Non-consensual intimate imagery (NCII) abuse describes the creation and dissemination of intimate images of a person without their consent and has become a significant and growing form of technology-facilitated gender-based violence (TFGBV). Advances in generative artificial intelligence (GenAI) have enabled the production of highly realistic imagery with minimal technical expertise, expanding the accessibility of NCII content. As a result, NCII abuse has soared.  

This Explainer outlines key definitions, examines NCII’s prevalence and gendered nature and explores different forms of NCII, including synthetic NCII (NSII). This Explainer also discusses NCII’s impact on victim-survivors, highlighting NCII as a form of gender-based violence (GBV) embedded in broader structural inequalities. Finally, this Explainer provides an overview of legislative, technological and societal responses to this form of violence. 

What is NCII? 

NCII refers to intimate images or videos of a person that are created, distributed or threatened to be disseminated without their consent. Images that were edited, altered or otherwise  manipulated also fall under the umbrella of NCII abuse.  Although public attention has often focused on high-profile cases, NCII abuse is not isolated to public figures. Rather, this phenomenon reflects broader patterns of gendered violence embedded in digital environments. As demonstrated by ISD research, perpetrator motivations for creating and sharing NCII extend beyond sexual gratification to include public humiliation and degradation, as well as the exertion of control, coercion and power more broadly. 

NCII is often used interchangeably with the term Image-Based Sexual Abuse (IBSA). However, NCII also encompasses images that depict nudity, partial nudity or otherwise private situations that are not sexual or sexualised. This can include a wide range of intrusive practices, such as non-consensually taking photos up a person’s skirt (‘upskirting’) or into a person’s clothing (‘downblousing’), as well as recording individuals in inherently private contexts, like bathrooms or changing rooms. As definitions of NCII often fail to account for differences in what is considered ‘intimate’ given distinct cultural or religious contexts, some experts advocate for broader conceptualisations of this abuse that centre the individuals’ right to define and control the boundaries of their own intimate lives. 

‘Consent’ refers to a person’s informed, voluntary and specific agreement to an act, and it should be understood as an ongoing process rather than a one-time occurrence. Consent can be revoked at any time. In turn, the term non-consensual covers any situation where such permission is absent, coerced or given for a different context or purpose. For example, an intimate image that was taken with consent in a specific setting can constitute NCII if it is later shared without consent, as the original agreement only applied to that particular use. This also applies to situations where the content was stolen (e.g. obtained through hacking).  

Many authorities subsume this form of NCII—in which intimate content was initially shared consensually but further disseminated without consent—under so-called ‘revenge porn’ offences. However, this term is widely criticised because it mischaracterises and minimises the nature of the harm, implies revenge as a necessary motive and shifts attention away from issues of consent and structural GBV. The term risks framing the perpetrator’s actions as an understandable or emotionally motivated response, thereby implicitly directing empathy toward the perpetrator and obscuring the coercive and abusive nature their actions. Finally, coerced consent is generally invalid, as any agreement produced under pressure cannot be considered free or voluntary. This element is particularly relevant in the context of ‘sextortion’, which involves coercing someone into providing intimate images and/or threatening to share intimate content to force compliance. 

The prevalence of NCII 

The literature on NCII reports varying statistics regarding the prevalence of NCII. Due to shame, stigma and pervasive victim-blaming narratives, many affected individuals do not report their experiences. Therefore, the absolute number of cases is almost certainly higher than official statistics suggest. 

A recent large-scale representative study on the prevalence of NCII found that one in four individuals experienced this form of abuse at least once. LGBTQ+ persons reported significantly higher rates of victimisation (38.5 percent) when compared to non-LGBTQ+ respondents (20.5 percent) and disproportionate impacts, which reflects the historical and persistent systemic marginalisation experienced by these groups.  

Respondents below the age of 35 were significantly more likely to report an experience of NCII (33 percent) than older respondents (19.7 percent for 35 to 49-year-old respondents). While women (22.3 percent) and men (22.7 percent) appear to experience similar rates of victimisation, women report higher levels of impact than men. This discrepancy is because the psychological impact of victim-survivors’ experiences is shaped by cultural expectations, as women are more frequently subjected to ‘slut-shaming’ narratives and social stigmatisation than men. Slut-shaming functions as a mechanism of victim-blaming by framing the victim-survivor’s appearance or perceived sexual behaviour as contributing to their victimisation.  

