develop professional ethics, standards, and AI regulatory systems across the globe.
But how do we translate and embrace these articulated values, principles and actions to
guide Nation States in the formulation of their regulatory systems, policies or other
legal instruments regarding AI?
As stated by Fjeld et al., the impact of a set of principles is “likely to depend on how
it is embedded in a larger governance ecosystem, including for instance relevant poli-
cies (e.g. AI national plans), laws, regulations, but also professional practices and eve-
ryday routines” [31]. That view also resonated with those of UNESCO. UNESCO has
advocated for Member States to put in place policy actions and oversight mechanisms
to operationalize the values and principles in the Recommendation.
Due to the challenges in enforcing ethical principles or frameworks, we are seeing
greater regulatory impetus and focus to complement the gaps to improve the public’s
trust. We are seeing growing awareness that our existing regulatory frameworks are not
evolving fast enough to keep pace with the rapid progress in AI. Recently, UNESCO
and the EU Parliament have set the regulatory train in motion.
One of the objectives of the UNESCO Recommendation is to provide a universal
framework of values, principles and actions to guide Member States in the formulation
of their legislation, policies or other instruments regarding AI: “Member States should
develop, review and adapt, as appropriate, regulatory and legal frameworks to achieve
accountability and responsibility for the content and outcomes of AI systems at the
different phases of their life cycle. Member States should introduce liability frame-
works or clarify the interpretation of existing frameworks to ensure the attribution of
accountability for the outcomes and behaviour of AI systems” [32].
6
UNESCO has strongly advocated that AI cannot be a no law zone: “There are some
legislative vacuums around the industry which needs to be filled fast. The first step is
to agree on exactly which values need to be enshrined, and which rules need to be
enforced. Many frameworks and guidelines exist, but they are implemented unevenly,
and none are truly global. AI is global, which is why we need a global instrument to
regulate it” [34].
In October 2020, the European Parliament adopts 3 resolutions to regulate AI, setting
the pace as a global leader in AI regulation. The resolutions cover the ethical and legal
obligations surrounding AI, civil liability setting fines of up to 2 million euros for dam-
age caused by AI; and intellectual property rights [33]. In response, the European Com-
mission has indicated that it will publish draft legislation in 2021 addressing AI by
obliging high-risk AI systems to meet mandatory requirements related to their trust-
worthiness.
2021 will prove to be an interesting year for AI regulatory developments. However,
what will unfold, time will tell. Some of these ethical principles or frameworks may
well be adopted alongside or incorporated in legislation.
3 Responsibility and Liability for damages caused by AI
How should regulators manage the complexity and challenges arising from the design,
development and deployment of robots and autonomous systems? What legal and social
responsibilities should we give to algorithms shielded behind statistically data-derived
‘impartiality’? Who is liable when robots and AI get it wrong?
There is much debate as to who amongst the various players and actors across the
design, development and deployment lifecycle of AI and autonomous systems should
be responsible and liable to account for any damages that might be caused. Would au-
tonomy and self-learning capabilities alter the chain of responsibility of the producer
or developer as the “AI-driven or otherwise automated machine which, after consider-
ation of certain data, has taken an autonomous decision and caused harm to a human’s
life, health or property” [35]?
Or has “inserting a layer of inscrutable, unintuitive, and statistically-derived code in
between a human decisionmaker and the consequences of that decision, AI disrupts our
typical understanding of responsibility for choices gone wrong”? [36] Or should the
producer or programmer foresee the potential loss or damage even when it may be dif-
ficult to anticipate—particularly in unusual circumstances, the actions of an autono-
mous system? These questions will become more critical as more and more autonomous
decisions are made by AI systems.
One of the more advanced regulatory developments in AI is in the trialling of auton-
omous vehicles [37] and in the regulatory frameworks for drones [38].
The rapid adoption of AI and autonomous systems into more diverse areas of our
lives—from business, education, healthcare and communication through to infrastruc-
ture, logistics, defence, entertainment and agriculture—means that any laws involving
liability will need to consider a broad range of contexts and possibilities.
7
We are moving rapidly towards a world where autonomous and intelligent AI sys-
tems are connected, embedded and integrated in complex environments, and with “the
plurality of actors involved, it can be difficult to assess where a potential damage orig-
inates and which person is liable for it. Due to the complexity of these technologies, it
can be very difficult for victims to identify the liable person and prove all necessary
conditions for a successful claim, as required under national law” [39]. That view is
also reflected in the more recent European Parliament resolution with recommendations
to the Commission on a civil liability regime for artificial intelligence [40]. The burden
of proof in a tort fault-based liability system in some countries could significantly in-
crease the costs of litigation.
We will need to establish specific protections for potential victims of AI-related in-
cidents to give consumers confidence that they will have legal recourse if something
goes wrong.
