ArticlePDF Available

Confucius, cyberpunk and Mr. Science: comparing AI ethics principles between China and the EU



We propose a comparative analysis of the AI ethical guidelines endorsed by China (from the Chinese National New Generation Artificial Intelligence Governance Professional Committee) and by the EU (from the European High-level Expert Group on AI). We show that behind an apparent likeness in the concepts mobilized, the two documents largely differ in their normative approaches, which we explain by distinct ambitions resulting from different philosophical traditions, cultural heritages and historical contexts. In highlighting such differences, we show that it is erroneous to believe that a similarity in concepts necessarily translates into a similarity in ethics as even the same words may have different meanings from a country to another-as exemplified by that of "privacy". It would, therefore, be erroneous to believe that the world would have adopted a common set of ethical principles in only three years. China and the EU, however, share a common scientific method, inherited in the former from the "Chinese Enlightenment", which could contribute to better collaboration and understanding in the building of technical standards for the implementation of such ethics principles.
1 3
AI and Ethics
Confucius, cyberpunk andMr. Science: comparing AI ethics principles
betweenChina andtheEU
PascaleFung1· HubertEtienne2
Received: 29 March 2022 / Accepted: 21 May 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022
We propose a comparative analysis of the AI ethical guidelines endorsed by China (from the Chinese National New Gen-
eration Artificial Intelligence Governance Professional Committee) and by the EU (from the European High-level Expert
Group on AI). We show that behind an apparent likeness in the concepts mobilized, the two documents largely differ in their
normative approaches, which we explain by distinct ambitions resulting from different philosophical traditions, cultural
heritages and historical contexts. In highlighting such differences, we show that it is erroneous to believe that a similarity
in concepts necessarily translates into a similarity in ethics as even the same words may have different meanings from a
country to another—as exemplified by that of “privacy”. It would, therefore, be erroneous to believe that the world would
have adopted a common set of ethical principles in only three years. China and the EU, however, share a common scientific
method, inherited in the former from the “Chinese Enlightenment”, which could contribute to better collaboration and
understanding in the building of technical standards for the implementation of such ethics principles.
Keywords AI ethics· Europe· China· Cyberpunk· Regulation
1 Introduction
The exponential development of artificial intelligence trig-
gered an unprecedented global concern for potential social
and ethical issues. Stakeholders from different industries,
international foundations, governmental organizations, and
standards institutions quickly reacted in improvising codes
of ethics for the purpose of establishing afirst layer of con-
trol in the absence of existing State laws. This exercise is
comparable to that of controlling the proliferation of nuclear
weapons in the 1960’s. Its objective was no less than to reach
a global agreement on common ethical standards to regulate
one of the most promising technologies whose crucial stra-
tegic implications on both business and political grounds
are already well acknowledged. The resulting profusion of
documents on AI ethical standards, as much as 84 identified
by Jobin etal. [14] and 160 in Algorithm Watch’s AI Eth-
ics guidelines Global Inventory [1], however, deserve to be
A major concern is the large homogeneity and presumed
consensualism around these principles. Jobin etal. [14]
identified 11 clusters of ethical principles among 84 docu-
ments and Fjeld etal. [9] found 8 key themes across 36
of the most influential of these. They both noted a general
convergence, which leads Fjeld etal. to conclude that the
conversation around principled AI is beginning to converge”
and that “these themes may represent the ‘normative core’
of a principle-based approach to AI ethics and governance”.
However, we argue that ethics, by nature, is not consensual.
While it is true that some ethical doctrines, such as Kantian
deontologism, aspire to universalism, they are however not
universal in practice. In fact, ethical pluralism is more about
differences in which relevant questions to ask rather than dif-
ferent answers to a common question. When people abide by
different moral doctrines, they tend to disagree on the very
approach to an issue. Therefore, even when people from dif-
ferent cultures agree on a set of common principles, it does
not necessarily mean that they share the same understanding
of these concepts and what they entail.
* Hubert Etienne
1 Centre forArtificial Intelligence Research (CAiRE),
The Hong Kong University ofScience andTechnology,
HongKong, China
2 Department ofPhilosophy, Ecole Normale Supérieure, Paris,
AI and Ethics
1 3
To better understand the philosophical roots and cultural
context underlying ethical principles in AI, we propose to
analyze and compare the ethical principles endorsed by the
Chinese National New Generation Artificial Intelligence
Governance Professional Committee (CNNGAIGPC) and
those elaborated by the European High-level Expert Group
on AI (HLEGAI). China and the EU have very different
political systems and diverge in their cultural heritages.
In our analysis, we wish to highlight that principles which
seem similar a priori may actually have different meanings,
derived from different approaches and reflect distinct goals
2 Promotional vs prohibitive approaches
At first glance, the Chinese ethical principles seem similar
to those of the EU in many aspects. Both notably promote
fairness, robustness, privacy, safety and transparency. Their
prescriptive approaches, however, reveal different cultural
perspectives associated with different objectives.
