Subverting Characters Stereotypes: Exploring the Role of AI in Stereotype Subversion

International Journal of Artificial Intelligence and Applications (IJAIA), Vol.14, No.5, September 2023
DOI: 10.5121/ijaia.2023.14502 13
SUBVERTING CHARACTERS STEREOTYPES:
EXPLORING THE ROLE OF AI IN STEREOTYPE
SUBVERSION
Xiaohan Feng 1
and Makoto Murakami 2
1
Graduate School of Information Sciences and Arts, Toyo University, Kawagoe,
Saitama, Japan
2
Dept. of Information Sciences and Arts, Toyo University, Kawagoe, Saitama, Japan
ABSTRACT
The Aim of this paper is to explore different ways of using AI to subvert stereotypes more efficiently and
effectively. It will also enumerate the advantages and disadvantages of each approach, helping creators
select the most appropriate method for their specific situations.
AI opens up new possibilities, enabling anyone to effortlessly generate visually stunning images without the
need for artistic skills. However, it also leads to the creation of more stereotypes when using large amounts
of data. Consequently, stereotypes are becoming more prevalent and serious than ever before. Our belief is
that we can use this situation in reverse, aiming to summarize stereotypes with AI and then subvert them
through elemental exchange.
In this study, we have attempted to develop a less time-consuming method to challenge character
stereotypes while embracing the concept of "exchange." We selected two character archetypes, namely the
"tyrant" and the "mad scientist," and summarized their stereotypes by generating AI images or asking
ChatGPT questions. Additionally, we conducted a survey of real historical tyrants to gain insights into
their behavior and characteristics. This step helped us comprehend the reasons behind stereotyping in
artwork depicting tyrants. Based on this understanding, we made choices about which stereotypes to
retain. The intention was to empower the audience to better evaluate the identity of the character. Finally,
the two remaining character stereotypes were exchanged, and the design was completed.
This paper documents the last and most time-consuming method. By examining a large number of sources
and examining what stereotypical influences were used, we were able to achieve a greater effect of
subverting stereotypes. The other method is much less time-consuming but somewhat more random.
Whether one chooses by subjective experience or by the most frequent choices, there is no guarantee of the
best outcome. In other words, it is the one that best guarantees that the audience will be able to quickly
identify the original character and at the same time move the two characters the furthest away from the
original stereotypical image of the original. In conclusion, if the designer has sufficient time, ai portrait +
research or chatGPT + research can be chosen. If there is not enough time, the remaining methods can be
chosen. The remaining methods take less time and the designer can try them all to get the desired result.
KEYWORDS
Multidisciplinary, Artificial Intelligence, Arts & Design History, Stereotypes
1. INTRODUCTION
Stereotypes are prevalent in every culture and media, especially in popular culture [1]. For
instance, in video games, female characters are often cast in the role of "Damsel in distress" to

14
fulfill game objectives and advance the storyline [2]. Similarly, print media frequently depicts
Latina women with a sexy yet uninhibited image [3]. These stereotypes are also present in the
film industry, with approximately 85% of the global box office revenue coming from Hollywood
films, influencing gender, race, and cultural stereotypes established by the U.S. film industry that
spread widely to other media [4].
Artificial Intelligence (AI) is a rapidly advancing field, making significant strides and evolution
in recent years. This technology has transformed various industries and impacted our lives. For
example, numerous AI image generators have become widely used in the creative and
technological sectors, allowing anyone to easily create visually appealing images without
requiring artistic skills. However, generating images from massive datasets also contributes to the
proliferation of stereotypes, making them more prevalent and serious than ever before. In this
regard, the study aims to utilize AI to group and exchange elements of stereotypes to overturn
prevailing stereotypes.
In past studies, we have used AI to compile stereotypes of pop culture "witches." We summarize
the stereotypical image of the Witch today. We then analyze the two major components of the
stereotype, "appendages" and "appearance," from a historical and social perspective, and attribute
the reasons for the formation and transformation of the Witch image. These past stereotypes will
be designed using the design approach of extraction, preservation, and transformation, and
finally, the pros and cons of this method will be summarized from a practical perspective.
The research results prove that this method is feasible. At the same time, however, the method
has several drawbacks. The most obvious drawback is the need to spend a lot of time on a
literature review.
Also, in the various methods of stereotype subversion, we have studied, whether replacing
stereotypes with image-based media with information obtained from textual media or replacing
current stereotypes with past stereotypes of characters, the exchange of stereotypes is essential.
Such an exchange of elements is not only effective in subverting stereotypes but is also used for
innovation.
Building on the concept of "exchange," the study proposes a method to overturn character
stereotypes more efficiently with less time investment. By selecting two character archetypes,
"tyrant" and "mad scientist," the stereotypes associated with them are grouped using AI image
generation or through questions posed to ChatGPT. The study then delves into the investigation
of historical real-life tyrants to deepen understanding of their actions and traits, which helps
comprehend the reasons behind the stereotypes depicted in the artwork. Based on the knowledge
about the reasons behind the character's stereotyping, the study chooses which stereotypes to
retain, enabling the audience to better judge the character's identity. Finally, the remaining
stereotypes of the two characters are exchanged, and the design is finalized.
This time-efficient method allows for overturning stereotypes of two different characters
simultaneously. Additionally, the approach offers flexibility, as the step involving historical
investigation can be omitted when time constraints are pressing. However, some historical
investigation may still be necessary to more accurately select the stereotypes to be retained.
The primary objective of this research is to explore various AI-based methods to efficiently
overturn stereotypes within a shorter time frame. It aims to enumerate the strengths and
weaknesses of these methods to empower creators to select the most suitable approach for their
specific situations. Moreover, the study provides a reference and inspiration for future research
related to stereotypes.

