AI & SOCIETY (2021) 36:319–329

https://doi.org/10.1007/s00146-020-01027-6

OPEN FORUM

Artificial intelligence and moral rights

Martin Miernicki1 · Irene Ng (Huang Ying)2

Received: 2 March 2020 / Accepted: 22 July 2020 / Published online: 3 August 2020
© The Author(s) 2020

Abstract
Whether copyrights should exist in content generated by an artificial intelligence is a frequently discussed issue in the legal
literature. Most of the discussion focuses on economic rights, whereas the relationship of artificial intelligence and moral
rights remains relatively obscure. However, as moral rights traditionally aim at protecting the author’s “personal sphere”, the
question whether the law should recognize such protection in the content produced by machines is pressing; this is especially
true considering that artificial intelligence is continuously further developed and increasingly hard to comprehend for human
beings. This paper first provides the background on the protection of moral rights under existing international, U.S. and
European copyright laws. On this basis, the paper then proceeds to highlight special issues in connection with moral rights
and content produced by artificial intelligence, in particular whether an artificial intelligence itself, the creator or users of an
artificial intelligence should be considered as owners of moral rights. Finally, the present research discusses possible future
solutions, in particular alternative forms of attribution rights or the introduction of related rights.

Keywords Copyrights · Moral rights · Artificial Intelligence · Authorship · Attribution · Contributorship

1 Introduction1 discussed legal questions include liability issues, especially

in the context of autonomous driving (Collingwood 2017),
Artificial Intelligence (AI)2 is often considered as a disrup- data protection (Kuner et al. 2018) as well as the protection
tive technology. The implications of this technology are not of AI and its products (Abbott 2016; Vertinsky and Rice
confined to the industrial sector, but do extent to numer- 2002) under copyright law (Grimmelmann 2016a, b). This
ous artistic fields, like the creation of music or works of art also relates to machine learning (Surden 2014). The copy-
(Bridy 2016; Niebla Zatarain 2018; Schönberger 2018). As right-related literature has, as far as can be seen, focused on
the technology promises great advances in a wide variety of the question whether economic rights exist or should exist in
contexts and areas of research, massive initiatives and invest- AI-generated content. However, the relationship between AI
ments are being undertaken; this also applies to the political and moral rights has not been studied to a comparable extent
level (European Commission 2018b). However, the use of (Miernicki and Ng 2019; Yanisky-Ravid 2017). Against this
AI can have far reaching consequences and many problems background, this research analyzes the relationship between
are still not fully explored. This relates, for instance, to the AI and moral copyrights. For the sake of completeness, we
technology’s philosophical and economic implications, but will also allude to economic rights in AI-generated content,
also to the legal framework that governs its use. Widely where appropriate.

1
The idea for this paper stems from a conference paper submitted to
* Martin Miernicki IRIS 2019 (Miernicki and Ng 2019).
2
[email protected] There is no universally accepted definition of AI; in fact, many
related concepts, like “robots”, “machine learning” and “AI” overlap
Irene Ng (Huang Ying)
(Lambert 2017). The European Commission, for instance, defines AI
[email protected]
as referring to “systems that display intelligent behavior by analyzing
1 their environment and taking actions – with some degree of auton-
Department of Business Law, University of Vienna, Vienna,
omy – to achieve specific goals” (European Commission 2018a). For
Austria
the purposes of this paper, we understand AI as the ability of com-
2
Singapore Management University Centre for AI and Data puter software to produce content that meets the requirements of
Governance, Singapore Management University, Singapore, copyright protection or that would be copyrightable if created by a
Singapore human being.

13
Vol.:(0123456789)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

320 AI & SOCIETY (2021) 36:319–329

The relation between AI and copyright can be studied of the work as well as the public in general will associate the
in two major respects. Commonly, the legal literature dis- work with its creator (Ciolino 1995), which is also linked to
tinguishes issues that arise “upstream” and those that arise a certain kind of social recognition and appreciation; recog-
“downstream”. Upstream refers to legal issues that arise nition for one’s work is sometimes deemed a “basic human
from making a machine capable of generating content desire” (US Copyright Office 2019). The right to integrity, in
which would include the implications of machine learning, turn, refers to the author’s interest not to have his or her work
or whether existing exceptions and limitations to copyright be altered drastically or used in a prejudicial way. Whether
would apply (or should be enacted) (Grimmelmann 2016a, there is an infringement of the right to integrity is very
b; Stewart 2014). Moreover, this relates to the questions dependent on the individual case as well as, importantly,
whether or how AI could be “taught” to avoid copyright the context of the use. Under this rule, moral rights could
infringement or “interpret” legal provisions (Schafer et al. be infringed if, for instance, a song is rearranged or used for
2015). However, for the purposes of this contribution, we a purpose completely different from the author’s intentions
will focus on the downstream issues of AI and copyright. (Ricketson and Ginsburg 2006).
This refers to the content generated by the AI and explores Moral rights came in special focus in the U.S. legal
the question whether or to what extent copyright law grants regime with the accession of the United States to the Berne
protection to this content; this naturally encompasses the Convention (Ciolino 1995). At that time, legislative changes
question to whom (or what) the law should allocate (exclu- were not made because the moral rights contained in the con-
sive) copyrights (Schafer et al. 2015; Schönberger 2018). vention were already, according to Congress’s opinion, pro-
vided for a sufficient extent under U.S. law (U.S. Copyright
Office 2019). Later, moral rights were explicitly recognized
2 Legal background in the Visual Artist Rights Act (17 U.S.C. § 106A); however,
the scope of the act is relatively narrow (U.S. Copyright
Moral rights acknowledge that authors have personal inter- Office 2019; cf. U.S. Court of Appeals 1st Circuit 2010). In
ests in their creations and the corresponding use that is made fact, the transposition of the Berne Convention’s require-
of them. These interests are conceptually different from the ments into U.S. law as regards moral rights has been a long
economic or commercial interests protected by the author’s source of controversy (Ginsburg 2004; Rigamonti 2006).
economic rights which are typically understood to enable the In any event, however, it is fair not to only look at the U.S.
author to derive financial gain from her creation (Rigamonti Copyright Act, but also at other laws on the federal level as
2007). Moral rights thus aim to protect the non-economic well as common law claims that can arise under state law,
interests of the author; this is often justified with reference for instance (Rigamonti 2007; U.S. Copyright Office 2019).
to a “presumed intimate” bond of the creator with his or her Against the background of the aim to strengthen the
creations (Rigamonti 2006). In this light, moral rights pro- internal market, European copyright law has focused on the
tect the personality of the author in the work (Biron 2014). harmonization of economic rights while moral rights have
Not surprisingly, different jurisdictions have varied takes never been harmonized to a comparable extent (Sirvinskaite
on this school of thought and do not interpret this ideol- 2010). To this end, a number of directives explicitly clarify
ogy to the same extent. In this regard, it is generally said that they are not meant to apply to moral rights which may or
that common law jurisdictions are more hesitant to granting may not be granted under the member states’ laws (Council
moral rights than civil law jurisdictions (Rigamonti 2007; of the European Communities 1993; European Parliament
Schére 2018). This might explain why—even though moral and Council of the European Union 1996, 2006; Euro-
rights can be found in various jurisdictions throughout the pean Parliament and the Council 2001, 2009, 2012, 2019).
world—the degree of international harmonization with The member states thus retain a great amount of freedom
regard to moral rights is rather low (Rigamonti 2006). The in respect of their moral rights legislation (von Lewinski
general principles are set forth by the Berne Convention and Walter 2010) and it appears that this area of copyright
(WIPO 1979); its article Art ­6bis states that “the author shall law will not be dealt with on the European level in the near
have the right to claim authorship of the work and to object future (cf. Commission of the European Communities 2004).
to any distortion, mutilation or other modification of, or At the same time, however, the lack of harmonization has
other derogatory action in relation to, the said work, which been called an “intrinsic obstacle” for the development of a
would be prejudicial to his honor or reputation.” As can be common European copyright legislation (Sirvinskaite 2010).
seen, the Berne Convention provides for two distinct moral As can be seen, there is—both on the international and
rights: The right of attribution and the right of integrity of the European level—little harmonization across borders
the work (Rigamonti 2006; cf. U.S. Court of Appeals 1st in the context of moral rights (Pettenati 2000) and even
Circuit 2010). The right of attribution generally includes the the degree of harmonization in the Berne Convention is
right to be recognized as the author of a work, so that users not uncontroversial. Strictly speaking, there is not even a