2.1 A collective vs anindividualist cultural heritage
Confucian philosophy has shaped the governing system in
China and the rest of East Asia for centuries. It emphasizes
the “rule for the people”, rather than “rule by the people”,
and favors an elitist leadership, associating political man-
dates with competence and merit. The Chinese govern-
ment’s belief in “doing the right thing” for its citizens
is informed by the Confucian ideas of virtuous authority
and exemplary person, grounded in ren (humaneness), yi
(appropriateness), li (rite), and zhi (wisdom). This philo-
sophical tradition explains the community-focused and
goal-oriented perspective, from which the Chinese guide-
lines derive, together with the promotions of principles,
such as “harmony and friendship”, “shared responsibili-
ties”, “tolerance and sharing”, and “open collaboration”.
A high sense of social responsibility and self-discipline”
is also expected from individuals to harmoniously par-
take in a community while promoting shared responsi-
bilities and open collaboration. The emphasis is explic-
itly informed by the Confucian value of “harmony” as an
ideal balance to be achieved through the control of extreme
passions to avoid conflicts. Other than a stern admonition
against “illegal use of personal data”, such value leaves
little room for constraining rules. These principles are
not paths to regulation, what would be detrimental to the
development of research and business opportunities in a
highly competitive environment where innovation is cru-
cial. Rather, they are framed to guide AI developers in a
way that would promote collective good for the Chinese
The European ethical principles, in contrast, emerge from
a more individual-focused and rights-based approach. They
express a different aspiration, rooted in the Enlightenment
tradition, and colored by the European history. Their primary
goal is to protect individuals against well-identified harms.
Whereas the Chinese principles emphasize the promotion
of good practices, the EU focuses on the prevention of evil
consequences. The former draws a direction for the devel-
opment of AI, so that it contributes to the improvement of
society. The latter sets limitations to its uses, so that it does
not happen at the expense of certain categories of people.
This distinction is clearly illustrated by the presentation of
fairness, diversity and inclusiveness. While the EU empha-
sizes fairness and diversity with regard to individuals from
specific demographic groups (specifying gender, ethnicity,
disability, etc.), Chinese guidelines urge for the upgrade of
“all industries”, reduction of “regional disparities” and pre-
vention of data monopoly. While the EU insists on the pro-
tection of vulnerable persons and potential victims, China
prescribes “inclusive development through better education
and training, support”.
The individualist perspective reflected by the European
approach to AI ethics should, however, not be mistaken for
a form of selfish moral individualism, but rather as a result
from the European history of individual reasoning. It is
worth noting that the first claim of the Enlightenment did
not target political self-determination nor the possibility for
people to partake in collective decision-making, but rather
ontological autonomy, freeing them from the subjection
to the king and the power to the State. This was famously
defined by Kant [13] as “man’s emergence from his self-
incurred state of immaturity”. This point is also illustrated
in the Declaration of Human and Citizen Rights which gives
prevalence of a citizen’s protection against political power
abuse — i.e., negative rights over positive rights. Finally, the
repeated clashes of European nationalism that culminated in
WWII and the trauma of totalitarianism acted as a powerful
reminder to the Europeans of the dangers of political holism.
Table 1 The ethical principles endorsed by the Chinese National New
Generation Artificial Intelligence Governance Professional Commit-
tee (CNNGAIGPC) and those elaborated by the European High-level
Expert Group on AI (HLEGAI)
Chinese ethical principles
EU key requirements [10]
Harmony and friendship Societal and environmental well-being
Fairness and justice Diversity, non-discrimination and fairness
Tolerance and sharing Human agency and oversight
Respect privacy Privacy and data governance
Safe and controllable Technical Robustness and safety
Share responsibilities Transparency
Open collaboration Accountability
Agile governance
AI and Ethics
1 3
Consequently, European societies have shown a clear pref-
erence for individualist and rights-based approaches of
2.2 Promotional vs prohibitive approaches
AI governance in both the EU and China are led by the
transnational and national governments in consultation with
industry stakeholders and academic experts. It is therefore
pertinent to compare the actual and perceived roles of their
governments in setting AI ethical guidelines. Philosophers
have debated the compatibility of Confucian values with
Western liberal democracies. There have also been debates
on normative versus empirical legitimacy of a government,
where scholars study the question of why “the observed
level of regime legitimacy under non-democratic regimes
has been substantially higher than either established or
emerging democracies” [3]. Commenting on Shin’s work
[21] based on the Asian Barometer Survey, Chu [4] states
that “the majority of East Asians in other countries with a
Confucian legacy also tend to be attached to ‘paternalistic
meritocracy’, prioritize economic well-being over freedom,
and define democracy in substantive (rather than procedural)
terms.” China is more of an assumed authoritarian technoc-
racy than an anti-democracy. Its political elite, composed of
civil servants mostly with backgrounds in science, technol-
ogy, engineering, and mathematics (STEM), has adopted a
pragmatic approach to AI ethics, grounded in existing appli-
cations and driven by society’s needs.