One of the proposals being debated is for the creation of a mandatory insurance
scheme to ensure that victims of incidents involving robots and intelligent AI systems
have access to adequate compensation. This might be similar to the mandatory com-
prehensive insurance that owners need to purchase before being able to register a motor
vehicle [41]. The EU Parliament has recently also proposed for deployers of high-risk
AI to have mandatory liability insurance (€10m in the event of death and physical harm
and €2m for damage to property) [40].
Another approach is for the creation of strict liability rules to compensate victims
for potential harm caused by AI and autonomous systems along the lines of current
product liability laws in the EU and Australia. Strict liability rules would ensure that
the victim is compensated regardless of fault. The EU Parliament has proposed that
deployers of AI designated as “high-risk” would be strictly liable for any damage
caused by it. But who amongst the various players and actors should be strictly liable?
Whether the existing mixture of fault-based and strict liability regimes are appropri-
ate is also subject to much debate.
Introducing a robust regulatory framework with relevant input from industry, poli-
cymakers and government would create greater incentive for AI developers and manu-
facturers to reduce their exposure by building in additional safeguards to minimise the
potential risks to humanity.
4 Transparency and Explainability of AI
Algorithms are increasingly being used to analyse information and define or predict
outcomes with the aid of AI. These AI systems may be embedded in devices and sys-
tems and deployed across many industries and increasingly in critical domains, often
without the knowledge and consent of the user. Should humans be informed that they
are interacting with AI, on the purposes of the AI, and on the data used for the training
and evaluation?
To ensure that AI based systems perform as intended, the quality, accuracy and rel-
evance of data are essential. Any data bias, error or statistical distortion will be learned
8
and amplified. In situations involving machine learning—where algorithms and deci-
sion rules are trained using data to recognize patterns and to learn to make future deci-
sions based on these observations, regulators and consumers may not easily discern the
properties of these algorithms. These algorithms are able to train systems to perform
certain tasks at levels that may exceed human ability and raise many challenging ques-
tions including calls for greater algorithmic transparency to minimise the risk of bias,
discrimination, unfairness, error and to protect consumer interests.
Over the last few years legislators have started to respond to the challenge. In the
EU, Article 22 of the General Data Protection Regulation (GDPR) [42] gives individu-
als the right not to be subject to a decision based solely on automated decision-making
(no human involvement in the decision process), except in certain situations including
explicit consent and necessity for the performance of or entering into a contract. The
GDPR applies only to automated decision-making involving personal data.
In the public sector, AI systems are increasingly being adopted by governments to
improve and reform public service processes. In many situations, stakeholders and us-
ers of AI will expect reasons to be given for transparency and accountability of govern-
ment decisions which are important elements for the proper functioning of public ad-
ministration. It is currently unclear how our regulatory frameworks would adjust to
providing a meaningful review by our courts of decisions undertaken by autonomous
AI systems, or in what circumstances a sub-delegation by a nominated decision-maker
to an autonomous AI systems would be lawful. We may need to develop new principles
and standards and “to identify directions for thinking about how administrative law
should respond … that makes sense from both a legal and a technical point of view.
[43].
As machine learning evolves, AI models [44] often become even more complex, to
the point where it may be difficult to articulate and understand their inner workings—
even to people who created them. This raises many questions: what types of explanation
are suitable and useful to the audience? [45] How and why does the model perform the
way it does? How comprehensive does the explanation need to be—is an understanding
on how the algorithmic decision was reached required, or should the explanation be
adapted in a manner which is useful to a non-technical audience?
In the EU, the GDPR explicitly provides a data subject with the following rights:
a) rights to be provided and to access information about the automated decision-
making; [46]
b) rights to obtain human intervention and to contest the decision made solely by
automated decision-making algorithm; [47] and
c) places explicit onus on the algorithmic provider to provide “meaningful infor-
mation about the logic involved” in algorithmic decision, the “significance”
and the “envisaged consequences” of the algorithmic processing [48].
But how would these rights operate and be enforced in practice? With recent and
more complex non-linear black-box AI models, it can be difficult to provide meaningful
explanations, largely due to the statistical and probabilistic character of machine learn-
ing and the current limitations of some AI models—raising concerns including account-
ability, explainability, interpretability, transparency, and human control.
9
What expertise and competencies would be required from a data subject to take ad-
vantage of the rights or for the algorithmic provider to provide the above rights?
“In addition, access to the algorithm and the data could be impossible without the
cooperation of the potentially liable party. In practice, victims may thus not be able to
make a liability claim. In addition, it would be unclear, how to demonstrate the fault of
an AI acting autonomously, or what would be considered the fault of a person relying
on the use of AI” [49].
This opacity will also make it difficult to verify whether decisions made with the
involvement of AI are fair and unbiased, whether there are possible breaches of laws,
and whether they will hamper the effective access to the traditional evidence necessary
to establish a successful liability action and to claim compensation.
Should organisations consider and ensure that specific types of explanation be pro-
vided for their proposed AI system to meet the requisite needs of the audience before
starting the design process? Should the design and development methodologies adopted
have the flexibility to embrace new tools and explanation frameworks, ensuring ongo-
ing improvements in transparency and explainability in parallel with advancement in
the state of the art of the technology throughout the lifecycle of the AI system?