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

AI & SOCIETY (2021) 36:319–329 321

common concept of moral rights in countries that follow Since there is no general EU copyright code but rather
the civil law tradition. In general, one distinguishes the several directives with their respective scope, it is not easy
“Monist Theory” and the “Dualist Theory” (Chisolm 2018; to distill the concept of authorship under EU law. However,
Rigamonti 2007). While moral rights are genuinely based on it is possible to infer some guidance from the language of the
copyright and are closely integrated with economic rights different directives. On the one hand, the “own intellectual
under the Monist Theory, the Dualist Theory derives moral creation” standard is set forth in respect of databases, com-
rights from a general personality right (Rigamonti 2007). puter programs and photographs (European Parliament and
Hence, a clear distinction between moral rights and eco- Council of the European Union 1996, art. 3(1); European
nomic rights is not always possible. In fact, an author could Parliament and Council of the European Union 2006, art.
derive financial value from her right to be identified as the 6; European Parliament and the Council 2009, art. 1(3); see
author (“traditional” moral right) (Hansmann and Santilli also European Parliament and the Council 2019, art. 14).
1997; Tang 2012) or use her right to copy (“traditional” On the other hand, the “Copyright Directive” (European
economic right) to prohibit the reproduction of her work Parliament and the Council 2001) refers to “authors” and
for various non-economic reasons. Moreover, national laws “works”. With regard to first standard, the ECJ establishes
can grant additional moral rights within the framework of the connection to the author’s personality (European Court
the Berne Convention (e.g., withdrawal rights) (Ricketson of Justice 2011a; European Court of Justice 2012), a line
and Ginsburg 2006). However, hereinafter, when referring to of argumentation that can also be found in connection with
moral rights, we refer to moral rights as traditionally under- the “Copyright Directive” (European Parliament and the
stood (attribution and integrity) in order to highlight the Council 2001) (European Court of Justice 2008; European
special issues that occur in context of AI-generated content. Court of Justice 2011b). Accordingly, many commentators
conclude the human authorship is required under European
copyright law (Handig 2009; Ihalainen 2018; Miernicki and
3 Protection of AI‑generated content Ng 2019; Niebla Zatarain 2018).
Conversely, member states may decide to expand copy-
The key question of much of the copyright-related debate right protection to content produced by machines and AI
on AI is whether or to what extent copyrightable works (Miernicki and Ng 2019). In this connection, the UK Copy-
require or should require human action. Under the Berne right, Designs and Patents Act (United Kingdom 1988)
Convention, there is no clear definition of the concept of should be mentioned. Under this act, the author of a “com-
“authorship”. However, it is strongly suggested that the puter-generated” work (such a work “is generated by com-
convention only refers to human creators (Ginsburg 2018; puter in circumstances such that there is no human author of
Ricketson 1991), thereby excluding AI-generated content the work”, see s.178) is considered as “the person by whom
from its scope. This means that the minimum standard set the arrangements necessary for the creation of the work
forth by the Berne Convention only applies to works made are undertaken” (s.9) (MacCutcheon 2013). What is meant
by humans. by “arrangements”, however, remains a source of debate
U.S. copyright law affords protection to “original works (Davies 2011; Perry and Margoni 2010; Lambert 2017).
of authorship” (17 U.S.C. § 102(a)). This language is Most importantly, however, the CDPA exempts computer-
understood as referring to creations of human beings only generated works from the moral rights framework so that no
(Clifford 1997; U.S. Copyright Office 2017); accordingly, attribution or integrity rights are conferred upon the authors
the law denies copyright protection for the “creations” of of such works (s. 79, 81).
animals, the so-called “monkey selfie” case (U.S. District
Court Northern District of California 2016)3 being a notable
example for the application of these principle in the courts. 4 Ownership in AI‑generated content
This equally applies to content produced by machines (U.S.
Copyright Office 2017) or, in the present context, AI-gener- The copyrightability of AI-generated content is inherently
ated content (Abbott 2016; Yanisky-Ravid 2017). In conse- linked to the ownership of the corresponding copyrights.
quence, such content is not copyrightable, unless a human Much has been written about which persons (if any) should
author is found to have contributed creative input. be considered when assigning those rights (see, e.g., Bridy
2012; Davies 2011; Galajdová 2018; Glasser 2001; Hristov
2017; Perry-Margoni 2010; Wu 1997; Yu 2017). In the fol-
3
(“[T]here is no mention of animals anywhere in the [Copyright lowing, we discuss the AI itself, the creators of the AI and
Act]. The Supreme Court and Ninth Circuit have repeatedly referred the users of the AI as possible holders of moral rights.
to "persons" or "human beings" when analyzing authorship under the
Act […] Naruto is not an "author" within the meaning of the Copy-
right Act”).