The Chinese leadership routinely holds workshops
with scientists to keep up to date with the latest trends in
advanced technologies. This proximity between political
leaders and scientific research, together with the greater con-
trol exercised by the government on the development of this
technology, are included in the centralized planning of the
economy to serve national strategic objectives. It explains
the great pragmatism in the Chinese governmental approach
to AI ethics, focusing on foreseeable harms that may derive
from the use of AI in the near future. The Chinese gov-
ernment is not averse to regulations, but in the case of AI
governance, it is unlikely to regulate with a broad brush
before AI has been widely applied and has found to be of
serious negative impact in specific areas or posing danger
to the society. Nevertheless, in areas of immediate societal
impact and concern, such as data privacy, China has also
been able to devise strict laws and regulations. Therefore,
even though the ethical guideline call for mere “respect for
privacy”, it is understood that companies must exercise a
great self-discipline in terms of user data protection.
In contrast, the traditional training of the European
political elite does not always allow the same interest,
nor understanding, for AI and new technologies in gen-
eral. Furthermore, the experience with totalitarianism in
Europe always serves as a reminder, calling for prudence.
It also explains the greater skepticism from the European
citizens about technologies, especially those that can be
used for surveillance purposes. A recent example is the
strong general reluctance of the French population to adopt
the official COVID-tracking application out of fear of what
it could be used for by the government. This is why the
European approach to AI ethics was, since the beginning,
conceived as both a way to regulate the private sector from
foreseeable risks, and to prevent systematic distrust against
AI by the public. It is intended to provide some sort of
guarantee against potential abuses from public–private
partnerships between governments and AI companies.
The differences between Chinese and European cultural
heritage, their respective historical contexts and the back-
ground of their political elites translate into two different
types of moral imperatives. The European requirements,
centered on satisfying initial conditions, dictate a strict
abidance by deontologist rules in the pure Kantian tradi-
tion. In contrast, the Chinese principles, referring to an
ideal to aim for, express rather softer constraints at differ-
ent levels, as part of a process to improve society. For the
Europeans the development of AI “must be fair”; for the
Chinese it should “eliminate prejudices and discrimina-
tions as much as possible”. The EU requires “processes
to be transparent” while China requires to “continuously
improve transparency”. The EU principles aim to protect
European citizens from vertical and horizontal abuses,
conscient of the danger of nationalism. The Chinese gov-
ernance system, in contrast, adopts a holistic approach,
holding that the social group it forms is not to be reduced
to the sum of its parts, but produces something more,
namely the Chinese nation. Its ethical principles thus
aim to benefit Chinese citizens through the service of the
Chinese nation, considered as common good citizens are
associated with.
3 A utopian vs adystopian vision
Other than roles of the governments, the two ethical sets
of guidelines are informed by opposing views from the
European and Chinese populations regarding AI. The main
fears expressed by western society toward AI are related
to privacy and surveillance, job automation [12], and the
possibility of a loss of control resulting in existential risks
for humanity [8]. These are greatly dependent on people’s
trust in political and technology leadership, on the narra-
tives surrounding the development of AI in mainstream
media, and on the representation of AI in science fiction.
AI and Ethics
1 3
3.1 The question oftrust intheGovernment
Public opinion studies show that Chinese people are largely
supportive of AI, which they associate with a great potential
to benefit society, and as an engine of economic growth.
Strong government support, a vibrant commercial market
for AI, and media content favorable to AI all contribute to
this positive perception [6]. A comparative study of Ger-
man, Chinese and UK participants to assess the Attitude
Towards Artificial Intelligence (ATAI scale) showed that the
Chinese scored the highest on the ATAI Acceptance scale
and lowest on the ATAI Fear scale [22]. Such findings are
supported by another survey conducted by Ipsos, according
to which 70 percent of Chinese respondents stated that they
trust artificial intelligence [23]. Overall, Asian public opin-
ion tends to be more favorable to AI. For example, a Pew
Research Center survey in 2020 found that “majorities in
most Asian public—Singapore (72%), South Korea (69%),
India (67%), Taiwan (66%) and Japan (65%)—surveyed
see AI as a good thing” for society, whereas more than half
of the EU population view AI as negative. “In France, for
example, views are particularly negative with only 37% of
survey people considering AI as good for society versus 47%
of them viewing it as bad. In the US and UK, about as many
say it has been a good thing for society as a bad thing.” [15].
