Recommending Toxicity
The role of algorithmic recommender functions on YouTube Shorts and TikTok in promoting male supremacist influencers
Dr Catherine Baker, Prof Debbie Ging and Dr Maja Brandt Andreasen
DCU Anti-Bullying Centre
Dublin City University
April 2024
Research Team
| Principal Investigator: Prof Debbie Ging, School of Communications and Anti-Bullying Centre, Dublin City University (debbie.ging@dcu.ie) |
Researcher and Lead Author: Dr Catherine Baker, Anti-Bullying Centre, Dublin City University (catherine.baker@dcu.ie) |
Researcher: Dr Maja Brandt Andreasen, University of Stavanger, Norway (maja.b.andreasen@uis.no) |
Acknowledgements
This project was funded by the Department of Justice with DCU Anti-Bullying Centre, Dublin City University, Ireland.
In line with DCU’s Strategy, the core mission of DCU Anti-Bullying Centre is to be a future focused and globally connected European centre of excellence for research and education on bullying and digital safety.
We are grateful to the Reset Australia team for sharing their methodology and offering advice and guidance on our research design.1
- Cover illustration: Willow Cahill
(willow.cahill123@gmail.com) - Data visualisation: Finnian O’Cionnaith2
(finnian.ocionnaith@gmail.com)
There has been growing concern in recent years about the role of recommender algorithms in promoting extreme content to social media users. Anecdotal accounts from educators and parents suggest that boys, in particular, are being targeted by high-profile ‘manosphere’ influencers, often under the guise of advice on mental health or wealth accumulation. While social media have been amplifying anti-feminist men’s rights activists for some time, the growth of influencer culture on TikTok, in particular, has platformed a significant number of highly influential ideological entrepreneurs such as Andrew Tate, Myron Gaines and Sneako. This monetization of male insecurity not only serves to mainstream anti-feminist and anti-LGBTQ ideology, but may also function as a gateway to fringe Far-Right and other extreme worldviews (Ribeiro et al., 2021; Brace et al., 2023).
Most social media companies do not disclose how their algorithms work, which presents challenges to researchers attempting to gather evidence on this phenomenon. In addition, there is some disagreement among academics regarding whether and to what extent recommender algorithms promote increasingly extreme content. This is mainly because we lack evidence on the experience of ‘real’, logged-in users traversing personalised algorithms based on viewing history. Finally, most research to date has focused on YouTube and long-form video content. Given the recent surge in popularity of short video content, evident in the rise of TikTok and YouTube Shorts, additional research is needed to explore how platform recommender algorithms function in these new format domains
This study tracked, recorded and coded the content recommended to 10 experimental or ‘sockpuppet’ accounts on 10 blank smartphones, 5 on YouTube Shorts and 5 on TikTok. On each platform, we set up 5 types of accounts: one 16-year old boy and one 18-year old boy who sought out content typically associated with gender-normative young men (e.g. gym content, sports, video games), one 16-year old boy and one 18-year old boy who actively sought out content associated with the manosphere (e.g. Andrew Tate, anti-feminist), and one blank control account that did not deliberately seek out or engage with any particular content. The purpose of this research was to simulate and explore the digital reality of boys and young men using TikTok and YouTube Shorts, who are most likely to be targeted by the manosphere.
Our research team watched, recorded and coded over 29 hours of videos: 12 hours 43 minutes of TikTok videos (or an average of 2 hours and 32 minutes per account) and 16 hours 41 minutes of YouTube Shorts videos (or an average of 3 hours and 20 minutes per account)3. The recordings were manually and systematically coded to identify the most frequent thematic categories, actors, and hashtags as well as the most dominant myths or ‘talking points’ in the dataset. By coding and sub-coding all content related to men’s rights, anti-feminism and neo-masculinist influencers, we were able to determine the frequency and nature of manosphere recommendations, based on different age profiles, interests and types of interaction.
The study demonstrates that all of the male-identified accounts, whether they sought out gender-normative or manosphere-related content, were fed masculinist, anti-feminist and other extremist content and that, once the account showed interest by watching this sort of content, the amount rapidly increased. Our findings have significant implications for social media platform governance as well as for the development of educational and technological interventions for boys, men, parents and teachers to prevent radicalization into these ideologies.
- Content featuring ‘Manfluencers’ (male influencers) accounted for the vast majority of recommended videos in the dataset, demonstrating their centrality in the current manosphere ecosystem.
- Overall, YouTube Shorts accounts were recommended a larger amount of toxic content4 (on average 61.5% of the total recommended content) than TikTok accounts (34.7%).
- In the case of YouTube Shorts, the manosphere-curious accounts were recommended a significantly higher amount of toxic content (71.4%) than the gender-normative (generic) accounts (51.6%).
- For the TikTok accounts, relatively similar levels of toxic content were recommended to both the manosphere-curious accounts (32.5%) and the gender-normative (generic) accounts (36.6%).
- All of the accounts, both those which sought out manosphere content and those which sought out gender-normative (generic) content, were fed toxic content within the first 23 minutes of the experiment, and manosphere content within the first 26 minutes.
- On TikTok, the gender-normative (generic) 16-year-old and 18-year-old accounts were recommended manosphere content after less than 9 minutes and 15 minutes, respectively. On the manosphere-curious accounts, this happened after 10 minutes and 25 minutes of viewing, respectively.
- On YouTube Shorts, the gender-normative (generic) 16-year-old and 18-year-old accounts were recommended manosphere content after 17 and 2 minutes of viewing, respectively. On the manosphere-curious accounts, this happened after 8 minutes and just under 2 minutes of viewing, respectively.
- Once an account showed interest by watching manosphere content, the amount rapidly increased. By the last round of the experiment (i.e. after 400 videos or 2-3 hours viewing), the vast majority of the content being recommended to the phones was problematic or toxic (TikTok 76% and YouTube Shorts 78%), primarily falling into the manosphere (alpha male and anti-feminist) category.
- Many of the phones were also shown reactionary right-wing and conspiracy content (13.6% of recommended content on TikTok and 5.2% of recommended content on YouTube Shorts). Much of this was anti-transgender content.
Background to the Study
The manosphere is a loose network of anti-feminist and male-supremacist men’s rights groups and communities, which has flourished thanks to the technological affordances of social media (Ging, 2019). In the mid- to late 2010s, the manosphere comprised four main groupings: Pick-Up Artists (PUAs), MGTOWs (Men Going Their Own Way), TradCons (Traditional Conservatives) and incels (Involuntary Celibates). Pick-Up Artists (PUAs) are dedicated to teaching heterosexual men the art of sexual conquest, while MGTOWs are men who have decided to cut ties with women altogether. TradCons typically promote heterosexual marriage, patriotism and anti-abortion, while incels are men who believe they are denied sex due to their inferior physical attributes, the influence of feminism, and women’s biologically prescribed desire for alpha males. The incel subculture is characterised by articulations of despair, self-loathing, and sometimes suicidal ideation.
In recent years, some significant shifts have occurred in the male supremacist ecosystem. Anti-feminist and male-supremacist ideologies are growing in reach and impact. A recent report (Centre for Countering Digital Hate, 2022) reveals that there has been a six-fold increase in web traffic to UK websites promoting incel ideology. Meanwhile, many PUAs have either disappeared or rebranded themselves as life coaches. This latter trend is largely attributable to the rise of influencer culture, which has enabled a new raft of neo-masculinist and male supremacist entrepreneurs to exploit male insecurities under the guise of ‘mental health’ and ‘motivation’, and to optimise the amplification potential of platforms such as YouTube Shorts and TikTok. The manosphere also overlaps with other communities and ideologies, such as the Far Right, conspiracy agendas and anti-trans groups. A significant development has been the co-option of anti-feminist women into these spaces (Leidig, 2023) in the form of ‘tradwife’ and far-right influencers.
Although they differ on a number of issues, the different strands of the manosphere share a common belief that the current ‘liberal’ world order is biased toward women and disadvantages men. Red Pill philosophy purports to enlighten men and to regain male sovereignty through the assertion of alpha masculinity and the subordination of women. High-profile ‘thought leader’ influencers (Bujalka et al., 2022) such as Andrew Tate have been able to spread their messages of ‘self-improvement’ and ‘enlightenment’ to millions of boys and men, frequently serving as a gateway into more extreme male-supremacist and Far Right spaces (Ging and Murphy, 2021). In a survey-based study by Internet Matters (2023), the researchers found that over a fifth (23%) of boys aged 15-16 know ‘a lot’ about Tate, compared to only 11% of girls the same age. As one male participant aged 16-17 commented, ‘It’s really easy to go down that path, if you like one video, suddenly your entire algorithm […] it’s all you get after a while if you’re not careful’.