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

322 AI & SOCIETY (2021) 36:319–329

4.1 AI as the owner of moral rights Schafer et al. 2015; Yanskiy-Ravid 2017). However, we
think it is fair to assume that machines have, already from
The concept of AI as a possible owner falls on the fact that today’s perspective, the potential to produce works which
AI is not simply an automatic system but could be an auton- would be indistinguishable from human creations (Holder
omous system. While the development of an AI system, et al. 2016) and thus would, in principle, meet the origi-
including data mining, machine learning and training pro- nality threshold for copyright protection. As can be seen,
cesses, are normally supervised by humans, recent advance- we consider the concept of creativity in strong proximity
ments in AI technology have enabled AI to learn from other to the originality threshold. This threshold is—generally
AIs, in a process called “kickstarting”.4 The “black box” speaking—rather low in the EU and the United States, for
of AI (Knight 2017), where developers of AI networks are instance.6 Thus, from this perspective, copyright protection
unable to explain why their AI programs have produced a should be considered.
certain result, further augments the belief of autonomy of However, in our opinion, it is questionable whether the
AI and its capability of having its own persona and decision- ability to produce copyrightable content is equivalent to hav-
making abilities. ing a personality sphere that moral rights aim to protect. It
However, it should be noted that any discussion centering seems that “creativity” or, as the law of many countries call
on AI as copyright owners is essentially a discussion de lege it, originality, embodied in a work is a basic prerequisite for
ferenda. This is because for machines to be granted rights, enjoying moral rights, as otherwise there would simply be no
some form of legal personality would be required (Bridy protectable subject matter (i.e. a work) (Rigamonti 2007).7
2012; Yu 2017). It is our impression that this is not the case Yet, moral rights are based on the additional concept of a
for many jurisdictions in Europe, America and Asia. In turn, “personality” that specifically addresses the author’s non-
whether machines should be granted some form of legal per- economic interests (Perry and Margoni 2010).8
sonality is in fact discussed on different levels. This relates,
Since the right of personality theory, much like the
of course, to academic literature (Günther et al. 2012; Pau-
moral rights orthodoxy, understands the work to be an
lius et al. 2017; Solaiman 2017), but also to legislative initia-
expression of the author’s personhood, it is easy to see
tives (Committee on Legal Affairs (2015).5 However, often
how authors complaining about unauthorized modifi-
times the discussion in this context centers on liability issues
cations of their works, for instance, could argue that
which are a different question from the grant of moral copy-
the modifications in question violated their personhood
rights (Miernicki and Ng 2019); the attribution of liability
expressed in their works (Rigamonti 2007).
is not the same as allocating exclusive rights, although the
former must certainly be considered when acknowledging This is, as stated above, a concept different from legal
AI as legal persons. This is especially true if a legislation personality in the sense of the attribution of liability. Thus,
granting AI the status of a legal person (similar to a company we consider the concept “personality” in strong proximity
or a partnership) would apply across other legislations unless to the non-economic interests protected by moral rights and
precluded by that specific legislation otherwise.
As aforementioned, moral rights, in the most basic sce-
nario, aim at protecting interests such as being recognized 6
“That being so, given the requirement of a broad interpretation
as the author of a work or object to certain modifications
of the scope of the protection conferred by Article 2 of Directive
to it. Hence, one must ask (1) whether an AI can have such 2001/29, the possibility may not be ruled out that certain isolated sen-
interests and (2) whether that the law deems such interest tences, or even certain parts of sentences in the text in question, may
worthy of protection against the background of the theoreti- be suitable for conveying to the reader the originality of a publication
such as a newspaper article, by communicating to that reader an ele-
cal basis of moral rights. The first aspect should be distin-
ment which is, in itself, the expression of the intellectual creation of
guished from the related question whether a machine can the author of that article. Such sentences or parts of sentences are,
be “creative” in the way humans are. This is, in fact, highly therefore, liable to come within the scope of the protection provided
debated (Bridy 2012; Bridy 2016; McCormack et al. 2019; for in Article 2(a) of that directive” (European Court of Justice 2008);
“These choices as to selection and arrangement, so long as they are
made independently by the compiler and entail a minimal degree of
creativity, are sufficiently original that Congress may protect such
4
Kickstarting here refers to the method of using a ‘teacher’ AI agent compilations through the copyright laws” (U.S. Supreme Court
to “kickstart the training of a new ‘student’ (AI) agent” (Schmitt et al. 1991).
7
2018). One could argue that, where moral rights are derived from a gen-
5
There, the Committee on Legal Affairs referred to a specific sta- eral personality right, copyrightable subject-matter is not a necessary
tus of “electronic persons”. It is noteworthy that the proposal received requirement.
8
serve criticism, see, e.g., Open Letter to the European Commis- To the extent that one considers economic rights to serve the pro-
sion, Artificial Intelligence and Robots, www.robot​ics-openl​etter​.eu tection of the author’s personal interests, this argument would in prin-
(accessed 14 January 2020). ciple apply also to those rights.

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

AI & SOCIETY (2021) 36:319–329 323

Table 1  Schematic view on the conceptual requirements for the grant for an AI, the question can be complicated by ethical and
of economic and moral copyrights religious concepts (Davies 2011).
Economic rights Moral rights Against this background, we can only give our opinion
whether moral rights for an AI would be consistent with
“Creativity” × ×
the development and rationales of moral rights as well as
Originality × ×
their foundations de lege lata. It appears to us that the moral
“Personality” (not in the sense of ×
rights ideology is closely connected to the creations made
attribution of liability)
by human beings (Miernicki and Ng 2019); thus, applying
Non-economic interests ×
moral rights to an AI would constitute an extension of the
underlying rationales of moral rights and would therefore
require additional justification.
distinguish these concepts from the originality—creativity
dichotomy. From a conceptual perspective, the requirements 4.2 Creators or users of an AI as the owners of moral
of both dichotomies must be fulfilled for moral rights to be rights
justified. This is illustrated by Table 1.
In light of the conceptual framework set forth above, Should humans that interact with the AI—i.e., the creators or
the question remains whether an AI can have a personal- the users—should have moral rights in the content produced
ity sphere and hence related non-economic interests. This by the AI? Where software is a tool used by humans, the
issue is, at first glance, related to technological developments general rules apply: Where the programmer contributes crea-
and contingent upon the definition of said terms.9 Against tive effort, she is awarded a copyright the extent of which
the background of moral rights theory, however, it seems is determined by national law. However, the situation is dif-
that this determination is not only based on strict logical ferent where the AI produces the content without original
reasoning. While economic rights can first and foremost input contributed a human being (Miernicki and Ng 2019).
be explained by the objective aim to incentivize the pro- In order to analyze this question, it is helpful to conceptu-
duction of creative works, moral rights appear to be highly alize the different roles of the AI and the involved humans as
based on ideological values, the legal recognition of which follows: The programmer creates the AI and holds, generally
is the task of the legislator and not lawyers or economists. speaking, a copyright in a literary work (cf. WTO 1994, Art
As already noted above, this is not to say that moral rights 9 et seq.). Thus, under ordinary circumstances, the AI con-
cannot be of economic value: e.g., the right to be named as stitutes copyrightable subject-matter and is referred to here-
the author can be used to build up the author’s reputation, inafter as “first generation work” because it directly stems
leading to increasing sales, or the author may (depending from the programmer’s creative effort (cf. Yanisky-Ravid
on the jurisdiction) choose not to execute this right against a and Velez-Hernandez 2018). Going a step further, if this
monetary payment when she serves as a ghostwriter (Davies work (software) generates new content, we could refer to this
2011). Furthermore, moral rights can also be seen as a way content as a “second generation work” since—provided that
to incentivize authors (cf. U.S. Court of Appeals 2nd Cir- the programmer did not intervene at all—it only indirectly
cuit 1995; U.S. Court of Appeals 1st Circuit 2010). How- stems from the programmer’s creative work.10 Now the ques-
ever, these examples appear to be rather ancillary effects; tion arises whether the creator has a special non-economic
the protection of the author’s personality remains the pri- connection to the “second generation work” that could be
mary objective and justification for the introduction of moral protected by moral rights. To answer this question, it is use-
rights (cf. U.S. Court of Appeals 2nd Circuit 1995; U.S. ful to recall the fundamental rationales of moral rights: these
Court of Appeals 1st Circuit 2010). The “author’s personal- rights are granted because the work represents “an extension
ity” is, of course, an elusive concept; indeed, not all juris- of the author’s personhood” (Rigamonti 2006); the author’s
dictions recognize the protection of non-economic interests personal traits are, so to say, “embodied” in the work (cf.
to the same extent. This can best be illustrated by the small Rosenthal Kwall 2010). Conversely, in absence of this cre-
degree of international harmonization with regard to moral ative endeavor, there is a lack of the intimate connection
rights (see above) and the inverse situation in respect of eco- between the creator of the AI and the produced content that
nomic rights. Especially with regard to a personality sphere moral rights have traditionally sought to protect (Miernicki