While rapid development methodologies may have been adopted by the IT Industry,
embedding transparency and explainability into AI system design requires more exten-
sive planning and oversight, and requiring input and knowledge from a wider mix of
multi-disciplinary skills and expertise.
New tools and better explanation frameworks need to be developed to instill the de-
sired human values and to reconcile the current tensions and trade-off between accu-
racy, cost and explainability of AI models. Developing such tools and frameworks is
far from trivial, warranting further research and funding.
5 Legal personhoods for AI
Historically, our regulatory systems have granted rights and legal personhood to slaves,
women, children, corporations and more recently to landscape and nature. Two of In-
dia’s rivers, the Ganga and the Yamuna, have been granted legal status. In New Zealand
legislation was enacted to grant legal personhoods to the Whanganui river, Mount Ta-
ranaki and the Te Urewera protected area. Previously, corporations were the only non-
human entities recognised by the law as legal persons.
“To be a legal person is to be the subject of rights and duties” [50]. Granting legal
personality [51] to AI and robots will entail complex legal considerations and is not a
simple case of equating them to corporations.
Who foots the bill when a robot or an intelligent AI system makes a mistake, causes
an accident or damage, or becomes corrupted? The manufacturer, the developer, the
person controlling it, or the robot itself? Or is it a matter of allocating and apportioning
risk and liability?
As autonomic and self-learning capabilities increase, robots and intelligent AI sys-
tems will feel less and less like machines and tools. Self-learning capabilities for AI
have added complexity to the equation. Will granting ‘electronic rights’ to robots assist
10
with some of these questions? Will human actors use robots to shield themselves from
liability or shift any potential liabilities from the developers to the robots? Or will the
spectrum, allocation and apportionment of responsibility keep step with the evolution
of self-learning robots and intelligent AI systems? Regulators around the world are
wrestling with these questions.
The EU is leading the way on these issues. In 2017 the European Parliament, in an
unprecedented show of support, adopted a resolution on Civil Law Rules on Robotics
[52] by 396 votes to 123. One of its key recommendations was to call on the European
Commission to explore, analyse and consider “a specific legal status for robots … so
that at least the most sophisticated autonomous robots could be established as having
the status of electronic persons responsible for making good any damage they may
cause, and possibly applying electronic personality to cases where robots make auton-
omous decisions” [53].
The EU resolution generated considerable debate and controversy, because it calls
for sophisticated autonomous robots to be given specific legal status as electronic per-
sons. The arguments from both sides are complex and require fundamental shifts in
legal theory and reasoning.
In an open letter, experts in robotics and artificial intelligence have cautioned the
European Commission that plans to grant robots legal status are inappropriate and “non-
pragmatic” [54].
The European Group on Ethics in Science and New Technologies, in its Statement
on Artificial Intelligence, Robotics and Autonomous Systems, advocated that the con-
cept of legal personhood is the ability and willingness to take and attribute moral re-
sponsibility. “Moral responsibility is here construed in the broad sense in which it may
refer to several aspects of human agency, e.g. causality, accountability (obligation to
provide an account), liability (obligation to compensate damages), reactive attitudes
such as praise and blame (appropriateness of a range of moral emotions), and duties
associated with social roles. Moral responsibility, in whatever sense, cannot be allo-
cated or shifted to ‘autonomous’ technology” [55].
In 2020, the EU Commission presented its “White Paper on Artificial Intelligence—
A European approach to excellence and trust for regulation of artificial intelligence
(AI)” [56] and a number of other documents including a “Report on the safety and
liability implications of Artificial Intelligence, the Internet of Things and robotics” [57]
for comments. The White Paper is non-committal on the question of endowing robots
with specific legal status as electronic persons. It proposes a risk-based approach to
create an ‘ecosystem of trust’ as one of the key elements of a future regulatory frame-
work for AI in Europe, so that the regulatory burden is not excessively prescriptive or
disproportionate.
I concur with the conclusions reached by Bryson et al [58] that the case for electronic
personhood is weak. With the current capabilities and state-of-the-art of AI systems, it
is essential that human stays ‘in the loop’. The negatives outweigh the benefits in the
current debate on shifting legal and moral responsibility to AI systems—at least for the
foreseeable future. That view is consistent with those reached by UNESCO: “when
developing regulatory frameworks, Member States should, in particular, take into ac-
count that ultimate responsibility and accountability must always lie with natural or
11
legal persons and that AI systems should not be given legal personality themselves”
[32].
In October 2020, the EU Parliament reversed its earlier resolution and makes it clear
that it would not be appropriate to grant legal personhood to AI [59].
As evidenced by the historical debates on the status of slaves, women, corporations
and, more recently landscape and nature, the question of granting legal personality to
autonomous robots will not be resolved any time soon. There is no simple answer to
the question of legal personhood, and one size will not fit all.