15
This paper first lists all methods then describes the actual process in detail using the most time-
consuming method as an example, and finally lists the results obtained from all methods and
summarizes the pros and cons of each method.
2. METHODS
The methodology of this study is divided into three major steps.
Step 1: Select two character archetypes.
We decided to choose characters with significant differences. We observed that even among
randomly generated character pairs, they were already distinct archetypes. The selected
archetypes are the "explorer" and the "jester," the "trickster" and the "tyrant," and the "hero" and
the "sage." For this study, we chose the "tyrant" and the "mad scientist."
Step 2: Next, choose AI image generation or chatGPT.
When inquiring about the differences between the "Tyrant" and the "Mad Scientist," chatGPT
provided information about the relative distinctions between these two character archetypes,
including their overall image, costumes, and facial expressions. This step reduced the time
needed for subsequent exchanges regarding stereotypes. However, some details, such as the
slimness of the mad scientist's face and the tyrant's lack of hair, were not mentioned, as they were
only compiled after observing the images.
Step 3:
For AI image generation:
Give preference to stereotypes that can clearly indicate identity. If none are available, select the
stereotypes that appear most frequently in the image.
Examine detailed information.
For chatGPT:
Give preference to stereotypes that can clearly indicate identity. If none are available, make a
subjective selection.
Examine detailed information.
This paper mainly documents the most time-consuming methods. Firstly, we summarize the
character stereotypes from the images generated by AI. Next, we analyze the data to determine
which stereotype influences can achieve a greater subversion of stereotypes. Finally, we select
the reserved stereotypes.
3. SELECTING RETAINED STEREOTYPES
3.1. Tyrant
The word "tyrant" has a very long history, dating back to ancient Greece. In the early days, the
word was considered more neutral. However, some, such as the Greek philosopher Plato, took the

