How does AI, specifically deepfake tools, contribute to the dangers women face on social media?

Word count: 1055

women have to think about their safety around the clock. It isn’t always as obvious of a hooded

man in a van to look out for. Rather, more hidden dangers that affect women every day.

Body dysmorphia, eating disorder, and perfectionism are known to be dangers of social

media. Multiple studies show, including a cross-sectional study of UK based 17 year olds, done

by Frontiers in Psychology, found that the more social media use young teens have, the more

body dysmorphia symptoms found(Tarlow et al., 2023). In addition, AI tools, specifically deep

fakes, pose the risk of taking advantage of young girls' perception of themselves and the world

around them.

To understand how dangerous AI tools can be, let's start with defining deep fakes.

According to the peer reviewed article, “The Word Real Is No Longer Real”: Deepfakes, Gender,

and the Challenges of AI-Altered Video”(Wagner & Blewer, 2019), it is a portmanteau, with

‘deep’ meaning learning, and ‘fake’ meaning content. It is a prototype of AI. How it is made is

through two low quality pictures of people's faces and then the computer uses a model to match

each other's facial features together. And the more times it does that, the more realistic it will

look. Eventually, the computer is able to make new fake images that have never existed at all.

Nude deep fakes of celebrities are perfect examples of deep fakes and just how dangerous they

are(Wagner & Blewer, 2019) . For example, celebrities like 17-year-old Marvel actor Xochitl

Gomez spoke about her image based abuse experience specifically, sexually explicit pictures

posted all over X and said “It made me weirded out and I didn’t like it and I wanted it taken

down……this has nothing to do with me. And yet it’s on here with my face.”(The Guardian,

2024). And to highlight, Xochitil is also a minor. Showing that even minors aren’t safe from the

dark sides of the web that need to be unveiled before more innocent women are hurt. Another
example would be british poet and novelist, Helen Mort who had deepfake images circulated on

the internet showing acts of extreme sexual violence that she is participating in. She reported

having nightmares and feeling extreme anxiety to even leave the house.(BBC, 2021).

Even porn websites like Naughty America encourage this type of behavior by showing their

ability to swap faces for these adult films. And not only are big companies doing this, it is

encouraging “user-generated work” where people can make their own fake videos as well.

(Wagner & Blewer, 2019) I have seen it happen to users and people on my own TikTok feed.

Where it is just a normal video of a pretty girl dancing like Brookemonk and then under the

video is a search bar of “Brooke Nudes” or “Brooke red shirt” where it will take you to fake

pictures of users being sexually explicit. When I went to find websites talking about this one of

the first to pop up was a website called BlipCut AI Video Translator and the headline was “How

to Create Brook Monk Deep Fake Videos in 2025.” It goes to talk about how to ​

“Best AI tool in the industry” to make these disturbing photos. If I, someone who has no

business or knowledge of making websites or AI, can find out how to make deep fakes with a

single search, imagine the kind of damage that millions of horrible people can do with this

information.

In addition, based on Diptrace Labs, now known as Sensity AI, their research shows 80% of the

Top 10 pornographic websites host and allow deep fakes. Not only this but they are 9 porn

websites that are dedicated to just deep fake content. (Chesney & Citron, 2019) And being put on

these websites can happen to anyone, not just celebrities. Apps like Deepnude take explicit

pictures of everyday women and the technology allows them to take the woman's clothes off.

And even more unsettling, the technology can only take off women's clothes- not mens. The

website received over half a million visits before it got taken down. However, there still is
technology online users can use independently. (Chesney & Citron, 2019) Deepfakes and image

based abuse is not a niche problem as 57% of women told an ESET Research survey that they

are worried of becoming victims of porn deep fakes. This isn’t a niche problem and it must be

regulated and solved with efficiency and empathy to the victims. (ESET UK, 2023).

​ Not only are pornographic deep fakes made to harm women and make them feel like they

must live in an objectified world, but there are other kinds of deep fakes that cause the same

sense of disturbance and invasion of privacy. For example, Nancy Pelosi was a victim of a

deepfake that made her appear to be drunk and slurring her words during an important event.

This caused people to question her being fit for office. And her being a woman in office, people

were looking for any kind of reason to make her look small and silence her as a woman in power.

These deepfakes can cause mental distress as well as “social, professional, and personal

ramifications.” In general when women have found that their content has been deepfaked, they

report feeling “humiliation, violation, fear, helplessness, and powerlessness, [and suicidal

ideation].” They also feel effects from the outside world from online and offline harassment and

damaged reputations. (Citron, 2019)

​ As felt by many women who have been victims to deep fakes, “cyber abuse is rooted in

unequal powers of control, resources, and social rules that seek to preserve a man's dominance

over their female counterpart” (Bourgeois, 2022). In simpler terms, regardless of if the goal is to

undermine a woman's intelligence or shame them for their sexuality, by putting dangerous

resources in the hands of these men, it will be used as a tool to silence women of the world.

Nothing about this kind of AI to be legal or ethical. Whether AI deep fakes are used to switch up

a congresswoman's words or take control of a young girl's body, it is an infringement on their

bodily autonomy and causes long lasting effects on their life and mental health.
 Citations:

Bourgeois, D. (2022). Deepfake technology and its implications for media literacy education. Digital

Learning and Leadership, 4(1), 38–48.

[Link]

Chesney, R., & Citron, D. K. (2019). Deep fakes: A looming challenge for privacy, democracy, and

national security. [Link]

Citron, D. K. (2019). Sexual privacy. Yale Law Journal, 128(7), 1870–1960.

[Link]

ESET UK. (2023, November 28). Nearly two-thirds of women worry about being a victim of deepfake

pornography, ESET UK research reveals. ESET.

[Link]

ut-being-a-victim-of-deepfake-pornography-eset-uk-research-reveals/

Hern, A. (2024, January 31). Inside the Taylor Swift deepfake scandal: 'It's men telling a powerful

woman to get back in her box'. The Guardian.

[Link]

s-men-telling-a-powerful-woman-to-get-back-in-her-box

Wagner, Travis L. and Blewer, Ashley. "“The Word Real Is No Longer Real”: Deepfakes, Gender, and

the Challenges of AI-Altered Video" Open Information Science, vol. 3, no. 1, 2019, pp. 32-46.

[Link]
West, S. M. (2019). Data capitalism: Redefining the logics of surveillance and privacy. Open

Philosophy, 2(1), 254–270. [Link]