9
The problems of definition appear to be especially challenging; a
10
comparison with human behavior (e.g., responses to prejudices to Even though a one might see a certain resemblance, these „second
the “moral sphere”) is difficult because every human is affected dif- generation works “should in principle not be considered as derivative
ferently and the infringement of personality rights are far harder to works of the AI (Butler 1982; Lambert 2017; Perry-Margoni 2010;
quantify than economic damages. Wu 1997).

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

324 AI & SOCIETY (2021) 36:319–329

and Ng 2019). In this light, the relationship between pro- One could also look at this issue from the perspective of
grammer and the AI’s output is “radically mediated” (Bridy the idea-expression dichotomy. This dichotomy describes
2012; cf. Perry and Margoni 2010). Thus, from a conceptual the well-known principle that only expression is gener-
perspective, the “second generation work” does not carry the ally protected by copyright, whereas ideas are generally
programmer’s personal traits to an extent comparable to the not. In the case of content autonomously produced by
“first generation work” (Miernicki and Ng 2019). an AI, while one could consider the content as “expres-
As can be seen, this proposition constitutes the reverse sion”, there is a conspicuous absence of an idea which the
situation as observed in connection with the AI as holder expression emanates from (Glasser 2001; Schaefer et al.
of moral rights: This time, humans who can have non-eco- 2015; McCormack et al. 2019). Thus, the fact that the
nomic interests in the moral rights sense are involved as programmer—by creating the AI—“laid the foundations”
potential right holders; however, the degree of creativity in for the generation of new content is not enough to justify
the production of the “second generation work” is question- moral rights protection, in other words: It is not enough
able. This lack of creativity also seems to be the reason why to have the abstract intention to create “something”; the
moral rights are expressly excluded from the protection of intention to create is by itself not an expression of an indi-
CGW in the UK (Miernicki and Ng 2019). The legislative vidual personality protected by moral rights and neither
materials contain a passage stating that: will be the intention to create an “artwork” or a “text”.
As mentioned before, the situation is different where
[m]oral rights are closely concerned with the personal
the AI serves as a tool that helps the author to materialize
nature of creative effort, and the person by whom the
her creative vision (Glasser 2001). Only in this situation,
arrangements necessary for the creation of a computer-
the recognition and appreciation connected to the grant
generated work are undertaken will not himself have
of moral rights is well founded. Clearly, it must be deter-
made any personal, creative effort (U.K. House of
mined in the individual case whether the programmer or
Lords 1988).
user of an AI contributed creative input; this analysis is
In a similar fashion (however, with a different justifica- not meant to limit copyrightability to a minimum and can
tion) the WIPO Committee of Experts argued: in fact extend further than one might think. Consider the
“monkey-selfie” case referred to above. One might argue
Since computer-produced works have no identifiable
that the owner of the camera that was used by the monkey
authors, it is necessary to include specific provisions
should be awarded copyright protection in pictures shot
in the possible Protocol concerning original owner-
by the animal because he undertook arrangements and a
ship and the term of protection of copyright in such
creative selection process (Guadamuz 2016).
works; for the same reasons, moral rights would not
be applicable in the case of such works (WIPO 1991).
Against this background, granting moral rights would
amount to not only protecting the author’s “spiritual child” 5 Attribution, delineation and evidence
(cf. Ciolino 1995; Yanisky-Ravid 2017) but also its “spiritual
grand-child”. One could of course carry the idea a bit fur- Some authors highlight delineation issues, stating that it
ther: What if the AI generates another AI that itself produces can be very hard to determine whether (or to what extent)
“creative content”? Are there non-economic interests in such a human being contributed original content (Butler 1982;
a “third generation work” that could be harmed, e.g., by Denicola 2010; Guadamuz 2017; Hernandez 2018; Yani-
its modification? In our view, the connection to the human sky-Ravid 2017). While this is undoubtedly true (even in
author becomes genuinely elusive and an “intimate bond” is full knowledge of the facts of the case) this does not mean
increasingly hard to observe. In this regard, we suggest that that copyright (and, specifically moral rights) protection in
the principle of the “first generation work” and “second gen- AI-generated should be granted just for the sake of circum-
eration work” should apply similarly—i.e. this would be an venting or “streamlining” this issue. Very similar issues
“x generation work” that is traceable to the original creator arise in connection with co-authorship, where it can easily
of the first AI that spawned the subsequent AI-progeny. This be unclear whether a person, based on her contributions to
is in fact sometimes analogized with the relation between a the final work, should be considered as co-authors. While
parent and her children’s creations (Yanisky-Ravid 2017). this determination can require a cumbersome analysis,
This is even more apparent with users/operators (Schaefer one should not treat every person involved in any minimal
et al. 2015) of AI, because in many situations they will not way in the process of the creation of the work should as
have contributed any creative input (not even in the process a co-author, just to resolve the issue of authorship in a
of creating the “first generation work”), but merely triggered more efficient way. The law requires this analysis not to
or instigate the AI’s productive process.