16
view that "tyrant" was a negative term [5]. Aristotle also defined a tyrant as a monarch who
commits extreme atrocities against people and others [6].
During the Enlightenment, the English philosopher John Locke argued that tyrants act beyond
their own power. Tyranny is not for the good of the people but for the personal benefit of the
tyrant. He contrasted "tyranny" with "human rights and democracy," a view that influenced later
writers[7]. In one of his books, he wrote, "Where law ends, tyranny begins."[8] Voltaire, another
Enlightenment writer, also believed that tyrants were capricious and ignorant of the law[9].
As history progressed, tyrants became closely associated with "cruelty and tyranny. The
Encyclopædia Britannica describes a tyrant as "an absolute ruler who does not obey the law" and
"one who usurps the sovereignty of a rightful ruler."[10] To further understand the role and
stereotypes of "tyrants," we divide "tyrants" into two main groups of subjects of study: tyrants as
recorded in works that existed in history, and tyrants as they appear in works of fiction.
Using Crayon, we generated portraits of 20 "tyrants". The results showed that the "tyrants" had a
number of obvious similarities. First, they are all middle-aged men. Ninety-five percent of the
tyrants have sparse hair. About 50% of the tyrants have beards. Fifteen percent of tyrants are
bald.15% of tyrants have eyes of a different color than humans. They all have serious, grim
expressions on their faces, and 20 percent of them are clearly angry. Twenty-five percent of the
tyrants were dressed in warlike attire, including military uniforms, medals, and armor.
In the previous section, we looked specifically at the backgrounds of the tyrants. The results
showed that they fit the AI-generated stereotypes of tyrants. In particular, their angry expressions
and war-reminiscent clothing indicated cruelty toward others.
Next, we surveyed several tyrants that existed in history and we will give two examples.
Xia Jie. The 17th and last ruler of the Xia Dynasty in China. A tyrant, he led an extravagant and
lewd life. He used the wealth of the people to satisfy his own desires. He is traditionally regarded
as one of the causes of the end of the Xia Dynasty and is portrayed as a tyrant and oppressor by
the people of his time and later generations [11].
His reign was often marked by indifference to internal affairs, and class conflicts became
increasingly severe. According to the Jukjeon, he used the wealth of the people to build palaces
and carve luxurious buildings [12]. He also sought out beautiful women from all over the country
and drank with them day and night [13]. Figure 1 depicts this scene. Not only was he a drunkard,
but he was also excessively strict about food, and any servant who made the mistake of eating or
drinking was immediately executed [14].
Xiajie relied on the vice minister and ostracized the virtue minister. The vicious minister Zhao
Liang gave advice to Xia Jie to please him, to extort and brutalize the people. Minister Cigu, who
saw Xiajie's extravagance, advised him that he would never be loved by the people unless he was
an industrious and frugal emperor who loved them. Another minister, Guan Longfang, repeatedly
advised Xia Jie, accusing him of extravagant spending and murderous habits, telling him that the
people wanted him to die quickly and that only by correcting his own faults could he win their
hearts. Enraged, Jie executed Guan Longbang[15].
In his later years, Xia Jie became even more lewd and even created a lake filled with liquor called
"Night Palace," said to be large enough for a boat to navigate, and his servants often drowned in
it when drunk [16]. As shown in Figure 2, a relief topiary from Shandong depicts him sitting on

17
top of two servants. This not only symbolizes his abuse of power but is also a well-known habit
of his [17].
Leopold II was the second king of Belgium from 1865 to 1909. He was called the "King of
Architecture" because he commissioned many buildings, city plans, and public works projects in
Belgium [18]. Many of these were done with the profits from the development of the Congo Free
State [19].
In the early years of his administration, he was convinced that overseas colonies were the key to
the country's strength. Therefore, he worked vigorously to secure colonial territories for Belgium.
On February 5, 1885, the Congo Free State was established under the personal rule and private
army of Leopold II [20].
Leopold II intended to accumulate wealth through the export of ivory, but it did not reach the
expected level. At that time, in the 1890s, the price of natural rubber was rising, and the free
government decided to put its labor force into the labor-intensive job of sap extraction for rubber
factories. For-profit, the Liberal government led by Leopold II forced local people to work. When
production quotas were not met, beatings, kidnappings, and mutilations of locals were
commonplace, as were widespread killings [21]. There is a photo of a Congolese man sitting on
the ground looking at the mutilated body of his daughter [22]. Estimates of the death toll range
from 2 to 15 million[23].
In 1890, George Washington described the actions of the Congolese Free State government led
by Leopold II as a "crime against humanity. These facts are corroborated by numerous witnesses
and evidence. For example, the on-site inspection by the International Commission of Inquiry and
the 1904 "Casement Report."[24]
From the AI-generated images, the tyrant stereotypes were summarized as follows. Male, middle-
aged, beard, sparse hair (or baldness), serious expression (or anger), eye color different from the
general population (e.g., red eyes), war-inspired clothing. In conjunction with the discussion of
real historical tyrants, we decide that "middle-aged" and "war-associated clothing" should be
retained. There are two reasons for this:
1. "beards" and "thinning or balding hair". These two stereotypes actually reflect the
stereotypical appearance of the middle-aged/villain. Especially the latter. The character
of the tyrant is the villain in most works. Studies have shown that male villains have a
higher percentage of thinning hair [25], which is a stereotype of the villain rather than the
tyrant. The hairstyles of tyrants vary from culture to culture. For example, Xia Jie is one
of the most famous tyrants in Chinese history. According to excavated artifacts and
documents, male lords of the time wore their long hair tied in various shapes on the top
and back of the head[26].In addition, past research[27] has shown that gender change is a
quick and effective way to overturn stereotypes. Therefore, we did not select "male" and
"beard" and left "middle-aged".
2. The three remaining stereotypes are all visual representations of the tyrant's brutal
characteristics. In particular, the "angry expression" and "different eye color from the
general public" are typical means of portraying a brutal character. The tyrants examined
in this study show a common characteristic of cruelty. They used inhumane means such
as torture and murder against others. In addition, an examination of historical documents
and artwork shows that they basically did not treat others as equals. Rather than being
angry, their attitude was to take things for granted. Therefore, we decided to leave