Should legal personhood for robots or autonomous systems eventuate in the future,
any right invoked on behalf of robots, or obligation enforced against them, will require
new approaches and significant recalibration of our regulatory systems. Legal person-
hood could potentially allow autonomous robots to own their creations, as well as being
open to liability for problems or negative outcomes associated with their actions.
6 AI and Implications of Employment
Over the past few years we have been inundated with predictions that robots and auto-
mation will devastate the workplace, replacing many job functions within the next 10
to 15 years. We have already seen huge shifts in manufacturing, mining, agriculture,
administration and logistics, where a wide range of manual and repetitive tasks have
been automated. More recently, cognitive tasks and data analyses are increasingly being
performed by AI and machines.
Historically, new technologies have always affected the structure of the labour mar-
ket, leading to a significant impact on employment, especially lower skilled and manual
jobs. But now the pace and spread of autonomous and intelligent technologies are out-
performing humans in many tasks and radically challenging the base tenets of our la-
bour markets and laws. These developments have raised many questions.
Where are the policies, strategies and regulatory frameworks to transition workers
in the jobs that will be the most transformed, or those that will disappear altogether due
to automation, robotics and AI?
Our current labour and employment laws, such as sick leave, hours of work, tax,
minimum wage and overtime pay requirements, were not designed for robots. What is
the legal relationship of robots to human employees in the workplace? In relation to
workplace safety— what liabilities should apply if a robot harms a human co-worker?
Would the ‘employer’ of the robot be vicariously liable? What is the performance man-
agement and control plan for work previously undertaken by human employees work-
ing under a collective bargaining agreement, now performed or co-performed with AI
or robots? How would data protection and privacy regulations apply to personal infor-
mation collected and consumed by robots? Who would be responsible for cyber secu-
rity and the criminal use of robots or AI? [60]
Are there statutory protection and job security for humans displaced by automation
and robots? Should we tax robot owners to pay for training for workers who are dis-
placed by automation, or should there be a universal minimum basic income for people
12
displaced? Should we have social plans, such as exist in Germany and France, if re-
structuring through automation disadvantages employees?
There are many divergent views on all these questions. All are being hotly debated.
Governments, policy makers, institutions and employers all have important roles to
play in the development of digital skills, in the monitoring of long-term job trends, and
in the creation of policies to assist workers and organisations adapt to an automated
future. If these issues are not addressed early and proactively, they may worsen the
digital divide and increase inequalities between countries and people.
ICT professionals are also being impacted as smart algorithms and other autonomous
technologies supplement software programming, data analysis and technical support
roles. With AI and machine learning developing at an exponential rate, what does the
future look like?
6.1 Case study - line between human and robo advisers in financial services
FinTech (financial technology) start-ups are emerging to challenge the roles of banks
and traditional financial institutions. FinTechs are rapidly transforming and disrupting
the marketplace by providing ‘robo-advice’ using highly sophisticated algorithms op-
erating on mobile and web-based environments. The technology is called robotic pro-
cess automation (RPA) and is becoming widespread in business, and particularly in
financial institutions. Robo-advice or automated advice is the provision of automated
financial product advice using algorithms and technology and without the direct in-
volvement of a human adviser [61].
Robo-advice and AI capabilities have the potential to increase competition and lower
prices for consumers in the financial advice and financial services industries by radi-
cally reshaping the customer experience. They are designed, modelled and programmed
by human actors. Often they operate behind the scenes 24/7 assisting the people who
interact with consumers. There are considerable tasks and risks involved in writing al-
gorithms to accurately portray the full offerings and complexity of financial products.
In 2017 Australia, after a number of scandals, introduced professional standards leg-
islation for human financial advisers [62]. These regulations set higher competence and
ethical standards, including requirements for relevant first or higher degrees, continuing
professional development requirements and compliance with a code of ethics. The ini-
tiatives were introduced into a profession already under pressure from the robo envi-
ronment.
Because robo-advice is designed, modelled and programmed by human actors,
should these requirements also apply to robo-advice? Should regulators also hold ICT
developers and providers of robots and autonomous systems to the same standards de-
manded from human financial advisers? What should be the background, skills and
competencies of these designers and ICT developers?
Depending on the size and governance framework of an organisation, various play-
ers and actors could be involved in a collaborative venture in the development, deploy-
ment and lifecycle of AI systems. These might include the developer, the product man-
ager, senior management, the service provider, the distributor and the person who uses
the AI or autonomous system. Their domain expertise could be in computer science, or
13
mathematics or statistics, or they might be an interdisciplinary group composed of fi-
nancial advisers, economists, social scientists or lawyers.
In 2016 the Australian regulator laid down sectoral guidelines [63] for monitoring
and testing algorithms deployed in robo-advice. The regulatory guidance requires busi-
nesses offering robo-advice to have people within the business who understand the “ra-
tionale, risk and rules” used by the algorithms and have the skills to review the resulting
robo-advice. What should be the competencies and skills of the humans undertaking
the role?