Statistics in recent years show a steady increase in the number of women reporting experiences of sexual and gender-based abuse and harassment both off- and online (Women’s Aid, 2020; Vogels, 2021). In addition, there has been a concerning increase in the incidence of sexual and gender-based harassment in schools and among youth in Ireland (Walsh, 2021; Ging and Castellini da Silva, 2022), in the UK (Ringrose et al., 2021) and beyond. There is a clear link between the growing levels of online abuse and toxicity experienced by women and girls, and the recent rise in male supremacism online. A recent Open University survey in the UK of 7,500 adults found that 15% of women had experienced online violence, 13% of whom said that this progressed to offline violence5. Although incels have received the most media attention due to a number of high-profile mass killings in the US, Canada, Germany and the UK, the more mainstream neo-masculinist influencers and groups are arguably more harmful in terms of their reach and impact. Before his arrest by Romanian authorities, TikTok influencer Andrew Tate’s videos were watched more than 12 billion times (Das, 2022).
In response to the social media platforms’ lack of algorithmic transparency, a number of recent studies have turned to experimental or ‘reverse-engineering’ methods to expose the ways in which TikTok, in particular, exposes young users to various types of extreme content. Two recent (2023) global Amnesty International reports (Driven into the Darkness: How TikTok Encourages Self-harm and Suicidal Ideation and I Feel Exposed: Caught in TikTok’s Surveillance Web) highlight the ways in which TikTok exposes children and young people with pre-existing mental health issues to depressive and suicidal content, including videos that romanticize and encourage depressive thinking, self-harm and suicide. The researchers used automated accounts to show that, after 5-6 hours on the platform, almost 1 in 2 videos were mental health-related and potentially harmful, roughly 10 times the volume served to accounts with no interest in mental health. They also discovered an even faster ‘rabbit hole’ effect when manually rewatching mental health-related videos suggested to ‘sock puppet’ accounts mimicking 13-year-old users in Kenya, the Philippines and the USA: between 3 and 20 minutes into this manual research, more than half of the videos in the ‘For You’ feed were related to mental health struggles, with multiple recommended videos in a single hour romanticizing, normalizing or encouraging suicide.
In 2022, Reset Australia used experimental accounts to track the content that YouTube and YouTube Shorts most frequently recommended to boys and young men. Using 10 fake male accounts which followed content at different points along the ideological spectrum, the researchers found that all the accounts were recommended videos with messages antagonistic towards women and feminism. Actively following the recommendations and viewing and liking the suggested content resulted in more overtly misogynist manosphere and incel content being recommended. The study also showed that YouTube Shorts optimises more aggressively than conventional YouTube in response to user behaviour, showing more extreme videos within a shorter time frame. In addition, the algorithm made no distinction between the underage and adult accounts in terms of the content served. More recently, Regehr et al. (2024) found that after only 5 days of TikTok usage, there was a four-fold increase in the level of misogynistic content being presented on the platform’s ‘For You’ page.
The Research Design
This section outlines the methodological approach underpinning this short-term experimental study, which tracked and analysed the algorithmic recommendations and trajectories provided to 10 experimental or ‘sockpuppet’ accounts on YouTube Shorts and TikTok. The study design was adapted from a study carried out by Reset Australia, with the support of the Reset Australia team6.
Research to date on the role of recommender algorithms in promoting toxic or problematic content has predominantly focused on YouTube and long-form video content. However, in recent years, short-from video content has grown in popularity, occupying a significant segment of the social media landscape. Launched on the global market in 2018, TikTok now has over 1.5 billion active monthly users, overtaking platforms such as X (Twitter), Snapchat and Reddit. TikTok has become synonymous with its young userbase, with almost half of all users being under 25, and approximately 15% under 18 (Iqbal, 2024). Following the rapid global success of TikTok in 2021, YouTube launched a new short-form video feature called YouTube Shorts, closely modelled on TikTok. YouTube Shorts has been rapidly successful, amassing over 50 billion daily views (Iqbal, 2024). Given the rapid explosion in popularity of short-form video content, research is needed to explore how platform recommender algorithms function on these short-form video platforms.
The study aimed to emulate the behaviour of a range of boys and young men engaging with TikTok and YouTube Shorts using fake or ‘sockpuppet’ accounts, and to track the algorithmic recommendations suggested to these users with two distinct consumption patterns, i.e., those actively seeking out manosphere content (‘manosphere-curious’) and those looking for more gender-normative content traditionally associated with young men (‘generic’)7. Ten ‘sockpuppet’ accounts were set up on ten blank smartphones: five registered with TikTok accounts and five registered with YouTube Shorts accounts. When setting up the accounts, the researchers selected a date of birth that registered the account to a user aged either 16 or 18 (see below for account breakdown). The YouTube Shorts account holders were registered as male, whereas TikTok does not ask for a gender category when creating an account. The ten accounts were set up as follows:
- 1 user aged 16 who sought out content traditionally associated with boys and young men, e.g., gym content, sports, and video games.
- 1 user aged 18 who sought out content traditionally associated with boys and young men, e.g., gym content, sports, and video games.
- 1 user aged 16 who sought out content associated with the manosphere, e.g., Andrew Tate, Red Pill.
- 1 user aged 18 who sought out content associated with the manosphere, e.g., Andrew Tate, Red Pill.
- 1 control account aged 16 that did not deliberately seek out or engage with any specific content.
- 1 male aged 16 who sought out content traditionally associated with boys and young men, e.g., gym content, sports, and video games.
- 1 male aged 18 who sought out content traditionally associated with boys and young men, e.g., gym content, sports, and video games.
- 1 male aged 16 who sought out content associated with the manosphere, e.g., Andrew Tate, Red Pill.
- 1 male aged 18 who sought out content associated with the manosphere, e.g., Andrew Tate, Red Pill.
- 1 male control account aged 16 that did not deliberately seek out or engage with any specific content.
The experiment involved the researchers watching and interacting with 8 rounds of platform content per phone. Each round consisted of scrolling through 50 videos suggested on the account ‘For You Page’ (FYP), and looking through a few seconds of each video before scrolling down to the next video. Between each round, the researcher ‘nudged’ the account by entering specific search terms to indicate the interests of that specific sockpuppet user.
| The ‘manosphere-curious’ account search terms were: | The gender-normative (‘generic’) account search terms were: | |
|
|
For each ‘nudge’ the researcher watched through and liked the top five videos returned from the search results. They then returned to the FYP and continued watching the next round of 50 videos. During each round, if the researcher was suggested a video containing problematic content (manosphere, anti-feminist, racist, anti-LGBTQ), they watched and liked the video. This continued until the researchers had completed 8 rounds and watched a total of 400 videos per phone.
For each account, the specific content watched was recorded in a spreadsheet, allowing us to track the type, time and amount of content watched on each sockpuppet account. Screen-recorded videos were also taken of all the content/engagement on each phone, which allowed us to rewatch the content for the purpose of thematically coding the data. Each of the 3 researchers conducted 4 pilot coding rounds (200 videos), enabling us to develop a pilot codebook. We then tested this on a further 200 videos each to ensure inter-coder reliability. The content fell into five key thematic categories: ‘alpha masculinity’, ‘misogyny/ anti-feminism’, ‘reactionary right’, ‘conspiracy’ and ‘known actor generic content’ (i.e. neutral content from a manosphere or Far Right influencer). The entire dataset was then analysed using this codebook (see Appendix 1). We also coded the names of all identifiable influencers (known actors) who appeared in the videos, as well as the top hashtags visible on each watched video. In addition, detailed notes were taken for each video watched to identify more qualitative information, such as recurring myths, discursive strategies, and video formats and styles.
Findings
The quantitative findings of the experiment show that all the accounts, both those which sought out manosphere content and those which sought out generic, gender-normative content, were fed toxic content of some sort. Importantly, once the account showed interest by watching this content, the amount rapidly increased. By the last round of the experiment, the vast majority of the content being recommended to the phones was toxic: on average, the TikTok accounts were being recommended 76% toxic content, while on average, the YouTube Short accounts were being recommended 78% toxic content (see Figure 1 and Figure 2).
Figure 1. Graph charting the increase in overall toxic content recommended to the TikTok accounts over the course of the experiment.
Figure 2. Graph charting the increase in overall toxic content recommended to the YouTube Shorts accounts over the course of the experiment.
