The Men Behind The Screens: How online communities drive nonconsensual sexual deepfake creation 

Feature Image Credit: Sora Shimazaki on Pexels. 

Taylor was with her boyfriend when she received the message. A link to a Pornhub profile and the words “I’m really sorry but I think you need to see this.” 

When she opened the link, she was met with a collection of pornographic videos she had never seen before… but they all featured her face. 

Photos from her social media had been posted in an online forum by an anonymous user asking others to edit her into explicit content.  

And she’s not the only one.

Nonconsensual sexual deepfakes are spreading online. There are around 15 websites dedicated to deepfake porn, with over 300 million views.  

They aren’t hidden in dark corners of the web: they are easily accessible. It’s an industry, it’s a culture.  

At the heart of this culture are private online forums where members share techniques and tools. 

These are ‘the men behind the screens’ driving this troubling trend. 

Inside the forums

AI generators can make synthetic sexual content from scratch. The images look realistic, but they don’t feature real people, and are often used to create fantasy content. 

Now, this phenomenon has developed into a culture of using real images of real women. 

Specialised subgroups on Reddit, 4chan, Discord, and Telegram host hundreds of users creating sexual content with AI, hidden behind anonymous profiles and VPNs. 

These communities have fostered a ‘collector culture,’ swapping and sharing images of women to be inserted into crude content. 

It is often framed as an exchange.

Unlike other forms of intimate image abuse like revenge porn, the perpetrators often have no intention for the women featured to ever find out. 

Motivations and Culture

So, why do they do this?

Journalist Emanuel Maiberg has spent hours inside of these forums, investigating these communities.

He said: “There is a technical skill-sharing and image sharing, but there’s also a culture of ‘we’re just watching pornography,’ without really acknowledging the harm. 

“You’re just sitting on a computer, clicking around, it’s so easy to do, and it’s normalised in those communities and also outside.”  

AI tools developed deliberately for the creation of sexual deepfakes, such as nudification apps, advertise themselves openly on social media. 

“These communities are telling you it’s okay, porn sites are telling you it’s okay, Instagram is telling you it’s okay,” Maiberg said.  

A survey of 1522 members of deepfake pornography websites found that 74% do not feel guilty about it, listing reasons like “I know it’s not really the person” and “I don’t think it hurts anyone as long as it’s only for my personal interest.” 

Over half said that viewing and creating deepfake content was driven by an interest in AI technology. Others explained it as fulfilment of sexual fantasies.

One framework: Toxic Geek Masculinity

Researcher and professor Mel Stanfill is interested in the ideologies that drive the behaviours of these communities. 

They connect them to the broader ‘manosphere,’ an umbrella term for online content ranging from figures like Andrew Tate to the incel community.

Stanfill describes this culture as fostering a ‘toxic geek masculinity,’ which refers to men with traditionally ‘marginalised’ interests like games, comics, and technology.

This ‘geek masculinity’ becomes toxic when technological skill is used to harm others.  

Stanfill explained that these men, feeling they lack traditional masculine power, perceive themselves as marginalised: “They have a harder time seeing the ways that they could hurt others because they’re in this idea of ‘we’re powerless, what could we possibly do.”  

The technology

How are these communities able to make this content?

Well, there is very little stopping them. While laws in the UK and USA are changing, enforcement lags behind the technology.  

There are platforms devoted to creating this content, but AI tools developed by companies like Google and Microsoft are also being misused.

Maiberg explained: “There is a lot of money being poured into generative AI as an industry, and there is a race to see who is going to be the dominant player in this space. 

“There’s pretty good evidence that companies are releasing these tools too fast without considering the consequences.” 

Many users in these communities are tech-savvy. They bypass built-in safeguards, use VPNs to remain untraceable, hide behind anonymous profiles, and have quickly learnt how to use disguised language to avoid detection. 

Maiberg explained that most AI tools now have a list of prohibited terms to try and stop the creation of explicit content, but users have found ways to describe scenarios without using overtly sexual language. 

He said: “They play these word games that would fool a machine, but not a person.”  

The language used in Reddit and 4chan forums is much more open and unapologetic. Once someone uploads an image of a woman, others see it as free rein to create and upload explicit content.  

If the chat says it’s okay, it must be okay, right?  

Nonconsensual sexual deepfakes exemplify a troubling crossover of rapid AI advancements and harmful online cultures.   

These manipulated images and videos are often created and shared without consequence.

However, counter-campaigns are emerging, calling for legal frameworks, greater responsibility from tech platforms, and increased public awareness.  

As AI continues to develop, the rise of nonconsensual sexual deepfakes raises important questions about how we are to balance innovation and regulation.

---