The European historical context somewhat led to a general
state of distrust in governments in many liberal democracies,
which is described as the “counter-democracy” by Rosanval-
lon [20]. In France [16], as in the US [19], for instance, more
than three quarters of the citizens think their political repre-
sentatives behave unethically. The fear of AI being used by
governments for mass surveillance is a major concern, and
public–private collaborations are also regarded with high
skepticism. From the private sector, multiple incidents and
scandals related to user privacy, surveillance and nudging
involving top technology companies in the past few years
severely dampened consumer enthusiasm, together with the
perception of these companies’ intentions to do good or to
operate responsibly [2].
This trust gap is particularly well illustrated by the per-
ception of “privacy”. Data privacy is promoted by both the
European and the Chinese ethical guidelines, but with dif-
ferent meanings. The European promotion of privacy, as
highlighted by General Data Protection Regulation (GDPR),
encompasses the protection of individual data from both
state and commercial entities. The Chinese privacy guide-
lines in contrast only target private companies and potential
malicious agents. Whereas personal data are strictly pro-
tected both in the EU and in China from commercial enti-
ties, the State retains full access in China. Such a practice
would be shocking in Western countries; it is however read-
ily accepted by Chinese citizens, accustomed to living in a
protected society and have consistently shown high trust in
their government [7]. It is within the social norm in China
where Chinese parents routinely have access to their chil-
dren’s personal information to provide guidance and protec-
tion. This difference goes back to the Confucian tradition of
trusting and respecting the heads of State and family. The
trust is nowadays strengthened by the great economic growth
the country’s leaders succeeded in achieving. A recent sur-
vey showed that Chinese government’s successful domestic
management of the COVID-19 crisis is likely to inspire more
trust by its citizens [25]. This suggests that the trust gap
would also be related to people’s perception of government
competency, and thus to the objectives these governments
aim to achieve with AI. The most developed European coun-
tries are former global powers, which gave up on their past
expansionist ambitions, and now focus on domestic policies
to solve their social issues, while trying not to be left behind
in the innovation race. In contrast, China has recently estab-
lished itself as a world leading economy. This rapid ascend
onto the world stage, together with the clear ambition to
challenge the US leadership, has played a significant role in
securing trust from the Chinese people in the actions of their
government, including the strategic support given to AI.
3.2 The influence ofthecyberpunk culture
The gap in the cultural representation of AI, perceived as
a force for good in China, and as a menacing force in the
dystopian technological future in the Western world, could
be rooted in the influence of popular culture. Robots are
assistants and companions in the Chinese vision of a tech-
nological future, whereas they tend to become insurrectional
machines as portrayed by a Western media heavily influ-
enced by the cyberpunk subgenre of science fiction and illus-
trated by success movies, such as 2001: A Space Odyssey
(Stanlet Kubrick, 1968), Blade Runner (Ridley Scott, 1982),
The Matrix (Lana & Lilly Wachowski, 1999) or Minority
Report (Steven Spielberg, 2002). Cyberpunk emerged in
the 1960s in the West, as a subgenre of science fiction. It
represents a view of a high-tech future where social orders
are broken down and renegade rebel forces battle against
a Big Brother government that uses technology to control
the people. This vision, embodied in the works of Philip
Dick and others, is a stark departure from a more positive
vision of a technological future espoused by Isaac Asimov
or Jules Verne in previous generations of science fiction.
The influence of popular culture in shaping public opinion
is well acknowledged. More particularly, Young and Car-
penter [24] found that “consumption of frightening armed
AI films is associated with greater opposition to autonomous
weapons”. Since lethal (fully) autonomous weapons sys-
tems have no official existence—or at least can we say that
their use is not common yet—people’s opinion about these
is greatly influenced by their representations from science
AI and Ethics
1 3
fiction literature and movies: “Sci-fi as a genre, and certain
iconic killer robot films in particular, appears most salient in
rhetorical arguments against such weapons. […] [and robo-
pocalyptic films themselves have been likelier to encourage a
cautionary rather than techno-optimistic sentiment on armed
AI, among at least sci-fi literate members of the American
public” conclude Young and Carpenter.