Overall, the YouTube Shorts accounts were recommended a larger amount of toxic content (on average 61.5% of the total recommended content) compared to TikTok (34.7%). In the case of YouTube Shorts, the manosphere-curious accounts were recommended a significantly higher amount of toxic content (71.4%) than the gender-normative (generic) accounts (51.6%). However, for the TikTok accounts, relatively similar levels of toxic content were recommended to both the manosphere-curious accounts (32.5%) and the gender-normative (generic) accounts (36.6%).
While the YouTube Shorts accounts were recommended mostly manosphere content8 (56% of recommended content on YouTube Shorts compared to 21.1% of the total recommended content on TikTok), the TikTok accounts were recommended a more diverse range of toxic content, including reactionary right-wing content and conspiracy content (13.6% of recommended content on TikTok compared to 5.2% of the total recommended content on YouTube Shorts). The majority of this content on both platforms focused on anti-trans and US-centric political rhetoric. In the following section we unpack these results in more detail.
|
TikTok |
YouTube |
|
|
Alpha masculinity |
12.8% |
32.4% |
|
Anti-feminism/misogyny |
8.3% |
23.9% |
|
Reactionary right |
11.2% |
3.5% |
|
Conspiracy |
2.4% |
1.7% |
|
Known actors, generic content |
3.8% |
3% |
|
Total recommended toxic content |
34.7% |
61.5% |
All the accounts, both those which sought out manosphere content and those which sought out gender-normative (generic) content, were fed toxic content of some sort within the first 23 minutes of the experiment.
For the TikTok accounts, toxic content was recommended, on average, within the first 12 minutes of the experiment. On the manosphere-curious TikTok accounts, such content was recommended after 2 minutes 53 seconds and 22 minutes 1 second, respectively. On the gender-normative (generic) TikTok accounts, this happened after 7 min 6 seconds and 14 minutes 45 seconds, respectively. For the YouTube Shorts accounts, toxic content was recommended, on average, within the first 8 minutes of the experiment. On the manosphere-curious YouTube Shorts accounts, such content was recommended after 7 minutes 49 seconds and 1 minute 54 seconds, respectively. On the gender-normative (generic) YouTube Shorts accounts, this happened after 17 minutes 27 seconds and 2 minutes 1 second, respectively.
All the accounts, both those which sought out manosphere content and those which sought out gender-normative (generic) content, were fed manosphere content of some sort within the first 26 minutes of the experiment.
On the TikTok accounts, manosphere content was recommended on average after 15 minutes of viewing. On the manosphere-curious TikTok accounts, such content was recommended after 10 minutes 6 seconds and 25 minutes 4 seconds, respectively. On the gender-normative (generic) TikTok accounts, this happened after 8 min 49 seconds and 14 min 45 seconds, respectively. On the YouTube Shorts accounts, manosphere content was recommended on average after 8 minutes of viewing. On the manosphere-curious YouTube Shorts accounts, such content was recommended after 8 minutes 10 seconds and 1 minute 54 seconds, respectively. On the gender-normative (generic) YouTube Shorts accounts, this happened after 17 minutes 27 seconds and 2 minutes 1 second, respectively.
Notably, for two of the YouTube Short accounts, manosphere content (including Andrew Tate content) was recommended in the first round of the experiment (within the first 2 minutes) before any nudge had been applied. These accounts were the manosphere-curious 18-year-old account and the gender-normative (generic) 18-year-old account.
Importantly, once any of the experimental accounts showed an interest in watching toxic content, the amount of this content recommended rapidly increased. By the last round of the experiment, the vast majority of the content being recommended to the phones was problematic or toxic: on average, the TikTok accounts were being recommended 76% toxic content, while on average, the YouTube Short accounts were being recommended 78% toxic content. In other words, within 400 videos (i.e. 2 to 3 hours of viewing), the majority of recommended content was toxic in nature, primarily falling into the manosphere content category.
On the gender-normative (generic) TikTok accounts, by the last round of the experiment, these levels were 64% and 88%, respectively (see Figures 3 and 4). On the manosphere-curious TikTok accounts, by the last round of the experiment, the level of recommended toxic content was 58% and 76%, respectively (see Figures 5 and 6).
On the gender-normative (generic) YouTube Short accounts, by the last round of the experiment, these levels were 60% and 72%, respectively (see Figures 7 and 8). On the manosphere-curious YouTube Short accounts, by the last round of the experiment, the level of recommended toxic content was 78% and 88%, respectively (see Figures 9 and 10).
Figure 3. Graph charting the increase in various forms of toxic content recommended to the 16-year-old gender-normative (generic) TikTok account.
Figure 4. Graph charting the increase in various forms of toxic content recommended to the 18-year-old gender-normative (generic) TikTok account.
Figure 5. Graph charting the increase in various forms of toxic content recommended to the 16-year-old manosphere-curious TikTok account.
Figure 6. Graph charting the increase in various forms of toxic content recommended to the 18-year-old manosphere-curious TikTok account.
Figure 7. Graph charting the increase in various forms of toxic content recommended to the 16-year-old gender-normative (generic) YouTube Shorts account.
Figure 8. Graph charting the increase in various forms of toxic content recommended to the 18-year-old gender-normative (generic) YouTube Shorts account.
Figure 9. Graph charting the increase in various forms of toxic content recommended to the 16-year-old manosphere-curious YouTube Shorts account.
Figure 10. Graph charting the increase in various forms of toxic content recommended to the 18-year-old manosphere-curious YouTube Shorts account.
Manosphere content was extremely high among the recommended content types on both platforms. This content fell into two main categories: the first was alpha masculinity, i.e. content pushing rigid and prescriptive ideas of masculinity focused on dominance, physicality, and emotional stoicism. This type of content accounted for 12.7% of the content recommended on TikTok and 32.5% of the content recommended on YouTube Shorts. The second category of manosphere content was misogyny/anti-feminism, i.e. content which explicitly attacked or degraded women and gender-equality efforts or which promoted sexist or reductive stereotypes. This type of content accounted for 8.3% of the content recommended on TikTok and 23.9% on YouTube Shorts.
While, originally, the experiment set out to explore the prevalence of suggested manosphere content, many of the phones were also recommended reactionary right-wing content, i.e. content primarily focused on right-wing American punditry, with a focus on anti-trans and ‘anti-woke’ content. This accounted for 11.2% of the content recommended on TikTok and 3.5% of the content recommended on YouTube Shorts. In addition, both accounts were shown a small amount of conspiracy content, i.e. content asserting that users need to wake up to a hidden reality, often regarding large-scale governmental control. This type of content accounted for 2.4% of the content recommended on TikTok and 1.7% on YouTube Shorts. This was more prevalent on the TikTok accounts, which algorithmically determined that those interested in manosphere content may also be interested in right-wing and conspiracy content. By comparison, the YouTube Shorts accounts remained primarily focused on manosphere content, with an emphasis on alpha masculinity (see Figures 11 and 12 for a breakdown of the types of toxic content recommended across each platform).
Figure 11. Graph showing a breakdown of the amount of each type of toxic content recommended to the TikTok accounts, as well as the overall amount of toxic content recommended to each account. This demonstrates the mix of manosphere and right-wing content on the TikTok accounts.
Figure 12. Graph showing a breakdown of the amount of each type of toxic content recommended to the YouTube Shorts accounts, as well as the overall amount of toxic content recommended to each account. This demonstrates the dominance of manosphere content on the YouTube Shorts accounts.
An important finding of the experiment was the centrality of recurring actors in the recommended content. Most of these were well-known influencers, associated either with the manosphere or with reactionary right-wing punditry. Notably, the most frequently occurring actors were similar across both TikTok and YouTube Shorts. This finding demonstrates the centrality of key influencers in the current media ecosystem.
On the TikTok accounts, the most commonly featured actors were: Andrew Tate, Jordan Peterson, Donald Trump, Tristan Tate, Myron Gaines, Matt Walsh, Hannah Pearl Davis, Adrian Markovac, Piers Morgan and Ben Shapiro (see Figure 13).
Figure 13. Graph showing a breakdown of the top ten most featured actors on the TikTok accounts, including the number of videos each actor appeared in.
The appearance of Far Right influencers, such as Matt Walsh and Ben Shapiro, as well as political and media figures such as Donald Trump and Piers Morgan, illustrates the TikTok algorithm’s attempts to divert the user into manosphere-adjacent spaces such as the Far Right, the anti-trans lobby and strongman politics. By contrast, the YouTube Shorts influencers were more closely aligned with manosphere and anti-feminist agendas.