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

AI & SOCIETY (2021) 36:319–329 325

Table 2  Schematic view on possible attribution rights of authors and “contributors” in AI-generated content
Purpose of attribution in general Purpose of attribution in case of AI-generated content

Author Acknowledgement of original contribution Acknowledgement of original contribution if an AI is used as a tool

“Contributor” Acknowledgement of non-original contribution (e.g., input Acknowledgement of non-original contribution (e.g., creation of
of abstract ideas, financial investments, administrative the AI; input of data; data selection to allow the AI to generate
tasks) the subsequent content)
AI – “Labelling” of non-human “creative” content

save costs, but rather to serve the greater goal of copy- situations, for example, if more than one person is involved
right law as well as a fair allocation of moral rights. 11 in the creation of a work. However, if one of these persons
Copyright law will always involve difficult delineation claims to be the sole author, there are other persons to con-
issues and many questions are—also in connection with test this assertion (Butler 1982). In the case of AI, it is, for
AI—not new (Grimmelmann 2016a, b). Alternative solu- the time being, hardly conceivable that the software would
tions have been proposed: This could relate, for instance, be able to claim and enforce any rights against its creator or
to a new category of attribution right that distinguishes user, even if the law grants the AI certain rights.
proper “authorship” from mere “contributorship” (Bently A straightforward solution would be to give copyrights
and Biron 2014).12 Applied to AI, this would mean that to “someone”, e.g., the users or creators of the AI (Samu-
the programmers or users of the AI would be considered elson 1986). However, similar concerns as observed with
as “contributors” rather than “authors”. This could be respect to substantive delineation problems arise. At least,
coupled with the indication that the work was produced different to the co-authorship situation, allocating the rights
by an AI. One could understand the latter not as an attri- in AI-generated content would not deprive any other per-
bution right for the AI, but rather as a form of “labelling sons of the rights because the alternative would be that
duty” when exercising the attribution right with regard to the output would fall in the public domain (Grimmelmann
“contributorship”. This solution could also address con- 2016a, b). However, the potentially enormous quantities of
cerns that human creativity would be endangered by AI- AI-produced content can generate a different kind of costs
generated content (cf. Schönberger 2018). The possible that relates to the enforcement of the rights that would exist
framework is outlined in Table 2 below.13 in this content. Consider, for instance, the “art experiment”
A practical problem that cannot be easily resolved—nei- centering around Qentis, a (fictive) company that claimed to
ther in connection with economic nor moral rights—relates have developed an AI that had generated virtually all pos-
to establishing whether a work is actually the product of sible texts between ten to four hundred words; the company
human creative endeavors. After all, why should the creator intended to collect royalties from anyone that would use
or user of an AI disclose that a content was AI-generated? In its content (Niebla Zatarain 2017). Even though it would
many cases, there will be a strong incentive to conceal this be practically infeasible to implement and maintain such a
fact to receive full protection under copyright law (Abbott system (Komuves et al. 2015), similar “business models”
2016; McCormack et al. 2019; Samuelson 1986) at least are not unlikely to arise if AI-generated was copyrightable
with regard to those jurisdictions that require human author- (Denicola 2016; Koboldt 1995).14 The opposing party would
ship. Admittedly, the evidentiary issues also occur in other then have to prove that the work was independently cre-
ated and not copied or modified. The costs of such potential
litigation should not be underestimated (Perry and Margoni
11
There are examples, however, where the law takes a pragmatic 2010; Wu 1997).
rather than an ideological approach to rights ownership; a prominent
example would be the copyright in cinematographic works (cf. WIPO
1979, art. 14bis).
12
In this different forms, this is already practiced in several contexts,
cf., e.g. “All contributors who do not meet the criteria for authorship 14
There are several other examples that suggest the ability of AI sys-
should be listed in an Acknowledgements section” (BMJ Author Hub tems to produce ungraspable amounts of content that are most likely
2018). too great to ever be consumed or used by human beings. Clearly, one
13
Note that we do not consider under which circumstances two per- could ask whether copyright is bluntly meant to foster larger and
sons can be considered as coowners of a work. This is essentially a larger quantities of protected works or, rather, to find the optimal
question of national law; see, e.g., for U.S. law 17 U.S.C. § 201(a); number of protected works. For the optimal level of copyright pro-
“A “joint work” is a work prepared by two or more authors with the tection, see, e.g. Koboldt 1995: “[…] the intensity of copyright pro-
intention that their contributions be merged into inseparable or inter- tection should not induce the production of the maximum number of
dependent parts of a unitary whole” (17 U.S.C. § 101). works […]”); see also Lunney 2014.

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

326 AI & SOCIETY (2021) 36:319–329

Conversely, there could be situations where the creator or 6 A related rights approach?
user of an AI does not want to be associated with the AI that
generated the content. This could involve situations where In light of the foregoing, we believe that granting moral
the AI has “run amok” and produces distasteful or illegal rights in AI-generated content is, in principle, not com-
content that the attribution to such content would result patible with the traditional rationale of these rights. Apart
in reputational damage or potential criminal charges. The from alternative models to the attribution right that should
“black box” of AI makes such a situation even more likely be considered, it remains to discuss whether or to what
as developers of AI networks are unable to explain or con- extent there should be economic rights and how such rights
trol the results of their creations; in fact, Microsoft’s experi- would fit in the copyright system. This is in many respects
ment to train its Twitter chatbot, Tay, on the social media a question of whether such rights can produce beneficial
platform Twitter, is an example of such an AI system that incentives (either for investing in the development of AI or
went rogue. Tay was intended to learn how to understand publishing its results), a question which has been discussed
conversations by mimicking the language of Twitter users; at length (Davies 2011; Glasser 2001; Grimmelmann 2016a,
however, less than 24 h after the chatbot was put online, b; McCutheon 2013; Perry and Margoni 2010; Samuelson
Microsoft stopped posts from that account due to it post- 1986; Yu 2017); we have already made some arguments
ing “obscene statements” (Victor 2016). Should a criminal above that would also apply to economic rights and do not
case be lodge against an AI developer for creating an AI delve further into this debate in this research. For the present
system that produces illegal content, the concept of whether purposes, and from the moral rights perspective, it would
an AI can be attributed moral copyright and whether the be a regulatory perspective to grant—similar to the solution
test to determine when an AI developer would be liable for found in the UK—certain economic rights but no or only
his creation becomes relevant. However, it should be high- limited moral rights; this resembles the legal situation with
lighted that the attribution of liability and moral rights are regard to related rights (Miernicki and Ng 2019).15 Since
separate and distinct legal concepts, and different laws would such a “middle-ground solution” might be more easily com-
be administered under both the common law and civil law patible with the different views that exist on moral rights,
systems. e.g., with regard to their scope, transfer or ownership by
A radical solution to the evidentiary problems with legal persons (Miernicki and Ng 2019; cf. Denicola 2016;
respect to the contribution of a human being would be to Ory and Sorge 2019), it might be more likely to find a con-
shift the burden of proof to the person who claims to be the sensus for the international harmonization of this matter.
author, requiring her in a given case of dispute to establish However, also a related rights solution runs the risk of trig-
that the content in question stems from a process of human gering a potential proliferation of protected content.
creation in case of doubt or upon the other party’s request.
Clearly, this appears to be a burdensome task. However, it
might just be the case that lawyers and courts currently lack 7 Conclusion
the experience with respect to such proceedings in prac-
tice, but such experience can be acquired over time. What is As a general principle, we believe that no moral rights
more, if the content was actually created by a human being, should be granted in AI-generated content, based on existing
it should not be too hard to establish this fact. In any event, laws and principles. While there is extensive debate on the
the shift of the burden of proof might not appear to be the possibility of AI having its own legal personality and thus
most pressing issue today, it could be more relevant in case the possibility of AI being capable of moral rights, much
machine-made contents proliferates; in fact, already from of this discussion is centered around AI liability—which
today’s perspective, there is no safe way to tell how much has different considerations vis-à-vis moral rights. Even in
AI-generated content has already been disseminated (Grim- light of arguments for AI to be granted moral rights, there
melmann 2016a, b). is nuance in the type of moral rights that should or should
A further issue relates to unjustified claims of authorship; not be granted to AI. First, it might be necessary to distin-
in this light, absent a copyright-based attribution right, it is guish between the different forms of moral rights: While
worth considering whether such claims can be addressed right to integrity is perhaps fundamentally rooted in the
by employing instruments offered by other legal fields, e.g. personal sphere of the author, the attribution right can also
the laws against unfair competition or potentially trademark
law (Tang 2017).
15
However, moral rights can also be found in this field. Yet, espe-
cially in case of performers’ rights, one could argue that a perfor-
mance is also something very individual which is connected to the
performer’s personality.