The EU General Data Protection Regulation (GDPR) [64] went further, by placing
an explicit onus on the algorithmic provider to provide “meaningful information about
the logic involved” [67]. In addition, GDPR provides an individual with explicit rights
including the rights to obtain human intervention, to express their point of view and to
contest the decision made solely by automated systems [66] that has legal or similarly
significant impact. GDPR applies only when AI uses personal data within the scope of
the legislation.
Revealing the logic behind an algorithm may potentially risk and disclose commer-
cially sensitive information and trade secrets used by the AI model and on how the
system works.
The deployment of robo-advice raises many new, interesting and challenging ques-
tions for regulators accustomed only to assessing and regulating human players and
actors.
7 Conclusion
This paper raises some of the major topical issues and debates relating to ethics of AI,
AI liability, transparency and meaningful AI explanation, aspects of data protection and
privacy, legal personhood, job transition and employment law.
In the wake of the 2020 “black lives matter” protests, a number of technology com-
panies have announced limitations on plans to sell facial recognition technology. There
have also been renewed calls for a moratorium on certain uses of facial recognition
technology that has legal or significant effects on individuals until appropriate legal
framework has been established [67].
The need to address AI and autonomous system challenges has increased in urgency
as the adverse potential impact could be significant in specific critical domains. If not
appropriately addressed, human trust will suffer, impacting on adoption and oversight
and in some cases posing significant risks to humanity and societal values.
From this brief exploration, it is clear that the values and issues outlined in the paper
will benefit from much broader debate, research and consultation. There are no defini-
tive answers to some of the questions raised—as for many, it is a matter of perspective.
I trust that this paper will embark you on your own journey as to what our future regu-
latory systems should encapsulate. Different AI applications create and pose different
benefits, risks and issues. The solutions that might be adopted in the days ahead, will
potentially challenge our traditional beliefs and systems for years to come. We are fac-
14
ing a major disruptive shift which is capable of dislodging some of our legal assump-
tions and may require significant rethink of some of our long-established legal princi-
ples—as we must now take into consideration, machine learning AI systems that have
the ability to learn, adapt and ‘make decisions’ from data and ‘life experiences’.
Technologists and AI developers understand better than most in relation to the trends
and trajectories of emergent technologies and their potential impact on the economic,
safety and social constructs of the workplace and society. Is it incumbent on them to
raise these issues and ensure they are widely debated, so that appropriate and intelligent
decisions can be made for the changes, risks and challenges ahead? Technologists and
AI developers are well placed to address some of the risks and challenges during the
design and lifecycle of AI-enabled systems. It would be beneficial to society for ICT
professionals to assist government, legislators, regulators and policy formulators with
their unique understanding of the strengths and limitations of the technology and its
effects.
Historically, our regulatory adaptations have been conservative and patchworked in
their ability to keep pace with technological changes. Perhaps the drastic disruptions
that COVID-19 has caused in our work, life and play beyond the normal will provide
sufficient impetus and tenacity to consider and re-think on how our laws and regulatory
systems should recalibrate with AI and autonomous systems, now and into the future.
1. This paper is for general reference purposes only. It does not constitute legal or professional
advice. It is general comment only. Before making any decision or taking any action you
should consult your legal or professional advisers to ascertain how the regulatory system
applies to your particular circumstances in your jurisdiction
2. United Nation, Report of the Secretary-General, Road map for digital cooperation: imple-
mentation of the recommendations of the High-level Panel on Digital Cooperation,
www.un.org/en/content/digital-cooperation-roadmap/, June 2020, p 3, www.un.org/en/con-
tent/digital-cooperation-roadmap/. Accessed January 2021
3. Neurotechnologies and AI, privacy, agency and identity, and bias, ABC TV News 24 No-
vember 2017; Re-engineering industries with Artificial Intelligence & the social contract -
The intended outcome, 26th International Joint Conference on Artificial Intelligence, Mel-
bourne August 2017, https://www.acs.org.au/insightsandpublications/media-re-
leases/00000121.html; Automation and The Nature of Work, ACS Canberra Conference,
August 2017; Robo-advice & FinTech: More Transparent, Honest & Reliable than Human
Actors? Chair and Panel, Sydney University Business School, November 2018; Future AI
Forum submission to the consultation, Artificial Intelligence, Australia’s Ethics Framework,
May 2019, https://consult.industry.gov.au/strategic-policy/artificial-intelligence-ethics-
framework/consultation/view_respondent?_b_index=60&uuId=309030657; Ethical Di-
mensions of AI & Autonomous Systems, Nan Tien Institute, Wollongong, Australia, August
24 2019
4. International ICT Infrastructure & Digital Economy Conference Sarawak (IDECS) 2017 –
Kuching, Sarawak, Malaysia, April 2017; Presentation to Global Conference on Computing
Ethics, Kuala Lumpur, Malaysia, August 2012.