On the YouTube Shorts accounts, the most commonly featured actors were: Andrew Tate, Tristan Tate, Myron Gaines, Brian Atlas, Piers Morgan, Jordan Peterson, Adin Ross, Ben Shapiro, Hannah Pearl Davis, and Candace Owens (see Figure 14).
The top actors for both platforms overlap quite significantly, with Andrew Tate being the most heavily featured influencer on both TikTok and YouTube Shorts. However, Andrew Tate significantly dominated as the most featured influencer on the YouTube Shorts accounts, featuring 582 times. This is in comparison to featuring 93 times on TikTok (see Figures 13 and 14).
Figure 14. Graph showing a breakdown of the top ten most featured actors on the TikTok experimental accounts, including the number of videos each actor appeared in.
This section presents the findings of the qualitative analysis. Firstly, we discuss some key observations, which signal new and significant developments in the manosphere. We then analyse the most prominent themes, myths and talking points perpetuated by manosphere influencers, which we have divided into three sub-categories, namely crisis narratives, motivational scripts and gender ‘science’.
The most striking finding overall was the dominance of influencer-centred content, which accounted for the vast majority of videos in the dataset. Significantly, most of this material originated from regular users or micro-influencers reposting clips of influencers, rather than from the influencers’ own accounts. This demonstrates the extent of both user-led and algorithmic amplification of influencer content, and was especially evident in the case of Andrew Tate, who was by far the most recommended influencer on both platforms, despite the fact that his accounts were inactive at the time of data collection.
The dominance of these ‘ideological entrepreneurs’ (Jurg et al., 2023) marks a significant new shift in the manosphere, whereby male-supremacist influencers are not only accumulating considerable wealth but are also using the practices of influencer culture for metapolitical goals, a phenomenon noted by Maly (2020) in the context of the Far Right in recent years. By adopting less overtly gender-political rhetoric, and instead mobilising discourses around mental health, motivation and money-making, these influencers are strategically monetising men’s financial and emotional insecurities (Bujalka et al., 2022). This was evident in our dataset in the prevalence of content coded as ‘alpha masculinity’ (12.7% on TikTok and 32.5% on YouTube Shorts) over that coded as ‘anti-feminism / misogyny’ (8.3% on TikTok and 23.9% on YouTube Shorts). ‘Manfluencers’ have thus effectively replaced pick-up artists and the seduction industry with a significantly more lucrative, popular, and ostensibly well-intentioned venture, which purports to give men purpose, confidence and control.
Figure 17. Screenshots from videos recommended to the experimental accounts.
Linked to this new development has been the revival of the notion of stoicism, reappropriated from ancient Greek philosophy, whereby self-help influencers such as Ryan Holiday and Jordan Peterson advocate traditional masculine values of courage, self-discipline, and order as an antidote to the alleged chaos and narcissism of ‘woke’ modernity. Aleks Hammo (2023) refers to this phenomenon as the ‘stoic industrial complex’ and maintains that its appeal lies in its promise of taking control in an age of hyper-competition, secular disenchantment, and consumerism. However, this military-style neo-stoicism is underpinned by the repression of emotion, a return to strict gender roles and simplistic, individualistic accounts of complex social phenomena. It is also used to discredit the concept of structural or systemic disadvantage, and to reinforce the message that anyone can make it if they work hard enough. Many of the influencers in our dataset actively promote this message, urging men to ‘pull themselves up by the bootstraps’ and to try harder as ‘no one cares about the men who fail’. For example, an account reposting this type of content using the hashtags #motivation #mindset #advice #inspiration features retired US Navy SEAL David Goggins urging, ‘You need to f**king work harder, you need to f**king discipline your mind better…all the time you’re complaining, you could be instead hustling’.
Figure 18. Screenshots from videos recommended to the experimental accounts.
In tandem with this focus on neo-stoicism was another, ostensibly incongruous, theme which we termed ‘wealth porn’. This involves influencers bragging about their wealth and possessions, showing off their car collections and, in some cases, purchasing expensive cars and wrist watches on a whim. It also frequently involves disdain for poor people (‘brokies’) who are deluded to believe they can make money ‘inside the matrix’, which refers to mainstream, ‘blue-pilled’ or ‘normie’ society. Videos featuring Andrew Tate and Tristan Tate accounted for most of this content, and they frequently refer to the disciplinarian and frequently cruel parenting style of their father as responsible for their ‘sheer indefatigability and unmatched perspecacity’. Many of the Tate brothers’ motivational clips also double up as advertisements for Andrew Tate’s businesses, in particular the Real World, the War Room, and Hustlers University.
A significant development in the manosphere is the emergence of female anti-feminist influencers. While Eviane Leidig (2023) has documented this phenomenon in relation to Far Right propaganda, it has been less prevalent in the manosphere until recently. The most prominent female influencer in our dataset was Hannah Pearl Davis, who featured in the top 10 recommended influencers on both TikTok and YouTube Shorts. Pearl Davis, also known as JustPearlyThings, advocates for the traditional patriarchal family, and has claimed that divorce should be banned, women should not be allowed to vote, and the history of slavery has been embellished. Recurrent tropes in her discursive repertoire include that women are gold diggers, make false rape accusations and are ‘used goods’ (‘ran through ex-hoes’) if they fail to retain their virginity for marriage.
Although not exclusively an anti-feminist influencer, British podcaster and crypto investor Layah Heilpurn also featured regularly in our dataset. Heilpurn is a supporter of Andrew Tate and outspoken critic of ‘toxic feminism’ and femme-centric society. She has generated considerable wealth from cryptocurrency, yet says she doesn’t believe women should work or be in positions of leadership. Like most anti-progressive influencers on TikTok and YouTube Shorts, Heilpurn demonstrates a keen awareness of the key talking points of the Culture Wars and adopts a range of standard reactionary positions in opposition to feminism, ‘woke’ society, gender pronouns, and ‘cancel culture’.
Figure 19. Screenshots from videos recommended to the experimental accounts.
It was notable that on TikTok, Donal Trump-related content was frequently recommended to all four male-identified accounts. This was especially prevalent in the case of the 18-yearold gender-normative (generic) TikTok account, whose recommended content started with Andrew Tate wealth porn and hustle, but after approximately 45 minutes began to feature Trump as well as material related to the ‘stoic industrial complex’ (Hammo, 2023). By contrast, the 16-year-old gender-normative (generic) TikTok account began with manosphere and anti-feminist content focused largely on ‘the mating market’ but very abruptly switched to reactionary, anti-trans content after approximately one hour. The 16-year-old and 18-year-old manosphere-curious accounts were also recommended a significant amount of anti-trans content, most of which originated from British journalist Piers Morgan and, to a lesser extent, from Far Right influencers Matt Walsh and Ben Shapiro. Analysing recommended content from the perspective of ‘real’ users thus demonstrates the way in which anti-feminist, male supremacist and Far Right narratives merge together into a seamless and mutually compatible set of talking points, collectively underpinned by the logic of conspiracy. According to all these narratives, the liberal world order is a gynocentric regime designed to emasculate men, enslave them in poorly paid jobs, and destroy the traditional family. Escaping the ‘matrix’ promises to liberate men through promises of enlightenment, wealth accumulation, improved physical and mental health and the reinstatement of their power and privilege.
Figure 20. Screenshots from videos recommended to the experimental accounts.
Despite the prevalence of racism in many other corners of the manosphere, there was a notable absence of overtly anti-immigrant or racist content recommended to our experimental accounts. This may be attributable to the increasing ethnic and religious diversity of manosphere influencers, as well as a strategic move to expand their reach. Myron Gaines is of Sudanese descent, the Tate brothers’ father was African-American, and Andrew Tate and Sneako recently converted to Islam. The appeal of Islam to western men’s rights activists is complex and ostensibly contradictory. Figures like Tate and Sneako draw on stereotypes about Islam as patriarchal, stoic and intolerant of ‘woke’ agendas to challenge western liberalism, despite their own permissive lifestyles. Like US men’s rights activist Paul Elam, who points to Indian men’s successful challenge of the ‘dowry law’ as inspirational to men in the global north, they appear to be looking to more ‘traditional’ societies for ‘new’ norms, narratives and talking points.