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

AI & SOCIETY (2021) 36:319–329 327

be explained on the basis of other foundations. In this con- Code availability Not applicable.
nection, intermediate solutions are conceivable, such as the
introduction of “contributorship” right, although there are Compliance with ethical standards
practical problems such as establishing whether the work is
indeed the product of human creative endeavors and associ- Conflict of interest The author(s) declare that they have no conflict of
interest.
ated evidentiary problems. One possible strand of thought16
in favor of acknowledging moral rights is that these rights Open Access This article is licensed under a Creative Commons Attri-
serve a higher public function—attributing moral rights to bution 4.0 International License, which permits use, sharing, adapta-
the original creators (who arguably have the greatest interest tion, distribution and reproduction in any medium or format, as long
in protecting their own works) could not only be understood as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
as serving the authors’ own interest but also the public inter- were made. The images or other third party material in this article are
ests in the integrity and societal status of creative works in included in the article’s Creative Commons licence, unless indicated
general.17 If such a public function exists, there may be an otherwise in a credit line to the material. If material is not included in
argument for protecting AI works for the greater good and the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
thus a responsible person should be designated under the law need to obtain permission directly from the copyright holder. To view a
to protect AI generated works. While national laws that pro- copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/.
vide for perpetual moral rights might point in this direction,
we would like to reserve this discussion for future research.
Clearly, it is not easy to see where the technological References
development will take us; however, in connection with moral
rights, a fundamental question becomes apparent: When will Abbott R (2016) I Think, Therefore I invent: creative computers and
AI systems have a “personality” that the law recognizes? In the future of patent law. Boston College L Rev 57:1079–1126
Bently L, Biron L (2014) Discontinuities between legal conceptions of
fact, why should anyone program an AI with interests of authorship and social practices: what, if anything, is to be done.
“personal nature” that could be protected by moral rights? In: van Eechoud M (ed) The work of authorship: creativity that
Perhaps, it is indeed as what some philosophers have argued counts. Amsterdam University Press/OAPEN library, Amsterdam,
on the topic of an AI being an artist—“creativity is, and pp 237–276
Biron L (2014) Creative Work and communicative norms. In: van
always will be, a human endeavor” (Kelly 2019). And, Eechoud M (ed) The work of authorship: creativity that counts.
even if this should be the case at some point we still have to Amsterdam University Press/OAPEN library, Amsterdam, pp
make the decision from an ideological standpoint whether 20–44
we should recognize AI, in this respect, as equal to us; in BMJ Author Hub (2018) BMJ policy on authorship. https​://autho​
rs.bmj.com/polic​ies/bmj-polic​y-on-autho​rship​/. Accessed 4 Nov
this light, granting moral rights to AI would truly start the 2019
“post human era” (Stewart 2014). However, if this era starts, Bridy A (2012) Coding creativity: copyright and the artificially intel-
“copyright will be the least of our concerns” (Grimmelmann ligent author. Stan Tech L Rev 5:1–28
2016a, b; see also Clifford 1997). Bridy A (2016) The evolution of authorship: work made by code.
Colum J L Arts 39:395–401
Butler TL (1982) Can a computer be an author—copyright aspects of
Acknowledgements Open access funding provided by University of artificial intelligence. Hastings Comm Ent L J 4:707–747
Vienna. Čerka P, Grigienė J, Sirbikytė G (2017) Is it is possible to grant legal
personality to artificial intelligence software systems? Comp L
Author contributions This paper was written in collaboration of the Secur Rev 33:685–699
two authors indicated above. Chisolm TE (2018) In Lieu of Moral Rights for IP-Wronged music
vocalists: personhood theory, moral rights, and the WPPT revis-
Funding Martin Miernicki: Not applicable; Irene Ng (Huang ited. St John’s L Rev 92:453–507
Ying): This research is supported by the National Research Founda- Ciolino DS (1995) Moral Rights and real obligations: a property-law
tion, Singapore under its Emerging Areas Research Projects (EARP) framework for the protection of authors’ moral rights. Tulane L
Funding Initiative. Any opinions, findings and conclusions or recom- Rev 69:935–995
mendations expressed in this material are those of the author(s) and Clifford RD (1997) Intellectual property in the era of the creative com-
do not reflect the views of National Research Foundation, Singapore. puter program. Tulane L Rev 71:1675–1703
Collingwood L (2017) Privacy implications and liability issues of
Data availability Not applicable. autonomous vehicles. Info Comm Tech L 26:32–45
Commission of the European Communities (2004) Commission Staff
Working Paper on the review of the EC legal framework in the
field of copyright and related rights. https​://data.consi​lium.europ​
16
We thank an anonymous reviewer for contributing this idea during a.eu/doc/docum​ent/ST-11634​-2004-INIT/en/pdf. Accessed 14
the review process. Jan 2020
17
Cf. Rushton 1998: “If the point of moral rights is to preserve those Committee on Legal Affairs (2015) Report with recommendations to
aspects of a culture that bind a nation together, then a case could be the Commission on Civil Law Rules on Robotics 2015/2103(INL),
made for a perpetual moral right.”