5. Artificial Intelligence, the Good, and the Ugly, Victoria Falls, Zimbabwe, November 2017
6. AI and Employment Law Implications, LAWASIA 2018, Siem Reap, Cambodia, November
2018
15
7. The Ethics & Regulation of Artificial Intelligence, Keynote to NITC 2019 “Embrace Digi-
tal” Colombo, Sri Lanka, 9th October 2019
8. Duty of Care and Ethics on Digital Technologies, Internet Governance Forum (IGF), Ge-
neva, December 2017; How do we maximise the benefits of Innovative 4.0 technologies,
without unnecessary risks and consequences?, World Summit on the Information Society
(WSIS), Geneva, May 2019; AI: Complementing Codes of Ethics with Law and Regulation,
Session 131 “Living the standard – how can the Information and Knowledge Society live to
an ethical and FAIR Standard”, World Summit on the Information Society (WSIS), Geneva,
July 2020
9. Ethics & Regulation of Artificial Intelligence, SECOMU 2020 - “Artificially Human or Hu-
manly Artificial? Challenges for Society 5.0”, Brazil 18th November 2020
10. Wong A. (2020) The Laws and Regulation of AI and Autonomous Systems. In: Strous L.,
Johnson R., Grier D.A., Swade D. (eds) Unimagined Futures – ICT Opportunities and Chal-
lenges. IFIP Advances in Information and Communication Technology, vol 555. Springer,
Cham. https://doi.org/10.1007/978-3-030-64246-4_4
11. “Ethics must travel as AI’s associate”, The Australian, September 13, 2016
12. “How far should AI replace human sense?”, The Australian, September 27, 2016
13. “We need plans for when robots are in driver’s seat”, The Australian, November 22, 2016
14. “Complex algorithms can use a little of that human touch”, The Australian, January 24, 2017
15. “AI: Are Musk and Hawking right, or is our future in our hands?”, The Australian, August
23, 2017
16. “Do robots and artificial intelligence think about copyright?”, The Australian, September 5,
2017
17. “Data frameworks critical for AI success”, The Australian, October 4, 2017
18. “Who is liable when robots and AI get it wrong?”, The Australian, September 19, 2017
19. Refer to examples in the AI Incident database https://incidentdatabase.ai/summaries/inci-
dents
20. Ethical Dimensions of AI & Autonomous Systems, Nan Tien Institute, Wollongong, Aus-
tralia, August 24 2019
21. Australian AI Ethics Framework (2019). https://www.industry.gov.au/data-and-publica-
tions/building-australias-artificial-intelligence-capability/ai-ethics-framework. Accessed
2020/6/6
22. European Commission: Ethics guidelines for trustworthy AI (2019). https://ec.eu-
ropa.eu/digital-single-market/en/news/ethics-guidelines-trustworthy-ai. Accessed 2020/6/6
23. OECD, OECD Principles on Artificial Intelligence (22 May 2019),
https://www.oecd.org/going-digital/ai/principles/. Accessed 2020/6/20
24. World Economic Forum: AI Governance: A Holistic Approach to Implement Ethics into AI,
https://www.weforum.org/whitepapers/ai-governance-a-holistic-approach-to-implement-
ethics-into-ai. Accessed 2020/6/20
25. Singapore Model AI Governance Framework, https://www.pdpc.gov.sg/-/me-
dia/files/pdpc/pdf-files/resource-for-organisation/ai/sgmodelaigovframework2.pdf. Ac-
cessed 2020/6/20
26. Jobin, A., Ienca, M., Vayena, E.: The global landscape of AI ethics guidelines. Nat Mach
Intell 1, 389–399 (2019). https://doi.org/10.1038/s42256-019-0088-2
27. Fjeld, J., Achten, N., Hilligoss, H., Nagy, A., Srikumar, M.: Principled Artificial Intelli-
gence: Mapping Consensus in Ethical and Rights-Based Approaches to Principles for AI
(January 15, 2020). Berkman Klein Center Research Publication No. 2020-1,
https://ssrn.com/abstract=3518482 or http://dx.doi.org/10.2139/ssrn.3518482
16
28. United Nation, Report of the Secretary-General, Road map for digital cooperation: imple-
mentation of the recommendations of the High-level Panel on Digital Cooperation,
www.un.org/en/content/digital-cooperation-roadmap/, June 2020 p 18, www.un.org/en/con-
tent/digital-cooperation-roadmap/. Accessed January 2021
29. The first draft of the recommendation submitted to Member States proposes options for ac-
tion to Member States and other stakeholders and is accompanied by concrete implementa-
tion guidelines. The first draft of the AI Ethics Recommendation is available at
https://unesdoc.unesco.org/ark:/48223/pf0000373434
30. See also the opinion of the High Level Panel Follow-up Roundtable 3C Artificial Intelli-
gence - 1st Session.
www.un.org/en/pdfs/HLP%20Followup%20Roundtable%203C%20Artificial%20Intellige
nce%20-%201st%20Session%20Summary.pdf
31. Fjeld, J., Achten, N., Hilligoss, H., Nagy, A., Srikumar, M.: Principled Artificial Intelli-
gence: Mapping Consensus in Ethical and Rights-Based Approaches to Principles for AI
(January 15, 2020). Berkman Klein Center Research Publication No. 2020-1, p.5.