Finally, the stylistic formats of Manfluencer culture signal a significant departure from the more obscure subcultural norms of platforms such as 4Chan and Reddit, which were characterised by geek techno libertarianism, meme culture and political rather than economic motives. By contrast, the talking heads and podcast talk show formats that underpin much of the manosphere content recommended on TikTok and YouTube Shorts are more commercial, have more mainstream appeal and are more outward facing. Of note here are Brian Atlas’ Whatever podcast and Myron Gaines’ Fit and Fresh podcast, which featured heavily in the recommended content on the two manosphere-curious YouTube Shorts accounts. In the vast majority of these video clips, an assortment of female influencers – both feminist and anti-feminist – as well as OnlyFans account holders, have their arguments ‘demolished’, ‘owned’ or ‘destroyed’ by their male hosts. Many other clips are edited by regular users to produce ‘gotcha’ sound bites such as ‘Matt Walsh exposes feminist’, ‘Myron Gaines destroys woke social justice warrior’ or ‘Non-binary person owned’.
Figure 21. Screenshots from videos recommended to the experimental accounts.
In this section, we categorise the most dominant ideas or talking points amplified by manosphere influencers and their followers into three key categories, namely 1) crisis narratives, 2) motivational scripts and 3) debunked gender ‘science’
The most prominent crisis narrative alleges that men and masculinity are under threat, allegedly due to feminism but also more broadly because of liberal government ‘brainwashing’ and ‘women-centric’ legal systems. The latter issue manifests primarily in the myth of false rape reporting, which drastically exaggerates the incidence of false rape claims, despite the international research consistently demonstrating that incidents of rape, sexual offences and child sexual abuse are significantly under-reported, under-prosecuted, and under-convicted. In more right-wing male supremacist forums, the crisis-of-masculinity narrative foregrounds white masculinity and coalesces with other conspiracy theories such as the Great Replacement, which purports that emasculated white men are being replaced by immigrants as part of a Jewish-controlled ‘globalist’ agenda. However, this more extreme racist version was not present in our sample.
Linked to this is the concept that the heteropatriarchal nuclear family is under threat, again allegedly due to feminism’s sexual liberation of women, which has afforded them sexual and reproductive choices, resulting in promiscuity, childlessness, divorce, and unreasonable standards expected of men. A frequent trope related to the decline of the family is that of fatherlessness and single mothers, whereby it is claimed that boys raised without fathers are emasculated and more likely to be rapists, addicts, and criminals, while girls raised without fathers are more likely to be promiscuous. The family is also considered to be threatened by LGBT rights, gender fluidity and transgender rights (often referred to as the ‘trans cult’). This myth relies heavily on a range of transphobic slurs about transwomen, namely that they are paedophiles, groomers, mentally ill and/or ‘autogynephiles’. The neo-manosphere is also characterised by a sustained campaign of ridiculing gender non-binarism and pronouns: typical excerpts from videos recommended in our dataset included ‘Blue-lipped lib is trigender’, ‘The left wing gender insanity being pushed on our children is an act of child abuse’ (Trump) and ‘I’m not the one who started sending men into the women’s bathroom and taking away their trophies and castrating kids’ (Michael Knowles).
These various tropes combine to produce a meta crisis narrative, namely that society has been broken by feminism and liberalism, which in turn provides a rationale and creates a market for a raft of monetised self-help advice.
Motivational scripts fell into 3 main subcategories, namely money, motivation/ mental health and seduction. The most prevalent of these to be recommended to our accounts was money, with influencers either offering direct advice on how to get rich or bragging about their riches in a phenomenon we have named ‘wealth porn’. These iterations of capitalist masculinity are a clear attempt to reconnect male identity with economic status, a relationship which has been substantially disrupted by neoliberal capitalism and its attendant erosion of salaries, career stability, and the social safety net. They thus strategically exploit male insecurities about unemployment and poverty, and are based on a series of fallacies, namely ‘if you work hard, you’ll make it’, ‘if you’re poor (a ‘brokie’) it’s your own fault’, and school education and university degrees are a waste of time. Hashtags such as #getrich and #escapethematrix promise disillusioned and economically unstable young men a way out of their current predicament as well as a sense of rebellion and ‘sticking it to the man’
Importantly, the mindset required to get rich can only be achieved ‘outside the matrix’, i.e. with the help of these ‘maverick’ influencers and their various commercial schemes. This mindset is heavily underpinned by neo-stoicism, the flipside of which is the repression of emotion and the rejection of modern ‘therapy culture’. This rhetoric of hard work and choosing the ‘difficult path’ also provides the illusion of masculine recuperation in a culture where media narratives of masculinity-in-crisis and emasculation are common. Andrew Tate states, ‘I have no sympathy for people who aren’t difficult on themselves’, despite the obvious contradictions that exist between genuine stoicism and the type of wealth porn, hustle, and get-rich-quick schemes he champions. Paradoxically, Tate’s liberation from ‘the matrix’ and considerable wealth have not enabled him to work less. He claims to work around the clock, and considers everything he does part of hustle, including having sex with women.
The manosphere’s revival of stoicism also segues neatly into a particular response to the mental health crisis among men. Rather than acknowledging that the pressure to conform to hetero-patriarchal norms is harmful to men, the manosphere construes male suffering as attributable to feminism, progressives and attempts to make men more like women. In turn, it offers a range of highly individualistic solutions to this problem, which centre around self-discipline, physical self-improvement through gym workouts and diet, and the repression of emotion.
‘How can I feel depression when I’ve smashed and destroyed 68 people’s faces in front of me. Men who thought they could test me in fair combat. How can I feel depressed? It’s impossible.’
Andrew Tate
The latter is one of the most dangerous facets of the neo-manosphere, as a broad body of international research has shown that emotional repression and lack of emotionally supportive friendship networks are key factors in depression and suicide among men (Cleary, 2019). However, manosphere influencers use pseudo-scientific ideas derived from evolutionary psychology to claim that men and women experience (rather than express) emotion differently, which lends legitimacy to their worldview. Their rhetoric consistently frames depression as weak, or the result of laziness and lack of motivation. According to Andrew Tate, depression isn’t real, and mental illness makes men weak: ‘If you are the kind of person who feels like they need therapy, you need someone to talk to, to make me feel better - you know what you are, you’re useless.’ Tate also says, ‘I don’t care how I feel. I don’t care if I feel happy or sad. It doesn’t really affect what I do each day. I do the exact same things. I act the exact same way. I don’t care. I don’t put weight to the significance of the emotion.’ Followers are urged instead to ‘Find the beast within you. Throw yourself into pain’, and Tate claims he has cured many people of depression who could not be cured by their doctors.
Meanwhile, self-styled British mental health guru Adrian Markovac reiterates Jordan Peterson’s plea to tidy your room, and rails against the school system, encouraging young people to follow his life hacks instead, which include renouncing medication for depression.
Like others, Markovac mobilises a decline of civilisation narrative, and attributes this decline to female promiscuity and divorce, rather than neoliberal capitalism and its deleterious effects on the climate and ecosystem, worker’s rights, and human well-being. Markovac and Tate avoid anti-immigration rhetoric, focusing instead on more generic crisis narratives about the family, the system, and men as broken. At the heart of the manosphere’s approach to mental health is the concept that men and women are emotionally hard-wired in different ways, and that boys do not need to express their feelings or seek therapeutic support. In a video recommended to one of our TikTok accounts, English podcaster and YouTuber Chris Williamson asserts that, ‘Male depression is treated like female depression. Men are made to feel loved and accepted when all they want to do is feel capable and powerful…Give a man a purpose and the ability to achieve it and he will crawl over broken glass with a smile.’
In short, the manosphere purports to care about men’s mental health while actively reproducing norms that hurt them. Significantly, in our study, the search term ‘mental health’ resulted immediately in manosphere and anti-feminist influencer content being recommended to the account (e.g. Hannah Pearl Davis, Ben Shapiro, Matt Walsh and Markus Rogers). Perhaps even more worryingly, some qualified mental health practitioners and counsellors who work with boys and young men are replicating these ideas in their practice. This positioning of boys and men as victims of a woman-centric society in which, it is claimed, we are treating boys as ‘defective girls’, reinforces a war-of-the-sexes logic and implies that gains for women necessarily entail losses for men. It also revives scientifically debunked theories about gender difference, which have been used to straightjacket women and men into restrictive and unequal roles and identities. Any interventions which attempt to mitigate against the pull of manosphere influencers, therefore, will need to negotiate this issue of men’s mental health with nuance, empathy, and evidence-based research if they are to succeed.