18
"warlike dress" behind. This maintains the stereotype of the tyrant as a brutal murderer,
while at the same time alluding to his identity to some extent.
3.2. Mad Scientist
We have prepared several character archetypes for the exchange of stereotypes. They are the
Lunatic, the Zealot, and the Evil Genius/Mad Scientist.
For continuity in the study of villain stereotypes, we have decided to once again use the villain as
the prototype. This is because both villain stereotypes can be subverted simultaneously. However,
for the practical application of this method, we are not limited to the villain character. It might
have worked better if we had used our positive characters. In most works, the antagonistic nature
of good and evil makes their stereotypes more contrasting.
Let me briefly explain why the pre-selected character prototypes were chosen.
First is the "Lunatic." The "tyrant" is motivated by the misuse of power for their own gain and
enjoyment. The Lunatic archetype is almost always characterized by pure madness. Their
motives and purposes are confused and incomprehensible to the average person. Sometimes their
actions also end up hurting themselves simply because they want them to. They might even
abandon their obsessive goals one moment and completely ignore logic and reason the next,
making this type of character archetype unpredictable [28]. The Joker from DC Comics is an
example of this character [29].
The "Zealot," on the other hand, is characterized by strong beliefs, and nearly all of their actions
are rooted in those beliefs. As villains, their beliefs and actions often harm others, but they
stubbornly believe that what they are doing is best for everyone [30]. In most productions,
fanatics are motivated by religion and distorted moral values, which drive them to pursue their
mission [31]. An example of this character is Melisandre in "Game of Thrones" [32].
The key difference between these two archetypes and the "tyrant" lies primarily in the characters'
motivations. The motives of the Lunatic are chaotic, varied, and sometimes unclear, whereas the
Zealot's motives are strong, stubborn, and self-sacrificing. The motives of the "tyrants" are less
fixed, yet still stable. Due to these differences, we believe that either the Lunatic or the Zealot is a
valid choice. However, the identities of these two archetypes are not as clear-cut as that of the
"tyrant." This is why we have introduced another archetype, the Mad Scientist, which
incorporates characteristics of both the "Lunatic" and the "Zealot."
The Mad Scientist (or Mad Doctor, Mad Professor) is an archetype of a scientist typically
portrayed as mad, dangerous, and evil [33]. These archetypes are also referred to as evil geniuses
when they play the role of villains in their works. They share many similarities with the older
archetype, the Absent-minded Professor. Both are intellectually and academically inclined, often
engrossed in research and experimentation. However, their brilliance often comes at the expense
of competence in other areas. For instance, Absent-minded Professors are frequently forgetful
and careless, while Mad Scientists often exhibit low ethical standards [34]. Mad Scientists are
often ambitious and use their talents to challenge the impossible or attempt to play God [35].
During World War II, fears of technology spiraling out of control due to the atomic bomb and
other technologies of the time led to the emergence of many "mad scientists" in popular culture.
Furthermore, the scientific and technological rivalry between the U.S. and Russia during the Cold
War emphasized scientific progress but also instilled a fear of falling behind technologically.