https://ssrn.com/abstract=3518482 or http://dx.doi.org/10.2139/ssrn.3518482
32. UNESCO first draft of the AI Ethics Recommendation, Resolution 68,
https://unesdoc.unesco.org/ark:/48223/pf0000373434
33. European Parliament three Resolutions on the ethical and legal aspects of Artificial Intelli-
gence software systems (“AI”): Resolution 2020/2012(INL) on a Framework of Ethical As-
pects of Artificial Intelligence, Robotics and related Technologies (the “AI Ethical Aspects
Resolution”), Resolution 2020/2014(INL) on a Civil Liability Regime for Artificial Intelli-
gence (the “Civil Liability Resolution”), and Resolution 2020/2015(INI) on Intellectual
Property Rights for the development of Artificial Intelligence Technologies (the “IPR for AI
Resolution”)
34. Elaboration of a Recommendation on ethics of artificial intelligence (unesco.org),
https://en.unesco.org/artificial-intelligence/ethics
35. The World Economic Forum; White Paper on AI Governance A Holistic Approach to Im-
plement Ethics into AI, p. 6. Geneva, Switzerland (2019), https://www.weforum.org/white-
papers/ai-governance-a-holistic-approach-to-implement-ethics-into-ai. Accessed 2020/6/9
36. Selbst, Andrew D.: Negligence and AI’s Human Users. In: Public Law & Legal Theory
Research Paper No. 20-01, p 1. UCLA School of Law (2018)
37. For a brief rundown of the regulatory frameworks and developments in selected countries
refer to the Australian National Transport Commission 2020, Review of ‘Guidelines for tri-
als of automated vehicles in Australia’: Discussion paper, NTC, Melbourne, pp. 16-18,
https://www.ntc.gov.au/sites/default/files/assets/files/NTC%20Discussion%20Paper%20-
%20Review%20of%20guidelines%20for%20trials%20of%20automated%20vehi-
cles%20in%20Australia.pdf, last accessed 2020/6/6. For examples of Australian legislation
refer to: Motor Vehicles (Trials of Automotive Technologies) Amendment Act 2016 (SA),
Transport Legislation Amendment (Automated Vehicle Trials and Innovation) Act 2017
(NSW), Road Safety Amendment (Automated Vehicles) Act 2018 (Vic)
38. For the new European Union drone rules refer to: https://www.easa.europa.eu/do-
mains/civil-drones-rpas/drones-regulatory-framework-background. For the Australia drone
rules refer to: https://www.casa.gov.au/knowyourdrone/drone-rules and the Civil Aviation
Safety Amendment (Remotely Piloted Aircraft and Model Aircraft—Registration and Ac-
creditation) Regulations 2019
39. European Commission: Report on the safety and liability implications of Artificial Intelli-
gence, the Internet of Things and robotics, COM (2020) 64 (Feb. 19, 2020), p 14,
17
https://ec.europa.eu/info/files/commission-report-safety-and-liability-implications-ai-inter-
net-thingsand-robotics_en. Accessed 2020/6/
40. European Parliament resolution of 20 October 2020 with recommendations to the Commis-
sion on a civil liability regime for artificial intelligence: www.europarl.eu-
ropa.eu/doceo/document/TA-9-2020-0276_EN.pdf
41. Australian National Transport Commission 2020, Review of ‘Guidelines for trials of auto-
mated vehicles in Australia’: Discussion paper, NTC, Melbourne, pp. 26-27,
https://www.ntc.gov.au/sites/default/files/assets/files/NTC%20Discussion%20Paper%20-
%20Review%20of%20guidelines%20for%20trials%20of%20automated%20vehi-
cles%20in%20Australia.pdf. Accessed 2020/6/6
42. General Data Protection Regulation (GDPR) art.22; Recital 71; see also Article 29 Data
Protection Working Party, 2018a, Guidelines on Automated individual decision-making and
Profiling for the purposes of Regulation 2016/679, 17/EN WP251rev.01, p.19. http://ec.eu-
ropa.eu/newsroom/article29/item-detail.cfm?item_id=612053. Accessed 2020/6/4
43. Cobbe, Jennifer.: Administrative Law and the Machines of Government: Judicial Review of
Automated Public-Sector Decision-Making. Legal Studies, p 3 (2019)
44. For the interpretability characteristics of various AI models, refer to ICO and Alan Turing
Institute: Guidance on explaining decisions made with AI (2020), annexe 2,
https://ico.org.uk/media/for-organisations/guide-to-data-protection/key-data-protection-
themes/explaining-decisions-made-with-artificial-intelligence-1-0.pdf. Accessed 2020/6/6
45. For the types of explanation that an organisation may provide, refer to ICO and Alan Turing
Institute: Guidance on explaining decisions made with AI (2020), p. 20,
https://ico.org.uk/media/for-organisations/guide-to-data-protection/key-data-protection-
themes/explaining-decisions-made-with-artificial-intelligence-1-0.pdf. Accessed 2020/6/6
46. General Data Protection Regulation (GDPR) art.15