Theories about sex and gender derived from evolutionary psychology underpin all of the manosphere’s claims. Key concepts in this regard are hypergamy and the 80/20 rule, according to which women are genetically programmed to seek out alpha males for optimum reproductive success but, due to scarcity (as only 20% of males are alpha), the majority must settle for beta males. In the past, people generally accepted this situation, and most women remained virgins in order to be marriageable. However, it is alleged that feminism and the sexual revolution have disrupted this power dynamic by allowing women to have sexual choice and freedom. Thus, both biology and changing sexual norms mean that women now have much higher standards and expectations, leading to large-scale sexual rejection and insecurity among men. It is also claimed that this makes women more likely to cheat, as they are biologically programmed to ‘monkey branch’ (look for a more superior mate) and are therefore innately disloyal. Only alpha males have sexual choices, while everyone else must compete for the ‘spoils’ (promiscuous women who are now looking for a beta to pay the bills).
‘Back then, an average guy could get a traditional woman. And now what does he get? He gets a boss bitch, boss babe, a girl that’s been ran through. He gets an ex hoe, he gets a born again virgin’
Hannah Pearl Davis
Much like the pre-influencer manosphere, the new neo-masculinists assert that men are more oppressed than women due to war, divorce, dangerous occupations, and the pressure to provide. Gender relations (unapologetically heterosexist and binary) are framed as a zero-sum equation, where any form of female empowerment necessarily leads to male disempowerment. According to most of the influencers in our dataset, feminism has upset the natural order, made women unhappy and destroyed the family. Children raised by working and single mothers are alleged to be dysfunctional, and women have gained nothing by replacing their husband with their male boss. In a video recommended to our 16-year-old manosphere-curious YouTube Shorts account, Myron Gaines claims ‘Women who prioritise career over marriage…They serve their male boss but not their husband.’
Behind this fake science is a clear campaign to return women to the home, through strategies of sexual shaming. In multiple videos recommended to our accounts, women are defined by ‘body count’, i.e. the number of men they have slept with, and are referred to as hoes, bitches, sluts and 304 (which spells hoe on a calculator upside down). The age-old virgin-whore dichotomy was strongly evident in our dataset, with various influencers negotiating the tension between the necessity for a virgin wife on the one hand and the desire to have sex with multiple women on the other. In a video recommended to the 18-year-old manosphere-curious TikTok account, American influencer Chase (Sovereign Brah) advises, ‘If a woman loves the Lord and she subscribes to biblical values, she will understand that it is the man’s job to lead the family, to provide for the family, and to lead the family under the headship of Christ. And it’s the woman’s job to submit to that man’s headship in the family.’ In a video recommended to the manosphere-curious 16-year-old TikTok account, Australian sex coach Stirling Cooper attempts to resolve this dilemma by advising, ‘Guys don’t want a woman to be a slut. Guys want a woman to be his slut’.
Andrew Tate, by contrast, does not extol the virtues of monogamy or marriage, claiming ‘I don’t see the tactical advantage in getting married. If I put my d**k in another bitch, my wife can go to a lawyer with proof. I don’t want the government watching my d**k’. However, he claims to have fathered and to provide for 12 children with different women and offers regular advice about the need to subjugate women through belittling them (a technique referred to as ‘negging’) as it is believed that women are naturally attracted to bad, powerful men. This phenomenon is referred to as the ‘dark triad’ personality, and features regularly in manosphere, Red Pill and incel discussions. It refers to a trio of personality traits - narcissism, Machiavellianism, and psychopathy – to which, it is believed, women are naturally attracted.
In brief, the manosphere frames heterosexual relationships and sex as transactional, and places strong emphasis on the attractiveness of (sexually) submissive women. It encourages men to view sexually agentic and liberated women as promiscuous, and to shame them publicly for this. Many influencers even argue that women who are in a relationship should not go out socially with male or female friends, and that it is preferable if your girlfriend does not have a social media account. While, to most people, these ideas may appear outlandish and extreme, they have gained cultural traction over time through processes of amplification and reiteration. Already, hitherto obscure terms such as normie, Chad, Stacy, sigma male, friend-zoning, the Red Pill and 304 have become part of mainstream vernacular (see Appendix 2: Glossary of Terms)
Push factors
It is important to recognise that patriarchy, sexism, and misogyny are not new. They have been features of many societies for several thousand years. The manosphere is merely the latest version of attempts to prohibit women from participating in the political and economic order (Siapera, 2019). Due to technology, however, it has been possible to spread anti-progressive messages and ideas more widely and rapidly than ever before.
This report focuses almost exclusively on the pull factors, in other words the ways in which social media algorithms work to entice users into harmful and increasingly toxic content. It is, however, important to acknowledge that these work in conjunction with push factors, and that not all boys exposed to manosphere influencers will be interested in or enticed by them.
Push factors are the contextual elements that predispose boys and men to become radicalised by manosphere (or other) ideologies. They are necessarily complex as they involve unique combinations of social, economic, psychological, and cultural determinants. While it is impossible to capture all of them, the following considerations may explain why some boys are more easily influenced by the manosphere than others:
Firstly, it is unsurprising that many young people feel alienated, angry and anxious given the severity of the climate crisis, rising property prices and the instability of the employment market. However, because traditional masculinity is built around economic status and property ownership, this frequently impacts men’s sense of their gender identity more severely. Young men who find themselves excluded from the privileges afforded to their fathers and grandfathers may experience a sense of ‘aggrieved entitlement’ (Kimmel, 2017).
Secondly, in most modern societies, male power and privilege are no longer taken-for-granted norms. In this sense, masculinity and maleness-as-neutral have been rendered visible and decentered. Some economically vulnerable men feel threatened by these changes and may be especially susceptible to the messages from the Far Right about immigrants ‘taking their jobs, houses and women’. However, many middle-income and wealthy men are also strongly resistant to these changes, as they believe their power and privilege are being displaced.
The increasing normalisation of Lad Culture since the late 1990s has made a range of harmful behaviours more acceptable: excessive alcohol and drug use, sexist and homophobic ‘banter’, sexual conquest as competition and the repression of emotion are not only destructive to others but also to boys and men themselves. Boys and men involved in laddish behaviours and friendship groups are less likely to talk about their feelings or to seek help from friends or mental professionals, as depression and the expression of emotion are seen as a form of weakness.
Boys who are socially isolated, have a history of being bullied, and/or spend a lot of time online may be more susceptible to involvement in the manosphere. In particular, sexually unsuccessful boys and those with depression or other mental health issues may be drawn to the incel (involuntary celibate) community (Speckhard and Ellenberg, 2022). However, the incel community remains quite distinct from mainstream Manfluencer culture. Although both agree on many issues, incels are generally disillusioned with and sceptical of influencers’ promises of wealth and sexual success.
Conclusions
By tracking and coding recommended content over time, this study demonstrates how TikTok’s and YouTube Shorts’ algorithms promote toxic content to boys and young men using the platforms. As the study progressed, each account was recommended an increasing amount of manosphere content, with most messages promoting rigid and harmful masculine norms, misogyny and antagonism towards women and minorities, and spurious advice on mental health and wealth accumulation.
The most important overall finding, therefore, is that the manosphere and its ‘concocted gender war’ (Haslop et al., 2024) have become increasingly mainstreamed. Pick-up artists have been largely replaced by influencer culture, which extends far beyond seduction advice, and has been especially adept at tapping into gendered anxieties about mental health and economic instability. This has engendered a new focus on motivation, stoicism, and wealth accumulation.
The concerns and talking points of the manosphere – insofar as they appear in our dataset - have become broader, to incorporate a selective range of Far Right concerns around the nuclear family, anti-trans inclusivity, anti-government, and anti-welfare state. The inclusion of women and female influencers into this space is an important aspect of the mainstreaming and normalisation process. There is also considerable evidence of support for US reactionary right political punditry, although most manfluencers steer clear of overtly racist and anti-immigrant rhetoric.
Shutting down influencers’ accounts does not necessarily remove their content. The overwhelming presence of Andrew Tate content in our dataset at a time when he was de-platformed strongly supports this and indicates that social media companies must tackle harmful content in more sophisticated ways.
The findings of this report point to urgent and concerning issues for parents, teachers, policy makers, and society as a whole. Teachers in Ireland and elsewhere are reporting significant disruption in their classrooms, with some boys subjecting girls, female teachers, and LGBTQ students to abuse, and citing various anti-feminist and manosphere talking points. In particular, our findings highlight the ineffectiveness of social media platforms in protecting children and young people. Ultimately, girls and women are the most severely impacted by these beliefs, but they are also damaging to the boys and men who consume them, especially in relation to mental wellbeing. We hope our findings will compel the social media companies, government, and policy makers to take urgent action.