Chinese science fiction in the early decades of the twen-
tieth century was mainly translated from Soviet literature
and designed for children. While in Chinese literature there
is no tradition of describing a utopian future, there is not
exactly a dystopian and cyberpunk influence either. Main-
land China was mostly closed to the outside world before
the 1980s, what prevented it from the influence of the cyber-
punk culture and the dystopian visions of a technological
future such as that conveyed by George Orwell’s work. The
influence of the Soviet Union stopped in 1960s when the
diplomatic relationship between China and the USSR was
severed, preventing the Chinese population from the dark
visions of Stanisław Lem, for example. It is interesting to
note an opposite trend to that of China in a country with a
similar Buddhist/Confucian culture: Japan. The Japanese are
found to have a relatively low level of trust in their govern-
ment, in particular following the Fukushima nuclear plant
crisis in 2011. Their trust in the government ranks below
that of many EU countries including Germany [18]. Despite
its world leading pioneer position in robotics and techno-
phile population, there is a general doomsday malaise from
a collective memory of the only atomic bomb detonation in
history. The cyberpunk animation classic Akira (Katsuhiro
Otomo, 1988) foretold a post-apocalyptic dystopian future
in 2019 rife with anti-government protests and gang vio-
lence, superpowers and government sponsored assassination
attempts, all in the shadow of an impending Olympic game.
Akira inspired a cult following and had a strong influence on
Western science fiction culture that followed, including the
Matrix series. Nevertheless, whereas the Japanese have suf-
fered a number ofdata breaches prompting their government
to amend the Act on the Protection of Personal Information
(APPI) in 2020, they are still relatively optimistic about AI.
This is likely due to the familiarity of most Japanese with the
long-standing use of AI and robotics in their manufacturing
and health care sectors.
4 A scientic common ground
These gaps in both cultural representations of technology
and levels of trust toward governments constitute valuable
signals to explain why the Chinese principles work as pater-
nalistic guidelines where trust is not an objective, because
mistrust is not an issue, while the European principles estab-
lish the conditions for AI to be “trustworthy”, as distrust has
become the norm. Despite the seemingly different, though
not contradictory, approaches on AI ethics from China and
the EU, the presence of major commonalities between them
points to a more promising and collaborative future in the
implementation of these standards.
Much of operationalization and implementation of ethical
standards in AI is in organizational governance, that is to
say the process and application choices we make. In addi-
tion to governance, ethical principles need to be incorpo-
rated into the design of AI systems and a significant part of
operationalizing these standards lies in improvements and
modifications to the methodology and the architecture of
modern AI software systems. AI systems research and devel-
opment is an open and collaborative process across nations.
Their designers from China, the US or the EU are all trained
in a similar computer science and engineering curriculum
based on the “scientific method”. This latter paradigm—
which consists in formulating hypothesis and devising
empirical experiments to verify these to arrive at a claim
or thesis—has underpinned research areas from statistics,
signal processing, optimization, machine learning, and pat-
tern recognition, all forming the multidisciplinary area that
is modern AI today. The scientific method was first adopted
by China among other Enlightenment values during the May
Fourth Movement in 1919. Coined the “Chinese Enlighten-
ment”, this movement resulted in the first repudiation of
traditional Confucian values, and it was then believed that
only by adopting Western ideas of “Mr. Science” and “Mr.
Democracy” in place of “Mr. Confucius” could the nation
be strengthened. In the years since the third generation of
Chinese leaders, the Confucian value of the “harmonious
society” is again promoted as a cultural identity of the Chi-
nese nation. Nevertheless, “Mr. Science” and “technologi-
cal development” continue to be seen as a major engine for
economic growth and livelihood improvement, hence lead-
ing to the betterment of the “harmonious society”. For both
governance and design, two leading international standards
bodies, namely the International Standards Organization
(ISO) and the Institute of Electrical and Electronic Engi-
neers (IEEE) are working on and publishing governance and
best practice guidelines for the industry. Since ISO and IEEE
standards lend credibility to products and services, they are
widely accepted and recognized by countries. Chinese as
well as European representatives are also actively involved
in these standards organizations ensuring that such standards
and best practice guidelines take into account cultural norms
and differences. It results that “ISO data security standards
have been widely adopted by cloud computing providers,
e.g., Alibaba, Amazon, Apple, Google, Microsoft, and Ten-
cent.” [5] while the working group on IEEE Guidelines for
Ethically Aligned Design keeps exploring established eth-
ics systems, including both philosophical traditions (utilitari-
anism, virtue ethics, and deontological ethics) and religious
AI and Ethics
1 3
and culture-based ethical systems (Buddhism, Confucian-
ism, African Ubuntu traditions, and Japanese Shinto) and
their stance on human morality in the digital age. In doing
so, […] [they] critique [ethical] assumptions [… and they]
attempt[ed] to carry these inquiries into artificial systems’
decision-making processes.” [11].