19
Consequently, "mad scientists" frequently appeared in science fiction and movies during that
period [36].
The Mad Scientist embodies both the strong and unyielding motives of the "Zealot" and the
madness of the "Lunatic." Hence, we have chosen this archetype to replace the "tyrant" in our
study.
Using Craiyon, we generated portraits of 20 "mad scientists". The results showed that the "mad
scientist" had a number of obvious similarities. First, the most obvious feature is a broad
forehead, which is found in 100% of the mad scientist portraits created. The next most obvious
feature is disheveled hair, which is present in 98% of mad scientist portraits.90% of mad
scientists wear glasses or similar goggles.80% of mad scientist faces are thin. In terms of
clothing, 80% of the mad scientists wore white shirts, 45% wore formal academic clothing, and
25% wore white coats. In addition, 25% of the mad scientists had scientific instruments such as
test tubes.
Next, several mad scientists were investigated. Here are examples:
Victor Frankenstein is a typical fictional mad scientist. He was born out of the 1818 novel
Frankenstein by English author Mary Shelley and is the protagonist of this novel [37]. In addition
to being a gothic horror novel,[38] the book is also considered by Brian Wilson to be a precursor
to science fiction[39]. After a successful experiment in which a young scientist, Victor
Frankenstein, creates an intelligent life form called "Frankenstein's monster" out of non-living
matter, Victor is disgusted by his creation and flees the scene in terror. The monster is also
frightened and wanders around, unaware of its true identity. No matter how many times he tries
to integrate into human society, he is always rejected and marginalized. So he decides to take
revenge on his creator: after killing Victor's brother, the monster threatens his creator and
demands that he create a female companion for himself. Victor agrees and begins to build it on an
isolated island. As the creation progresses, Victor begins to regret his decision and destroys the
nearly finished female monster. In a fit of rage, the monster kills Victor's bride and best friend.
Victor was stricken with grief and decided to destroy his creation with his own hands [40].
The novel has had a profound influence on literature and popular culture, and has been adapted
many times for theater and other media. In particular, the American film Frankenstein (1931),
starring Boris Karloff, is considered the most prominent depiction of Frankenstein [41].
However, there are adaptations that do not exactly match the original work. For example,
Frankenstein, the protagonist of the original film, is a more tragic figure. Both the previously
mentioned 1931 film and the film series starring Peter Cushing portray Frankenstein as a typical
"mad scientist" [42]. He is a ruthless egocentric man who uses blackmail, murder, and
intimidation of others to achieve his ends [43].
Frankenstein has evolved from alchemist to mad scientist through the enrichment of these
derivative works. He evolved from alchemist to mad scientist and became one of the archetypal
representatives of the archetype [44].
Rotwang is one of the most famous mad scientists of the early movie screen. He appears in Fritz
Lang's film Metropolis (1927) [45]. The film is a German Expressionist science fiction silent
film. It is on UNESCO's Memory of the World list[46] and is considered one of the first feature-
length science fiction films[47].
Rudolf Klein-Rogge plays Rotwang, a brilliant but evil scientist [48]. He is a mighty scientific
force, and the machines he creates are key to the functioning of Metropolis, but he is ruled by his

20
own fierce desire for revenge and lust [49]. In the film, he falls in love with a woman named Hel,
but Hel chooses to rule the city and dies after giving birth to the protagonist; Rotwang pretends to
have created the robot to resurrect Hel, ostensibly at the behest of the city lord, but in fact he is
secretly working to create a destructive desire among the wealthy elite of Metropolis secretly
instructs the robot to cause [50].
Rotwang's visual image has had a major influence on the stereotype of the "mad scientist" in
popular culture. First, the crude appearance of his laboratory contrasts with his cosmopolitan
fashion sense. The laboratory is filled with scientific tools: bubbling glass containers, twisted and
intricate pipes, and Tesla coils, dials, and controls of various sizes. These elements influenced the
design of many laboratory scenes in subsequent films [51]. Second, his appearance and dress
code are also groundbreaking. In the plot, Rotwang loses a hand in the process of developing a
"robot". He, therefore, fitted himself with a fully functional metal prosthetic hand and put on a
pair of black rubber gloves. This made the black rubber glove the symbol of a mad scientist for a
while. In addition, Rotwang's unkempt hair, crazy and ambitious eyes, and exaggerated smiles
and gestures were also used as references by later mad scientists. For example, Dr. Meirschultz
trying to bring the dead back to life in "Maniac" (1934) [52].
One study examined 1,000 horror films released in the United Kingdom between 1930 and 1980.
The researchers found that 30% of these films featured mad scientists or their creations as
villains, and even when they did not, 39% of the threats in the films appeared as a result of
scientific research. Scientists appear as heroes in only 11% of the films. Most of the evil geniuses
share the same fate, trying to "play God" by creating life, but with doom in the end [53].
The stereotype of the mad scientist generated from AI images is a broad forehead, disheveled
hair, thin, wearing glasses, white shirt, white robe, or formal attire, and carrying scientific tools.
In conjunction with further research on specific mad scientists, we categorized these elements
according to the characteristics of the mad scientist they are intended to display.
Stereotypes reflecting intellectual traits: a broad forehead, glasses.
Stereotypes that reflect madness: messy hair, skinny.
Stereotypes that reflect identity: white shirt, white coat or formal wear, having scientific props.
The stereotypes we ended up keeping were: skinny and having scientific tools. The reasons for
this are as follows:
1. Based on our study of Frankenstein, we thought it was important as a prototype to have
the character embody the madness of trying to play God. Therefore, our first step was to
keep the stereotype in such a way that it embodies the characteristics of madness.
Disheveled hair is a good representation of insanity, but at the same time, it is the
opposite of the stereotype of a tyrant with sparse hair. So we believe that interchanging
these two points should produce a more pronounced result.
2. stereotypes that reflect identity, and chose to have scientific tools. Rotwang's study
shows that even without a lab coat or formal attire, having scientific tools in the
character's hands or in the background can reflect the scientist's identity to some extent.
Therefore, no other clothing options were chosen.
3. stereotypes that reflect intellectual traits were not kept because they are also reflected in
the scientific tools.