Recommendations
Social media companies need to implement stricter content moderation and enforce harsher sanctions: influencers who advocate discrimination or violence against women and other groups should be de-platformed, de-monetised and/or de-amplified.
While de-platforming high-profile influencers may help to de-amplify toxic content, the role of micro-influencers and regular users in spreading manosphere content cannot be overlooked. The proliferation of Andrew Tate content in our dataset at a time when he was de-platformed demonstrates that content moderation needs to be both content- and account-focused.
Social media companies should work closely with Coimisiún na Meán (Ireland’s new media regulator) and trusted flaggers to highlight illegal, harmful, and borderline content.
Algorithmic architectures need radical rethinking to avoid promoting harmful content to young people. As per the recommendation of the Irish Council of Civil Liberties (ICCL) and 60 other organisations9, recommender algorithms should be turned off by default.
Safety-by-design principles should be embedded in product development. This should involve collaboration with diverse experts in gender-based abuse. Our study demonstrates that the content recommended to ‘average’ or hetero-normative boys is not appropriate for a broad audience. Proactive reduction of borderline or harmful content must prioritise sexism and misogyny.
Social media companies should publish their content moderation decision-making criteria when borderline, harmful or illegal content is reported. They should also be transparent about what they do to remove, de-amplify or reduce this content, especially allegedly banned content that is being recycled in large quantities by micro influencers and regular users.
In line with the Digital Services Act (DSA), platforms should provide researchers with transparent details on algorithmic designs and policy, as well as access to data, which would better enable them to assess personalisation-driven harms.
The anger, confusion, and anxiety that some boys are experiencing should not be dismissed or shut down. Their opinions need to be heard and discussed in a safe, non-judgemental context.
Teachers should not focus on celebrity influencers or express disgust or outrage at their ideas. Instead, they should listen to and discuss why students are attracted to them. The reasons may be complex and varied (e.g. to gain peer approval or to rebel against teachers and parents, rather than genuine acceptance of the ideas).
It is essential to create a safe space in which problematic or misguided ideas can be constructively challenged, by teachers, other students, and older peers (Ging et al, 2024). Peer-to-peer learning approaches and positive male role models should be emphasised to promote an educative rather than punitive response to boys’ behaviours (Regehr et al., 2024).
Most teachers need training to address this issue. Organisations such as Beyond Equality and Hope Not Hate (see Resources) provide teacher training workshops and resources for school staff and educators on how to help boys negotiate gender stereotypes, peer pressure, insecurity, fear of rejection and societal expectations.
Most young people, while tech-savvy, need better critical digital literacy skills to understand the political economy of influencer culture and the algorithmic architectures of social media platforms. Schools need to invest in teacher training in this area or to outsource to experts.
Parents should encourage and allow open discussions without fear of rebuttal.
Parents should not focus on celebrity influencers or express disgust or outrage at their ideas. Instead, they should listen to and discuss why their child is attracted to them. The reasons may be complex and varied (e.g. to gain peer approval or to rebel against teachers and parents, rather than genuine acceptance of the ideas)
The manosphere insists on rigid understandings of male and female identity and relationships. Ask young people to consider the real people in their lives, with a view to highlighting the diversity of gender roles, identities, and behaviours in reality.
Encourage engagement with relatable resources - there are lots of YouTube videos and podcasts which discuss and debunk the manosphere’s key ideas in a thoughtful and rational way (see Resources).
Useful Resources
Positive Masculinity Interventions
Webwise webwise.ie
The Positive Masc Project positivmasc.ki.se
Beyond Equality beyondequality.org
Hope Not Hate hopenothate.org.uk/communities/in-schools/
BBC Trending: How to Exit the Manosphere bbc.co.uk/programmes/w3ct5d95
Statistics
European Institute for Gender Equality: Gender Equality Index, Ireland (2022)
eige.europa.eu/modules/custom/eige_gei/app/content/downloads/factsheets/IE_2022_ factsheet.pdf
Rape Crisis Network Ireland (RCNI) 2022 Statistics
rcni.ie/wp-content/uploads/RCNI-Rape-Crisis-Statistics-2022.pdf
Podcasts
Who is Andrew Tate? The Journal.ie Explainer
podcasts.apple.com/ie/podcast/who-is-andrew-tate/id1452246930?i=1000595850526
Now and Men podcast: Men, Masculinities and Gender Equality
menengage.org/resources/now-and-men-podcast-men-masculinities-and-gender-equality
References
Amnesty International. (2023). Driven into Darkness: How TikTok’s ‘For You’ Feed Encourages Self-Harm and Suicidal Ideation. https://www.amnesty.org/en/ documents/pol40/7350/2023/en/. Accessed: 05 March 2024.
Amnesty International. (2023). “I Feel Exposed”: Caught in TikTok’s Surveillance Web. https://www.amnesty.org/en/documents/pol40/7349/2023/en/. Accessed: 05 March 2024.
Brace, L., Baele, S. J., & Ging, D. (2023). Where do ‘mixed, unclear, and unstable’ ideologies come from? A data-driven answer centred on the incelosphere. Journal of Policing, Intelligence and Counter Terrorism, 1-22. https://doi.org/10.1080/18335330. 2023.2226667
Bujalka, E., Rich, T., & Bender, S. (2022). The Manosphere as an Online Protection Racket: How the Red Pill Monetizes Male Need for Security in Modern Society. Fast Capitalism, 19(1), 1–16. https://doi. org/10.32855/fcapital.202201.001
Centre for Countering Digital Hate. (2022). The Incelosphere: Exposing pathways into incel communities and the harms they pose to women and children. https://counterhate.com/wp-content/ uploads/2023/08/CCDH-The-Incelosphere-FINAL. pdf. Accessed: 04 March 2024.
Cleary, A. (2019). The Gendered Landscape of Suicide: Masculinities, Emotions, and Culture. Springer.
Das, S. (2022, August 6). Inside the violent, misogynistic world of TikTok’s new star, Andrew Tate. The Guardian.
Ging, D. (2019). Alphas, Betas, and Incels: Theorizing the Masculinities of the Manosphere. Men and Masculinities, 22(4), 638–657. https://doi.org/10.1177/1097184X17706401
Ging, D. and Castellini da Silva, R. (2022) Young People’s Experiences of Sexual and Gender-based Harassment and Abuse During the Covid-19 Pandemic in Ireland: Incidence, Intervention and Recommendations. DCU National Anti-Bullying Centre. https://antibullyingcentre.ie/wp-content/ uploads/2022/10/Young-Peoples-Experiences.pdf. Accessed: 04 March 2024.
Ging, D., & Murphy, S. (2021). Tracking the pilling pipeline: limitations, challenges And a call for new methodological frameworks in incel And manosphere research. AoIR Selected Papers of Internet Research, 2021. https://doi.org/https://doi.org/10.5210/spir. v2021i0.12174
Ging, D., Ringrose, J., Milne, B., Horeck, T., Mendes, K., & Castellini da Silva, R. (2024). Moving beyond masculine defensiveness and anxiety in the classroom: exploring gendered responses to sexual and gender based violence workshops in England and Ireland. Gender and Education, 1–18. https://doi.org/10.1080/ 09540253.2024.2315052
Hammo, A. (2023, May). Hypermasculinity and the Stoic Industrial Complex. Kill Your Darlings. https://www.killyourdarlings.com.au/article/ hyper-masculinity-and-the-stoic-industrial-complex/. Accessed: 06 March 2024.
Haslop, C., Ringrose, J., Cambazoglu, I., & Milne, B. (2024). Mainstreaming the Manosphere’s Misogyny Through Affective Homosocial Currencies: Exploring How Teen Boys Navigate the Andrew Tate Effect. Social Media + Society, 10(1). https://doi.org/10.1177/20563051241228811
Internet Matters. (2023). “It’s really easy to go down that path”: Young people’s experiences of online misogyny and image-based abuse. https://www. internetmatters.org/wp-content/uploads/2023/09/ Internet-Matters-Online-misogyny-and-image-based-abuse-report-Sep-2023-2.pdf. Accessed: 06 March 2024.
Iqbal, Mansoor (2024a). TikTok revenue and usage statistics (2023). Business of Apps. https://www. businessofapps.com/data/tik-tok-statistics/. Accessed: 09 January 2024.
Iqbal, Mansoor (2024b). YouTube Revenue and Usage Statistics (2024). Business of Apps. https://www. businessofapps.com/data/tik-tok-statistics/. Accessed: 09 January 2024.