Another reason for China to acknowledge this common
scientific ground in its ethical principle of “open collabora-
tion” relates to the hundreds of thousands of Chinese stu-
dents who have gone to study in the US and the EU since the
1980’s, most of them in the STEM fields. American technol-
ogy companies, such as Microsoft, Amazon, Google, have
all established research centers in the PRC where Chinese
researchers are recruited to work with their counterparts
in the US headquarters. Chinese graduate students in AI
have one time or another worked as interns in these com-
panies in China. A sampled study from the authors of one
NeurIPS conference showed that nearly 30% of the authors
received their undergraduate degrees in China, more than
from any other country, meanwhile over 50% get their grad-
uate degrees from the US and 16% from the EU. A sig-
nificant number of Chinese AI researchers do not return to
China within five years of completing their graduate studies
abroad. In recent years, top Chinese AI companies, such
as Tencent, Baidu, Huawei, and latecomers, such as Didi
and Bytedance, also have established research labs in the
US and EU to attract AI talents. Such a collaborative melt-
ing between AI researchers from various parts of the world
should encourage discussion around AI ethics to help us
achieve more consensus around ethical principles.
5 Conclusion
We analyzed and compared AI ethical guidelines from China
and the EU both from the perspective of governmental roles,
of public opinion and popular cultures, as well as from the
scientific common ground for the research and development
of AI in China and the West. Whereas the EU framework is
rooted in the core Enlightenment values of individual free-
dom, equal rights and serves to protect against State abuses,
the Chinese guidelines are based on the Confucian values
of virtuous government, harmonious society, and targets
to protect against commercial exploitation. The EU ethical
framework is also built as a dialectic system between users
on one side, and AI developers and service providers on the
other side. These normative rules are perceived as necessary
to enable trust from users, as well as that of positive interac-
tions between these two poles. This system is dynamic and
includes effective feedback loops, allowing people to keep
control and improve the system via their ability to “con-
test and seek effective redress against decisions made by AI
systems and by the humans operating them” [10]. In other
words, the transparency and explicability of AI systems are
required for decisions to “be duly contested”. The EU prin-
ciples assume skepticism from users and attempts to assuage
such negative sentiment with protective rules. Although the
Chinese AI ethical principles seem similar to those of the
EU in many ways, they however largely differ in the overall
approach. The Chinese principles start with the assump-
tion that Chinese citizens trust the state to guide and protect
them against commercial and third-party abuses. They are
pointing a future direction for the development of AI, rather
than its limitations. Finally, the EU principles mostly refer to
deontologist normative rules—mainly negative obligations–,
whereas the Chinese principles, stemming from Confucian
values, tend to combine some strict deontologist normative
rules (e.g., prohibiting evil uses and illegal activities) with
softer constraints that could be satisfied on different levels
(e.g., promote shared and inclusive development) and even
some aspects of virtue ethics, referring to “vigilance” and
“self-discipline. Whereas the Chinese principles tend to sug-
gest directions to shape how AI should be developed and
applied, the EU principles aim to precisely define what it
should not be allowed to do.
1. Algorithm Watch. AI ethics guidelines global inventory. https://
inven tory. algor ithmw atch. org/ (2020). Accessed 22 June 2021
2. Buchholz, K.: Americans fear the AI apocalypse. Statista Info-
graphics. https:// www. s tati sta. com/ c hart/ 16623/ attit udes- of- ameri
cans- towar ds- ai/ (2019)
3. Chu, Y.-H.: Sources of regime legitimacy in Confucian societies.
J. Chin. Gov,1(2), 195–213 (2016)
4. Chu, Y.-H.: Sources of regime legitimacy and the debate over the
Chinese model. China Rev. 13, 1–42 (2013)
5. Cihon, P.: Standards for AI governance: international standards to
enable global coordination in AI research & development. Future
of Humanity Institute’s technical report (2019)
6. Cui, Di., Wu, F.: The influence of media use on public perceptions
of artificial intelligence in China: evidence from an online survey.
Inf. Dev. 37(1), 45–57 (2019)
7. Cunningham, E., Saich, T., Turiel, J.: Understanding CCP resil-
ience: surveying Chinese public opinion through time. Ash Center
for Democratic Governance and Innovation. https:// ash. harva rd.
edu/ publi catio ns/ under stand ing- ccp- resil ience- surve ying- chine
se- public- opini on- throu gh- time (2020)
8. Gherheş, V.: Why are we afraid of artificial intelligence (AI)? Eur.
Rev. Appl. Sociol. 11(17), 6–15 (2018)
9. Fjeld, J., Achten, N., Hilligoss, H., Nagy, A>, Srikumar, M.:
Principled artificial intelligence: mapping consensus in ethical
and rights-based approaches to principles for AI. Berkman Klein
Center Research Publication n° 2020–1, (2020)
10. High Level Expert Group on Artificial Intelligence. Ethics guide-
lines for trustworthy AI, European Commission. https:// ec. europa.
eu/ futur ium/ en/ ai- allia nce- consu ltati on.1. html (2021). Accessed
22 June 2021
AI and Ethics
1 3
11. IEEE Standards Association. IEEE ethics in action in autono-
mous and intelligent systems, IEEE. https:// ethic sinac tion. ieee.