21
4. CONCLUSIONS
we once again summarize our methodology and its strengths and weaknesses.Sort by order of
shortest time used:
ChatGPT + Leave stereotypes to show identity
Pros: can be used by people with no design experience. Fastest method.
Cons: May be subjective if the selected character prototype does not have a stereotype to indicate
identity.
AI image generation + select most frequent stereotypes.
Pros: very fast
Cons: need to summarize stereotypes from image results. The highest percentage of stereotypes
are likely not reflective of character archetypes.
ChatGPT + Research
Pros: chatGPT can provide corresponding stereotypes, which saves time.
Cons: Takes time to research information. It doesn't always give the stereotypes that can be
summarized visually.
AI image generation+ research
Pros: Sometimes can summarize stereotypes not mentioned in chatGPT.
Cons: most time-consuming method.
This paper documents the last and most time-consuming method. By examining a large number
of sources and examining what stereotypical influences were used, we were able to achieve a
greater effect of subverting stereotypes. As a result, it was concluded that it would be better to
keep the tyrant's "middle-aged" "war-inspired clothing" and the mad scientist's "skinny" "carrying
scientific tools" and replace the rest of the stereotypes. This would allow the audience to
immediately recognize the original characters and would best distance both characters from their
original stereotypes.
The other method is much less time consuming but somewhat more random. Whether one
chooses by subjective experience or by the most frequent choices, there is no guarantee of the
best outcome. In other words, it is the one that best guarantees that the audience will be able to
quickly identify the original character and at the same time move the two characters the furthest
away from the original stereotypical image of the original.
In conclusion, if the designer has sufficient time, ai portrait + research or chatGPT + research can
be chosen. If there is not enough time, the remaining methods can be chosen. The remaining
methods take less time and the designer can try them all to get the desired result.
ACKNOWLEDGEMENTS
The authors would like to thank everyone, just everyone!