47. General Data Protection Regulation (GDPR) art.22
48. General Data Protection Regulation (GDPR) arts.13-14
49. European Commission: Report on the safety and liability implications of Artificial Intelli-
gence, the Internet of Things and robotics, COM (2020) 64 (Feb. 19, 2020), p 15,
https://ec.europa.eu/info/files/commission-report-safety-and-liability-implications-ai-inter-
net-thingsand-robotics_en. Accessed 2020/6/9
50. Smith, B.: Legal personality. Yale Law J 37(3), 283-299 (1928), p 283
51. For a discussion on the concept and expression ‘‘legal personality’’ refer to Bryson, J. J.,
Diamantis, M. E., Grant, T.D.: Of, for, and by the people: the legal lacuna of synthetic per-
sons. Artificial Intelligence and Law. 25(3) (2017), p. 277
52. European Parliament: European Parliament resolution of 16 February 2017 with recommen-
dations to the Commission on Civil Law Rules on Robotics (2015/2103(INL)), https://eur-
lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52017IP0051. Accessed 2020/6/9
53. Ibid paragraph 59(f)
54. Refer http://www.robotics-openletter.eu/. Accessed 2020/6/9
55. European Group on Ethics in Science and New Technologies: Statement on Artificial Intel-
ligence, Robotics and ‘Autonomous’ Systems, p 10. European Commission, Brussels
(2018), http://ec.europa.eu/research/ege/pdf/ege_ai_statement_2018.pdf. Accessed
2020/6/9
56. European Commission: White Paper on Artificial Intelligence - A European approach to
excellence and trust, COM(2020) 65 (Feb. 19, 2020), https://ec.eu-
ropa.eu/info/sites/info/files/commissionwhite-paper-artificial-intelligence-feb2020_en.pdf.
Accessed 2020/6/9
18
57. European Commission: Report on the safety and liability implications of Artificial Intelli-
gence, the Internet of Things and robotics, COM (2020) 64 (Feb. 19, 2020), https://ec.eu-
ropa.eu/info/files/commission-report-safety-and-liability-implications-ai-internet-thing-
sand-robotics_en. Aaccessed 2020/6/9
58. Bryson, J. J., Diamantis, M. E., Grant, T.D.: Of, for, and by the people: the legal lacuna of
synthetic persons. Artificial Intelligence and Law. 25(3) (2017), pp. 273-291
59. Resolution 2020/2014(INL) on a Civil Liability Regime for Artificial Intelligence (the
“Civil Liability Resolution”), “any required changes in the existing legal framework should
start with the clarification that AI-systems have neither legal personality nor human con-
science, and that their sole task is to serve humanity” (Civil Liability Resolution, Annex,
(6)); European Parliament resolution of 20 October 2020 on intellectual property rights for
the development of artificial intelligence technologies (2020/2015(INI)), “that it would not
be appropriate to seek to impart legal personality to AI technologies and points out the neg-
ative impact of such a possibility on incentives for human creators”, para 13
60. This section on employment implications is based on the presentation to LAWASIA 2018
at Siem Reap, Cambodia - 4 November 2018: Artificial Intelligence & Employment Law
Implications
61. Definition from the Australian Securities & Investments Commission: Regulatory Guide
255 - Providing digital financial product advice to retail client, https://asic.gov.au/regula-
tory-resources/find-a-document/regulatory-guides/rg-255-providing-digital-financial-prod-
uct-advice-to-retail-clients/. Accessed 2020/6/6
62. Corporations Amendment (Professional Standards of Financial Advisers) Act 2017
63. Australian Securities & Investments Commission: Regulatory Guide 255 - Providing digital
financial product advice to retail client, https://asic.gov.au/regulatory-resources/find-a-doc-
ument/regulatory-guides/rg-255-providing-digital-financial-product-advice-to-retail-cli-
ents/. Accessed 2020/6/6
64. Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016
on the protection of natural persons with regard to the processing of personal data and on
the free movement of such data, and repealing Directive 95/46/EC (General Data Protection
Regulation), 2016 O.J. (L 119/1) [GDPR].
65. Ibid art. 15(1)(h)
66. Ibid art. 22(3)
67. Australian Human Rights Commission: Discussion Paper on Human Rights and Technology
(2019), p 104, https://humanrights.gov.au/our-work/rights-and-freedoms/publications/hu-
man-rights-and-technology-discussion-paper-2019, last accessed 2020/6/20; For a US per-
spective, refer to Flicker, Kirsten: The Prison of Convenience, The Need for National Reg-
ulation of Biometric Technology in Sports Venues In: 30 Fordham Intell. Prop. Media &
Ent.L.J. 985 (2020), p 1015, https://ir.lawnet.fordham.edu/iplj/vol30/iss3/7/. Accessed
2020/6/20