org/ (2021)
12. Johnson, C., Tyson, A.: People globally offer mixed views of the
impact of artificial intelligence, job automation on society. Pew
Research Center (2020)
13. Kant, I.: Beantwortung der Frage: Was ist Aufklärung?. English
trad. from http:// donel an. facul ty. writi ng. ucsb. edu/ enlig ht. html
14. Jobin, A., Ienca, M., Vayena, E.: The global landscape of AI ethics
guidelines. Nat. Mach. Intell. 1, 389–399 (2019)
15. Johnson, C., Tyson, A,: People globally offer mixed views of the
impact of artificial intelligence, job automation on society. Pew
Research Center (2020)
16. Lévy, J.-D., Bartoli, P.-H., Hauser, M.: Les perceptions de la cor-
ruption en France. Harris interactive (2019)
17. National Governance Committee for the New Generation Artificial
Intelligence. Governance principles for the new generation arti-
ficial intelligence--developing responsible artificial intelligence,
China Daily. http:// www. china daily. com. cn/a/ 201906/ 17/ WS5d0
7486b a3103 dbf14 328ab7. html (2021). Retrieved 22 June 2021
18. OECD. “Trust in government”, The OECD data. https:// data. oecd.
org/ gga/ trust- in- gover nment. htm (2021)
19. Pew Research Center. Why Americans don’t fully trust many who
hold positions of power and responsibility (2019)
20. Rosanvallon, P.: La Contre-Démocratie. La politique à l’âge de la
défiance, Paris, Seuil
21. Shin, D.C.: Confucian Legacies and the Making of Democratic
Citizens: Civic Engagement and Democratic Commitment in Six
East Asian Countries. Cambridge University Press, Cambridge
22. Sindermann, C., Sha, P., Zhou, M., Wernicke, J., Schmitt, H.S.,
Li, M., Sariyska, R., Stavrou, M., Becker, B., Montag, C.: Assess-
ing the attitude towards artificial intelligence: introduction of a
short measure in German, Chinese, and English language. KI-
Künstliche Intelligenz 35(1), 109–118 (2021)
23. Slotta, D.: Share of people trusting AI in China 2018. Statistica.
https:// www. stati sta. com/ stati stics/ 947661/ c hina- share- of- people-
trust ing- artifi cial- intel ligen ce/ (2018)
24. Young, K.L., Carpenter, C.: Does science fiction affect political
fact? Yes and no: a survey experiment on ‘Killer Robots.’ Int.
Stud. Quart. 62(3), 562–576 (2018)
25. Wu, C.: How Chinese citizens view their government’s corona-
virus response. The Conversation. https:// theco nvers ation. com/
how- chine se- citiz ens- view- their- gover nments- coron avirus- respo
nse- 139176 (2020)
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Conference Paper
Full-text available
Our research aims to contribute to the concept of responsible artificial intelligence (AI), a topic under significant discussion in EU politics, further emphasized by recent EU publications. Primarily, AI, while beneficial, can be a potential weapon, necessitating responsible use and prevention against misuse or misalignment. In recognizing the critical role of AI research in aiding legislators and machine learning practitioners, our work aims to help prepare for future AI advancements. To the best of our knowledge, we establish the first unified definition of responsible AI. As part of a structured literature review, we clarify the current state of the art in the context of responsible AI. Based on the knowledge of the analysis part we also have discussed an approach for developing a future framework for responsible AI. The results demonstrate that responsible AI should be a human-centered approach, encompassing ethical considerations, explainability of models, privacy, security, and trust.
Full-text available
In the context of (digital) human–machine interaction, people are increasingly dealing with artificial intelligence in everyday life. Through this, we observe humans who embrace technological advances with a positive attitude. Others, however, are particularly sceptical and claim to foresee substantial problems arising from such uses of technology. The aim of the present study was to introduce a short measure to assess the Attitude Towards Artificial Intelligence (ATAI scale) in the German, Chinese, and English languages. Participants from Germany (N = 461; 345 females), China (N = 413; 145 females), and the UK (N = 84; 65 females) completed the ATAI scale, for which the factorial structure was tested and compared between the samples. Participants from Germany and China were additionally asked about their willingness to interact with/use self-driving cars, Siri, Alexa, the social robot Pepper, and the humanoid robot Erica, which are representatives of popular artificial intelligence products. The results showed that the five-item ATAI scale comprises two negatively associated factors assessing (1) acceptance and (2) fear of artificial intelligence. The factor structure was found to be similar across the German, Chinese, and UK samples. Additionally, the ATAI scale was validated, as the items on the willingness to use specific artificial intelligence products were positively associated with the ATAI Acceptance scale and negatively with the ATAI Fear scale, in both the German and Chinese samples. In conclusion we introduce a short, reliable, and valid measure on the attitude towards artificial intelligence in German, Chinese, and English language.