22
REFERENCES
[1] Fedorov, Alexander (2015). "Media Stereotypes Analysis in the Classroom at the Student
Audience". European Journal of Contemporary Education. 12 (2): 158–162.
[2] Knight, Gladys L. (2010). Female Action Heroes: A Guide to Women in Comics, Video Games,
Film, and Television. ABC-CLIO. p. 62. ISBN 978-0-313-37612-2.
[3] Davison, Heather K.; Burke, Michael J. (2000). "Sex Discrimination in Simulated Employment
Contexts: A Meta-analytic Investigation". Journal of Vocational Behavior. 56 (2): 225–248.
[4] Ajrouch, Kristine J.; Antonucci, Toni C.; Janevic, Mary R. (1 March 2001). "Social Networks
Among Blacks and Whites: The Interaction Between Race and Age". The Journals of Gerontology.
56 (2): S112–S118. doi:10.1093/geronb/56.2.S112. PMID 11245365.
[5] "Constitutional Rights Foundation". www.crf-usa.org. Retrieved 2023-05-17.
[6] The Internet Classics Archive | Politics by Aristotle". classics.mit.edu. Retrieved 2023-05-17.
[7] "John Locke – Biography, Treatises, Works, & Facts". britannica.com. Archived from the original
on 19 July 2017.
[8] John Locke, Two Treatises of Government (199)
[9] Voltaire (1824). A Philosophical Dictionary, Volume 1. J. and H.L. Hunt. p. 76.
[10] Chisholm, Hugh, ed. (1911). "Tyrant". Encyclopædia Britannica. Vol. 27 (11th ed.). Cambridge
University Press. p. 548.
[11] Liu Wei. (2002) Chinese civilization in a new light. Publishing Company. ISBN 962-07-5314-3.
Pp.150.
[12] Virginia.edu. "Virginia.edu contents of Bamboo annal." Xia chapter. Retrieved on 2023-5-18.
[13] Wang Heng wei. (2005) Lecture Hall of Chinese History # 1 from ancient times to the spring and
Autumn period. Zhonghua Book Company. ISBN 962-8885-24-3. P 26.
[14] Wang Heng wei. (2006) Lecture Hall of Chinese History # 1 from ancient times to the spring and
Autumn period. Zhonghua Book Company. ISBN 962-8885-24-3. Karen. P 27.
[15] Dai Yi, Gong Shu duo. [2002](2003) general history of China. Prehistoric Xia, Shang and Western
Zhou dynasties. Intelligence Press. ISBN 962-8792-80-6. P 51.
[16] S.J Marshall. (2002) The Mandate of Heaven: Hidden History in the I Ching. Columbia university
press. ISBN 0-231-12299-3, ISBN 978-0-231-12299-3. p 177-178.
[17] James Legge (1865), The Chinese Classics, Volume 3, part 1. page 125.
[18] "Leopold II". The Belgian Monarchy. www.monarchie.be/history/leopold-ii. Retrieved 4 may 2023.
[19] One or more of the preceding sentences incorporates text from a publication now in the public
domain: Chisholm, Hugh, ed. (1911). "Leopold II., King of the Belgians". Encyclopædia Britannica.
Vol. 16 (11th ed.). Cambridge University Press. p. 461.
[20] Hochschild, Adam (1998). King Leopold's Ghost: A Story of Greed, Terror and Heroism in
Colonial Africa. Mariner. ISBN 978-0-330-49233-1. OCLC 50527720. pp123-124.
[21] Neal Ascherson, The King Incorporated: Leopold the Second and the Congo (Granta Books, 1999)
[22] Thompson, T. Jack (October 2002). "Light on the Dark Continent: The Photography of Alice Seely
Harris and the Congo Atrocities of the Early Twentieth Century". International Bulletin of
Missionary Research. 26 (4): 146–9.
[23] Forbath, Peter (1977). The River Congo: The Discovery, Exploration and Exploitation of the
World's Most Dramatic Rivers. Harper & Row. p. 278. ISBN 978-0-06-122490-4.
[24] Hochschild, A. King Leopold's Ghost: A Story of Greed, Terror and Heroism in Colonial Africa.
Houghton Mifflin, 1999. pp. 111–112.
[25] Kyriakou, G., Drivelou, V. and Glentis, A. (2021), Villainous hair: ba(l)d to the bone – would they
be so evil if they had hair?. Br J Dermatol, 184: 156-157. https://doi.org/10.1111/bjd.19508
[26] Zhang Chunxin, Gou Shixiang zhu, “Fa ji shang de Zhongguo”, Chongqing Publisher, 2011, pp. 21.
[27] Xiaohan Feng, Makoto Murakami, “DESIGN THAT USES AI TO OVERTURN STEREOTYPES:
MAKE WITCHES WICKED AGAIN”, International Journal of Artificial Intelligence and
Applications (IJAIA), Vol.14, No.2, March 2023, pp.51.
[28] Shennandoahdiaz, " Villain Archetype: The Lunatic". August 11, 2010,
https://shennandoahdiaz.wordpress.com/2010/08/11/villain-archetype-the-lunatic/.
[29] [29] Stewart, Tom (August 2009). "The Joker's Not So Wild! The Clown Prince of Crime in his
Own Magazine!". Back Issue!. Raleigh, North Carolina: TwoMorrows Publishing (#35): 40–43.
[30] Cole Salao, " 12 Types of Villains in Fiction". https://www.tckpublishing.com/types-of-villains/，
Retrieved June 22, 2023.