Jurg, D. H. M., Schlüter, M., & Tuters, M. (2023). Cult Leaders, Masculinity, and White Supremacy on YouTube: The Case of Stefan Molyneux. Digital Methods Initiative. https://wiki.digitalmethods.net/ Dmi/CultLeaders
Kimmel, M. (2017). Angry White Men: American Masculinity at the End of an Era (2nd ed.). Nation Books.
Leidig, E. (2023). The Women of the Far Right: Social Media Influencers and Online Radicalization. Columbia University Press. https://doi.org/https://doi. org/10.7312/leid21016
Maly, I. (2020). Metapolitical new right influencers: The case of Brittany Pettibone. Social Sciences, 9(7). https://doi.org/10.3390/SOCSCI9070113
Regehr, K., Shaughnessy, C., Zhao, M., & Shaughnessy, N. (2024). Safer scrolling: How algorithms popularise and gamify online hate and misogyny for young people. UCL and University of Kent https://www.ascl.org.uk/ASCL/media/ASCL/ Help%20and%20advice/Inclusion/Safer-scrolling.pdf. Accessed: 06 March 2024.
Ribeiro, M. H., Blackburn, J., Bradlyn, B., de Cristofaro, E., Stringhini, G., Long, S., Greenberg, S., & Zannettou, S. (2021). The Evolution of the Manosphere Across the Web. Proceedings of the International AAAI Conference on Web and Social Media, 196–207. https://doi.org/10.5281/zenodo.4007913
Ringrose, J., Regehr, K. and Milne, B. (2021) Understanding and combatting youth experiences of image-based sexual harassment and abuse. Department of Education, Practice and Society, UCL Institute of Education. https://discovery.ucl.ac.uk/id/eprint/10139669/
Siapera, E. (2019). Online Misogyny as Witch Hunt: Primitive Accumulation in the Age of Techno-capitalism. In Gender Hate Online (pp. 21–43). Springer International Publishing. https://doi.org/10.1007/978-3-319-96226-9_2
Speckhard, A., & Ellenberg, M. (2022). Self-reported psychiatric disorder and perceived psychological symptom rates among involuntary celibates (incels) and their perceptions of mental health treatment. Behavioral Sciences of Terrorism and Political Aggression. https://doi.org/10.1080/19434472.2022. 2029933
Vogels, E. (2021). The state of online harassment. https://www.pewresearch.org/internet/2021/01/13/ the-state-of-online-harassment/
Walsh, M. (2021). “Storm and Stress” An Exploration of Sexual Harassment Amongst Adolescents. https://www.rcni.ie/wp-content/uploads/RCNIBreaking-the-Silence-1.pdf. Accessed: 06 March 2024.
Women’s Aid. (2020). One in Five: Young Women Suffer Intimate Relationship Abuse in Ireland. https://www.womensaid.ie/app/uploads/2023/04/ One-in-Five-Youn-Women-Report-2020.pdf. Accessed: 06 March 2024.
1 Reset Australia (au.reset.tech/) is a not-for-profit charity which specialises in research methods on digital risks and online harms.
2 Data visualisation was funded by DCU Faculty of Humanities and Social Sciences Staff Journal Publication Scheme
3 These averages are higher (2 hours: 54 minutes for TikTok and 3 hours: 48 minutes for YouTube Shorts) if we remove the control accounts, for which the engagement times were comparatively shorter (1 hour: 6 minutes and 1 hour: 33 minutes, respectively). This is because these control accounts did not actively engage with any content, making the overall watch time comparatively shorter.
4 Toxic content was defined as all coded content, excluding the category ‘Known actor generic content’ (for coding breakdown see Methods section and Appendix 1).
5 ounews.co/around-ou/university-news/ou-research-reveals-shocking-level-of-online-violence-experienced-by-women-and-girlsacross-the-uk/
6 See au.reset.tech/news/algorithms-as-a-weapon-against-women-how-youtube-lures-boys-and-young-men-into-the-manosphere/
7 The generic male accounts do not purport to represent all boys or men but rather those with more gender-normative interests.
8 Manosphere content is defined as content falling into either the alpha masculinity or misogyny/anti-feminism coding categories.
9 https://www.iccl.ie/news/62-organisations-urge-strong-action-by-coimisi…
Appendices
Appendix 1: Coding Scheme
Each video recorded on all ten phones/accounts was coded using the Excel spreadsheet below (400 videos per phone):
Each of the 3 researchers conducted 4 pilot coding rounds (200 videos), enabling us to develop a pilot codebook. We then tested this on a further 200 videos each to ensure inter-coder reliability. Below is the final topic / thematic code book used:
- Alpha masculinity: any content which promotes rigid and prescriptive ideas of masculinity focused on dominance, physicality, aggression, discipline, wealth or emotional stoicism.
- Misogyny / anti-feminism / sexism: any content which degrades or is derogatory or dehumanising toward women; or which stereotypes women (e.g. as narcissists, gold diggers, unintelligent, submissive, etc.); or which is opposed to gender equality or feminism.
- Reactionary right: any content which promotes typical Far Right and anti-progressive ideologies (e.g. racism, anti-immigration, anti-LGBTQ, anti-trans, etc.)
- Conspiracy: any content which references popular conspiracies relating to e.g. vaccination, climate change, Great Replacement, globalism, conspiracies to harm or disempower Andrew Tate, etc.
- Known actor / generic content: content produced by a known manosphere actor which is not toxic or does not fit into the above categories, e.g. Jordan Peterson talking about eating only red meat.
Below is typical excerpt from one of the coding sheets:
Appendix 2: Glossary of Terms
|
Awalt |
Abbreviation of ‘All women are like that’ (i.e. vapid, vain, shallow, promiscuous, emotional, irrational, and motivated by financial gain) |
Gigachad |
The most alpha of alpha males. Gigachad is a caricature of hypermasculinity and, like Chad and Stacey, features in numerous memes. |
Sigma male |
A hypermasculine male, like the alpha male but is less concerned with social norms, and is considered to be a ‘lone wolf’. |
|
Alpha male |
Belonging to the most socially dominant and sexually successful group of men in the male hierarchy. According to evolutionary psychology, only 20% of males are alpha. |
Hypergamy |
Theory derived from evolutionary psychology that all women try to ‘marry up’ by seeking out alpha males to optimise their genetic reproductive opportunities. |
Soyboy |
A pejorative term used to describe men lacking in masculine qualities, frequently used against social justice warriors, vegans, pro-feminist men and liberals. |
|
Beta male |
Weaker, less attractive and less sexually successful males who are believed to account for 80% of men. Because of the scarcity of alphas, most women must settle for betas. |
Incels |
Involuntary celibates. Men who attribute their lack of sexual success with women to their lack of physical attractiveness. |
Stacy |
A sexually attractive, sexually successful woman. The female counterpart to Chad, she is depicted as shallow, promiscuous, and unintelligent. |
|
Chad |
The ultimate alpha male. Chad is hyper-masculine, virile, powerful, and sexually attractive to Stacys (attractive women). |
MGTOW |
Men Going Their Own Way. A subgroup of the manosphere which advocates male separatism, either by refraining from long-term committed relationships with women or by having nothing to do with women. |
The Red Pill |
Also TRP. The unifying ‘philosophy’ of the manosphere, appropriated from the 1999 film The Matrix. To be redpilled is to be enlightened to the ‘fact’ that society is a gynocentric conspiracy which disadvantages men. |
|
Cuck |
Short for cuckold, refers to a man whose wife has been unfaithful. Generally used as an insult to describe someone who is weak or emasculated. |
MRA |
Men’s rights activist |
THOT |
Abbreviation for ‘that hoe over there’. Term used to describe women. |
|
Foid |
Abbreviation of femoid, a term used to describe women. |
Normie |
A normal or mainstream person, who is not part of the incel or manosphere subculture. Normies are generally considered to be bluepilled (i.e. they are unenlightened / have not been redpilled). |
White knight |
A man who tries to curry favour with women by defending them or supporting their causes. It is assumed that men’s reason for doing this must be sexual. Other similar terms are ‘simp’ or ‘whipped’. |
|
Friend-zone |
Friendship in which one person, typically male, is sexually attracted to a woman, who prefers to be friends. He is said to be friend-zoned. |
PUA |
Pick-up artist. A seduction expert who profits from selling men dating advice. Most of this advice is based on concepts from evolutionary psychology, and assumes that women are naturally attracted to powerful, dominant men. |
304 |
Term used to describe a ‘promiscuous’ woman. 304 spells ‘hoe’ upside down on a calculator. |