Full-text available
In the past five years, private companies, research institutions and public sector organizations have issued principles and guidelines for ethical artificial intelligence (AI). However, despite an apparent agreement that AI should be ‘ethical’, there is debate about both what constitutes ‘ethical AI’ and which ethical requirements, technical standards and best practices are needed for its realization. To investigate whether a global agreement on these questions is emerging, we mapped and analysed the current corpus of principles and guidelines on ethical AI. Our results reveal a global convergence emerging around five ethical principles (transparency, justice and fairness, non-maleficence, responsibility and privacy), with substantive divergence in relation to how these principles are interpreted, why they are deemed important, what issue, domain or actors they pertain to, and how they should be implemented. Our findings highlight the importance of integrating guideline-development efforts with substantive ethical analysis and adequate implementation strategies.
Full-text available
The study presents the results regarding the attitudes of students from humanities and technical specializations in Timișoara towards the emergence and development of artificial intelligence (AI). The emphasis was on the most likely consequences of the development of artificial intelligence in the future, especially the negative consequences that its development would entail. The method used for data collection was the sociological survey and the information gathering tool was the questionnaire. It was applied to a total of 929 people, ensuring a sample representativity margin of ± 3%. The analysis reveals that the participants in the study predict that due to the emergence and development of AI, in the future, interpersonal relationships will be negatively affected, there will be fewer jobs, economic crises will emerge, it will be used to make intelligent weapons, to increase military conflicts, to take control of humanity and, last but not least, to destroy mankind. The results revealed differences in responses depending on the type of specialization (humanities or technical) and the gender of the respondents.
With support from government and business, artificial intelligence is growing quickly in China. However, little is known of how media use shapes the Chinese public’s perception of artificial intelligence. Based on a national online survey (N = 738), this pilot study explored the linkages between media use and people’s risk perception, benefit perception, and policy support of artificial intelligence. Results showed that respondents perceive artificial intelligence as more beneficial than risky. Newspaper use was negatively associated with benefit perception and policy support, whereas television and WeChat use positively predicted both. Analyses of interaction effects showed that personal relevance could partly mitigate the influence of media use.
Some scholars suggest popular culture shapes public attitudes about foreign policy in ways that can affect real-world political outcomes, but relatively few studies test this proposition. We examine whether—and more importantly how—popular culture affects public opinion on foreign policy through a survey experiment on American attitudes toward fully autonomous weapons. We queried respondents about their consumption of popular culture—including a number of iconic science-fiction films featuring armed artificial intelligence (AI)—before or after questions about autonomous weapons. We find that science fiction “priming” exerts no independent effect on political attitudes, nor does referring to autonomous weapons as “killer robots.” However, consumption of frightening armed AI films is associated with greater opposition to autonomous weapons. This “sci-fi literacy” effect increases for the highest consumers of science fiction if they are “primed” about popular culture before reporting their attitudes—what we call the “sci-fi geek effect.” Our project advances current understanding of how popular culture affects public opinion on foreign policy and suggests avenues for further inquiry.
In this article, we examine the on-going debate over the legitimacy of the East Asia regime through a rigorous analysis of a recent Asian Barometer Survey data from 13 countries and territories. First, we compare the relative explanatory power of the elements ordained by Confucian political theory with that of the elements emphasized by the Western liberal tradition in explaining differences in level of regime support across different types of regimes. Next, we compare Confucian societies with non-Confucian societies to see if the Confucian propositions about regime characteristics and performances abode even better in societies that inherited Confucian cultural legacy. Third, we examine how traditional political–cultural predispositions interact with perceived regime characteristics and performances to affect respondents’ support for their country’s current regime. We found that the social and historical relevance of Confucian political theory is not limited to Confucian societies. Much like Communitarian theory, Confucian political theory presents a credible challenge (as well as an alternative) to Western liberal tradition in a much wider socio-cultural space beyond Confucian societies.
This article reexamines the ongoing debate over the legitimacy of the Chinese Communist Party regime through a rigorous analysis of a recent Asian Barometer Survey data from mainland China. It evaluates the empirical evidences related to three key issues—to what extent the CCP regime is facing a legitimacy crisis, what factors help sustain the regime’s political support, and whether the regime can defy the gravitation toward liberal democracy in the process of rapid socioeconomic modernization. There is no strong evidence suggesting that the regime’s popular foundation is highly or exclusively dependent on its superior economic performance or its manipulation of nationalist zeal. Our analysis lends its support to the culturalist argument about the prevailing influence of the traditional concepts of political legitimacy as well as the institutionalist argument about the importance of perceived characteristics of the political system.
AI ethics guidelines global inventory
  • Algorithm Watch
Algorithm Watch. AI ethics guidelines global inventory. https:// inven tory. algor ithmw atch. org/ (2020). Accessed 22 June 2021