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

328 AI & SOCIETY (2021) 36:319–329

P8_TA(2017)0051. https:​ //www.europa​ rl.europa​ .eu/doceo/​ docum​ Grimmelmann J (2016a) Copyright for literate robots. Iowa L Rev
ent/TA-8-2017-0051_EN.pdf. Accessed 14 Jan 2020 101:657–681
Council of the European Communities (1993) Council Directive 93/83/ Grimmelmann J (2016) There’s no such thing as a computer-authored
EEC of 27 September 1993 on the coordination of certain rules work—and it’s a good thing, too. Colum J L Arts 39:403–416
concerning copyright and rights related to copyright applicable to Guadamuz A (2016) The monkey selfie: copyright lessons for origi-
satellite broadcasting and cable retransmission. O.J. L 1993/248, nality in photographs and internet jurisdiction. Internet Policy
15–21 Rev 5:1–12
Davies CR (2011) An evolutionary step in intellectual property Guadamuz A (2017) Do androids dream of electric copyright? Com-
rights—artificial intelligence and intellectual property. Comp L parative analysis of originiality in artificial intelligence generated
Secur Rev 27:601–619 work. Intellectual Property Quarterly 2:169–186
Denicola RC (2018) Ex Machina: copyright protection for computer- Günther J, Münch F, Beck S, Löffler S, Leroux C, Labruto R (2012)
generated works. Rutgers U L Rev 69:251–287 Issues of privacy and electronic personhood in robotics. In: IEEE
Durham AR (2002) The random muse: authorship and indeterminacy. (ed.) The 21st IEEE International Symposium on Robot and
Wm Mary L Rev 44(2):569–642 Human Interactive Communication
European Commission (2018a) Communication from the Commission Handig C (2009) The Copyright Term “Work”—European harmonisa-
to the European Parliament, The European Council, The Council, tion at an unknown level. IIC 40:665–685
The European Economic and Social Committee and the Commit- Hansmann H, Santilli M (1997) Authors’ and artists‘ moral rights:
tee of the Regions, Artificial Intelligence for Europe. COM(2018) a comparative legal and economic analysis. J Legal Stud
237 final 26:95–143
European Commission (2018b) Communication from the Commission Holder C, Khurana V, Hook J, Bacon G (2016) Robotics and law: key
to the European Parliament, the European Council, the Council, legal and regulatory implications of the robotics age (part II of
the European Economic and Social Committee and the Commit- II). Comp L Secur Rev 32:557–576
tee of the Regions, Coordinated Plan on Artificial Intelligence. Hristov K (2017) Artificial Intelligence and the Copyright Dilemma.
COM(2018) 795 final IDEA 57:431–454
European Court of Justice (2008) Infopaq Int’l A/S v. Danske Dag- Ihalainen J (2018) Computer creativity: artificial intelligence and copy-
blades Forening C-5/08 right. J Intellectual Property L Practice 13:724–728
European Court of Justice (2011a) Painer v. Standard VerlagsGmbH Kelly S (2019) A philosopher argues that an AI can’t be an artist.
C-145/10 MIT Tech Rev, 21 Februar 2019. https​://www.techn​ology​revie​
European Court of Justice (2011b) Football Association Premier w.com/s/61291​3/a-philo​sophe​r-argue​s-that-an-ai-can-never​-be-
League Ltd and Others v QC Leisure C-403/08. an-artis​t/. Accessed 14 Jan 2020
European Court of Justice (2012) Football Dataco Ltd v. Yahoo! UK Knight W (2017) The dark secret at the heart of AI, MIT Technol-
Ltd C-604/10 ogy Review, 11 April 2017, https ​ : //www.techn ​ o logy ​ r evie​
European Court of Justice (2013) Deckmyn v. Vandersteen C-201/13 w.com/s/60408​7/the-dark-secre​t-at-the-heart​-of-ai/. Accessed 14
European Parliament and the Council (2001) Directive 2001/29/EC Jan 2020
of the European Parliament and of the Council of 22 May 2001 Koboldt C (1995) Intellectual property and optimal copyright protec-
on the harmonisation of certain aspects of copyright and related tion. J Cultural Econ 19:131–155
rights in the information society. O.J. L 2001/167, 10–19 Komuves D, Niebla Zatarain J, Schafer B, Diver L (2015): Monkeying
European Parliament and the Council (2009) Directive 2009/24/EC Around with Copyright—Animals, AIs and Authorship in Law,
of the European Parliament and of the Council of 23 April 2009 CREATe Working Paper 2015/02. https​://www.creat​e.ac.uk/publi​
on the legal protection of computer programs. O.J. L 2009/111, cation​ s/monkey​ ing-around​ -with-copyri​ ght-animal​ s-ais-and-autho​
16–22 rship​-in-law/. Accessed14 Jan 2020
European Parliament and the Council (2012) Directive 2012/28/EU of Kuner C, Cate FH, Lynskey O, Millard C, Ni Loideain N, Svantesson
the European Parliament and of the Council of 25 October 2012 DJB (2018) Expanding the artificial intelligence-data protection
on certain permitted uses of orphan works. O.J. L 2012/299, 5–12 debate. Int Data Privacy L 8:289–292
European Parliament and the Council (2019) Directive (EU) 2019/790 Lambert P (2017) Computer-generated works and copyright: selfies,
of the European Parliament and of the Council of 17 April 2019 traps, robots, AI and machine learning. EIPR 39:12–20
on copyright and related rights in the Digital Single Market and Lunney GS (2014) Copyright’s Mercantilist Turn. Fla St U L Rev
amending Directives 96/9/EC and 2001/29/EC. O.J. L 2019/790, 42:95–150
92–125 McCormack J, Gifford T, Hutchings P (2019) Autonomy, authenticity,
European Parliament and Council of the European Union (1996) Direc- authorship and intention in computer generated art. In: Ekárt A,
tive 96/9/EC of the European Parliament and of the Council of 11 Liapis A, Luz Castro Pena M (eds) Computational intelligence in
March 1996 on the legal protection of databases. O.J. L 2001/167, music, sound, art and design. Springer, Berlin, pp 35–50
10–19 McCutheon J (2013) The vanishing author in computer-generated
European Parliament and Council of the European Union (2006) Direc- works: a critical analysis of recent australian case law. Melb U
tive 2006/116/EC of the European Parliament and of the Council L Rev 36:915–969
of 12 December 2006 on the term of protection of copyright and Miernicki M and Ng I (2019) Machines, attribution and integrity: arti-
certain related rights. O.J. L 2006/372, 12–18 ficial intelligence and moral rights. Jusletter IT 21. February 2019.
Galajdová D (2018) Deadlock in protection of software developed by https:​ //juslet​ ter-it.weblaw
​ .ch/en/issues​ /2019/IRIS/machin​ es_-attri​
AI. In: Schweighofer E, Kummer F, Saarenpää A, Schafer B (eds) butio​_a400b​1d060​.html__ONCE&login​=false​.
Data protection/legal Tech. Editions Weblaw, Bern, pp 601–608 Niebla Zatarain JM (2018) A similarity assessment in copyright works:
Ginsburg JC (2004) The right to claim authorship in U.S. copyright the insertion of intelligent technology to provide certainty to
and trademark law. Hous L Rev 41:263–308 rights holders and the public sector. Eur J L Tech 9:1–28
Ginsburg JC (2018) People not machines: authorship and what it means Ory S, Sorge C (2019) Schöpfung durch Künstliche Intelligenz? NJW
in the Berne convention. IIC 49:131–135 72:710–713
Glasser D (2001) Copyrights in computer-generated works: whom, if
anyone, do we reward? Duke L Tech Rev 1:24