23
[31] MasterClass. "T 8 Villain Archetypes: How to Write Different Types of Villains". Sep 30, 2021.
https://www.masterclass.com/articles/how-to-write-different-types-of-villains
[32] "Game of Thrones Cast and Crew: Melisandre played by Carice van Houten". HBO. Archived from
the original on April 7, 2015. Retrieved June 22, 2023.
[33] Frayling, Christopher – Mad, Bad and Dangerous?: The Scientist and the Cinema (Reaktion Books,
2005) ISBN 1-86189-255-1
[34] Corbett, Robert (2001). "Romanticism and Science Fictions". Romanticism on the Net (21): 0.
doi:10.7202/005970ar.
[35] Haynes, Roslynn Doris (1994). From Faust to Strangelove: Representations of the Scientist in
Western Literature. Baltimore: Johns Hopkins University Press. ISBN 0-8018-4801-6.
[36] G., Fraser (1998-01-01). The Particle Century. ISBN 9781420050332. Retrieved 24 January 2017.
[37] Clemens, Valdine. Return of the Repressed, The: Gothic Horror from The Castle of Otranto to
Alien. p. 93. ISBN 9780791499276. Retrieved 10 November 2015.
[38] Tweg, Sue; Shelley, Mary Wollstonecraft; Edwards, Kim (August 2011). Frankenstein. p. 13. ISBN
9781921411397. Retrieved 10 November 2015.
[39] Aldiss, Brian Wilson (1995). The detached retina : aspects of SF and fantasy. Internet Archive.
Syracuse, N.Y. : Syracuse University Press. ISBN 978-0-8156-2681-7.
[40] Mary Shelley, with Percy Shelley (2008). Charles E. Robinson (ed.). The Original Frankenstein.
Oxford: Bodleian Library. ISBN 978-1-851-24396-9. Archived from the original on 25 September
2015.
[41] Anderson, John (25 January 2022). "'Boris Karloff: The Man Behind the Monster' Review: A Very
Different Creature". The Wall Street Journal. ISSN 0099-9660. Retrieved 26 January 2022.
[42] Geraghty, Lincoln (2009-10-01). American Science Fiction Film and Television. ISBN
9780857850768. Retrieved 28 June 2023.
[43] Legget, Paul (2018). Good Versus Evil in the Films of Christopher Lee. McFarland, Incorporated,
Publishers Ltd. pp. 09–12. ISBN 978-1-476669-63-2.
[44] Corbett, Robert (2001). "Romanticism and Science Fictions". Romanticism on the Net (21): 0.
doi:10.7202/005970ar.
[45] "Metropolis (1927)" Archived 10 October 2017 at the Wayback Machine Science Fiction Film
History. Retrieved 4 July 2023.
[46] "Memory of the World: Metropolis – Sicherungsstück Nr. 1: Negative of the restored and
reconstructed version 2001" UNESCO website. Accessed: 4 July 2023.
[47] Magid, Annette M. "Better than the Book: Fritz Lang's Interpretation of Thea von Harbou's
Metropolis". Spaces of Utopia. No. 2. Universidade do Porto. pp. 129–149, 2006.
[48] Hall, Mordaunt. "MOVIE REVIEW Metropolis (1927) A Technical Marvel". The New York Times,
1927.
[49] Geraghty, Lincoln (2009-10-01). American Science Fiction Film and Television. ISBN
9780857850768. Retrieved 4 July 2023.
[50] Hahn, Ronald M.; Jansen, Volker. Die 100 besten Kultfilme (in German). Munich, Germany: Heyne
Filmbibliothek. p. 396, 1998.
[51] Schulze-Mittendorff, Bertina. "The Metropolis Robot – Its Creation". Archived from the original on
3 September 2014. Retrieved 17 February 2016.
[52] Claxton, Oliver. "The Current Cinema". The New Yorker. New York: F-R Publishing Company. pp.
80–81, 1927.
[53] Christopher Frayling, New Scientist, 24 September 2005.

Subverting Characters Stereotypes: Exploring the Role of AI in Stereotype Subversion

More Related Content

More from gerogepatton (20)

Recently uploaded (20)

Subverting Characters Stereotypes: Exploring the Role of AI in Stereotype Subversion