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

AI & SOCIETY (2021) 36:319–329 329

Paulius Č, Jurgita G, Gintarė S (2017) IT is possible to grant legal U.S. Copyright Office (2019) Authors, attribution, and integrity: exam-
personality to artificial intelligence software systems? Comp L ining moral rights in the United States. https​://www.copyr​ight.
Secur Rev 33:685–699 gov/polic​y/moral​right​s/full-repor​t.pdf. Accessed 14 Jan 2020
Perry M, Margoni T (2010) From music tracks to Google maps: U.S. Court of Appeals 2nd Circuit (1995) Carter v. Helmsley-Spear
who owns computer-genereated works? Comp L Secur Rev Inc., 71 F.3d 77.
26:621–629 U.S. Court of Appeals 1st Circuit (2010) Mass. Museum of Contem-
Pettenati LA (2000) Moral rights of artists in an international market- porary Art Found., Inc. v. Buchel 593 F.3d 38
place. Pace Int L Rev 12:425–450 U.S. District Court Northern District of California (2016) Naruto v.
Ricketson S (1991) People or machines. The Berne convention and the Slater Case No. 15-cv-04324-WHO, U.S. Dist. LEXIS 11041
changing concept of authorship. Colum- VLA J L- Arts 16:21–22 U.S. Supreme Court (1991) Feist Publ’n, Inc. v. Rural Telephone Ser-
Ricketson S, Ginsburg J (2006) International copyright and neighbour- vice Comp., Inc. 499 U.S. 340
ing rights, 2nd edn, vol I. Oxford Univ. Press, Oxford Vertinsky L, Rice TM (2002) Thinking about thinking machines:
Rigamonti CP (2006) Deconstructing moral rights. Harv Int L J implications of machine inventors for patent law. BUJ Sci Tech
47:353–412 L 8:574–613
Rigamonti CP (2007) The conceptual transformation of moral rights. Victor D (2016): Microsoft Created a Twitter Bot to Learn From Users.
Am J Comp L 55:67–122 It Quickly Became a Racist Jerk., The New York Times, March
Rosati E (2015) Just a laughing matter? Why the decision in Deckmyn 24 2017. https:​ //www.nytime​ s.com/2016/03/25/techno​ logy/​ micro​
is broader than parody. Common Market L Rev 52:511–529 soft-creat​ed-a-twitt​er-bot-to-learn​-from-users​-it-quick​ly-becam​
Rosenthal Kwall R (2010) The soul of creativity. Stanford University e-a-racis​t-jerk.html. Accessed 15 Jan 2020
Press, Stanford Von Lewinski S, Walter M (2010) Rights of authors. In: Von Lewinski
Rushton M (1998) The moral rights of artists: Droit Moral ou Droit S, Walter M (eds) European copyright law. Oxford University
Pécuniaire? J Cultural Econ 22:15–32 Press, Oxford
Samuelson P (1986) Allocating ownership rights in computer-gener- Walter MM (2008) Österreichisches Urheberrecht, vol I. Medien und
ated works. U Pittsburgh L Rev 47:1185–1228 Recht, Vienna
Schafer B, Komuves D, Niebla Zatarain JM, Diver L (2015) A fourth WIPO (1979), Berne convention for the protection of literary and artis-
law of robotics? Copyright and the law and ethics of machine co- tic works, September 9, 1886, revised at Paris on July 24, 1971
production. Artif Intell Law 23:217–240 and amended in 1979, S. Treaty Doc. No. 99–27, (1986). https​://
Schére E (2018) Where it the Morality? Moral rights in international wipol​ex.wipo.int/en/text/28369​8. Accessed 14 Jan 2020
intellectual property and trade Law. Fordham Int L J 41:773–784 WIPO (1991) Committee of experts on a possible protocol to the Berne
Schmitt S, Hudson JJ et al. (2018) Kickstarting deep reinforcement convention for the protection of literary and artistic works, ques-
learning. https​://arxiv​.org/pdf/1803.03835​.pdf. Accessed 14 Jan tions concerning a possible protocol to the Berne convention part
2020. I, Doc. No. BCP/CE/I/2.
Schönberger D (2018) Deep copyright: up- and downstream questions WTO (1994) Agreement on trade-related aspects of intellectual prop-
related to artificial intelligence (AI) and machine learning (ML). erty rights, Apr. 15, 1994, marrakesh agreement establishing the
Intellectual Property J 24:35–58 world trade organization, Annex 1C, 1869 U.N.T.S. 299, 33 I.L.M.
Sirvinskaite I (2010) Toward copyright “Europeanification”: European 1197.
Union moral rights. J Int’l Media Ent L 3:263–288 Wu AJ (1997) From video games to artificial intelligence: assigning
Solaiman SM (2017) Legal personality of robots, corporations, copyright ownership to works generated by increasingly sophis-
idols and chimpanzees: a quest for legitimacy. Artif Intell Law ticated computer programs. AIPLA QJ 25:131
25:155–179 Yanisky-Ravid S (2017) Generating rembrandt: artificial intelligence,
Stewart D (2014) Do androids dream of electric free speech? Visions of copyright, and accountability in the 3A Era—the human-like
the future of copyright, privacy and the first amendment in science authors are already here—a new model. Mich St L Rev 659–726
fiction. Comm L Pol’y 19:433–461 Yanisky-Ravid S, Velez- Hernandez LA (2018) Copyrightability of art-
Surden H (2014) Machine learning and law. Washington L Rev works produced by creative robots and originality: the formality-
89:87–115 objective model. Minn J L Sci Tech 19:1–53
Tang X (2012) The artist as brand: toward a trademark conception of Yu R (2017) The machine author: What level of copyright protection
moral rights. Yale L J 122:218–257 is appropriate for fully independent computer-generated works?
Teubner G (2018) Digitale Rechtssubjekte? Zum privatrechtlichen Sta- U Pa L Rev 165:1245–1270
tus autonomer Softwareagenten. AcP 218:155–205
U.K. House of Lords (1988) Lords Sitting of 25 February 1988 HL Publisher’s Note Springer Nature remains neutral with regard to
Deb vol 493 col 1305. https​://api.parli​ament​.uk/histo​ric-hansar​ d/ jurisdictional claims in published maps and institutional affiliations.
lords​/1988/feb/25/copyr ​ight-desig​ns-and-paten​ts-bill-hl#colum​
n_1305. Accessed 14 Jan 2020.
United Kingdom (1988) Copyright, designs and patents Act 1988,
c. 48. https​://www.legis​latio​n.gov.uk/ukpga​/1988/48/conte​nts.
Accessed 14 Jan 2020
U.S. Copyright Office (2017) Compendium of US. Copyright Office
Practices, 3rd edn. https:​ //www.copyri​ ght.gov/comp3/​ docs/compe​
ndium​.pdf. Accessed 14 Jan 2020

13

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

[email protected]