In the context of NCII, researchers found that victim-survivors who consensually created intimate images prior to being victimised are attributed with greater blame for their abuse than those who did not. The study found that this effect was much more pronounced for women in comparison to men who experienced NCII abuse. This differential treatment reflects broader gendered stigma, stereotypes and double standards that shift attention away from the perpetrator’s abusive conduct. 

NCII also includes the recording or capture of sexual offences themselves, where the creation of the image forms part of the abusive conduct. Sexual violence disproportionately affects women, girls and LGBTQ+ individuals, with other intersecting markers of vulnerability and marginalisation further shaping who is at risk. Likewise, these populations are disproportionately harmed by other forms of digital violence, such as doxing, which often occurs alongside NCII. Taken together, these dynamics situate NCII within broader structures of gendered violence and highlights its continuity with other forms of coercion and harm. 

Patterns of publicly observable content show a pronounced gendered asymmetry. Notably, according to data from 2023, women are disproportionately represented in non-consensual deepfake material and other forms of abusive synthetic imagery, as estimates suggest that around 99 percent of online deepfake content features women. In addition, multiple high-profile or publicly visible cases that were brought forward in recent years involve women as victims. This distribution is closely linked to the ways in which such material operates within broader gendered power relations in digital environments, where sexualised representations of femininity are more readily produced, circulated and mobilised as forms of targeted harm and intimidation. 

Perpetrator demographics vary depending on the type of abuse, methodology and definitions used. However, the researchers and authorities most frequently identify men as the primary perpetrators. Although perpetrators of NCII abuse can operate anonymously, they are typically identified as intimate partners, friends, relatives or otherwise acquainted with their victims. Emerging evidence suggests that NCII is also increasingly prevalent among youth, including in school environments. This finding demonstrates that like other forms of GBV, NCII abuse should not be considered as isolated acts by strangers, but as part of a wider continuum of violence that often involves trusted individuals. 

Synthetic NCII (NSII) 

NSII is produced through methods such as image-editing, face-swapping and voice cloning. Other terms used interchangeably with NSII include synthetic intimate image abuse and AI-generated non-consensual intimate imagery. Another commonly used term is ‘deepfake’, which originated in 2017 from a Reddit community where non-consensual content was widely shared. Because the term refers to a specific deep learning technique, rather than the abusive nature of the content, it does not adequately capture the full range of NSII abuse. 

While NSII commonly refers to digitally created or manipulated content that portrays a person in nude, semi-nude or sexually explicit contexts without their consent, the use of such manipulated content to inflict harm often also extends to other forms of intimate or identity-based violation. For example, NSII can include the removal of religious or cultural markers, as well as the alteration of appearance in ways intended to humiliate, degrade or delegitimise the subject. For example, ‘nudification’ tools were weaponised specifically against Muslim women to remove their hijabs in several documented cases. 

The key difference between NSII abuse and other forms of NCII abuse lies not in its severity, as victim-survivors often describe NSII’s impacts as “akin to those resulting from physical sexual assault.” Rather, the key difference lies in NSII’s distinct identity violations, as its fabricated likeness infringes on victim-survivors’ sexual and social identities, as well as their general sense of self. 

In addition, due to the wide availability of GenAI systems, the entry barriers to committing image-based abuse have lowered. Today, perpetrators increasingly rely on generative AI systems and so-called nudification apps. These tools have become more readily available, require minimal technical expertise and allow for the creation of more realistic content, thereby enabling abusive content to be produced quickly and at scale. 

The creation of NSII through deepfakes often relies on depictions of real humans that are used as source material without their consent. In many cases, this process relies on face-swap technology, in which one person’s face is superimposed onto another person’s body. These bodies typically belong to sex workers, who tend to face professional, financial and emotional consequences when their content is used without consent for that use. In some cases, the material has also been linked to sex trafficking, which demonstrates that cases of deepfakes often involve multiple victims, regardless of whether they can be identified.  

The effects and consequences of NCII 

NCII infringes on a range of fundamental human rights, including the right to respect for private and family life, the protection of personal data, psychological integrity and personal autonomy. Simultaneously, NCII also undermines freedom of expression and the economic and moral rights of individuals whose bodies and labour are used without consent. These violations are reflected in the profoundly damaging effects on victim-survivors’ lives. Victim-survivors of NCII abuse have described their experiences as life-altering events that divide their lives into a ‘before’ and ‘after’, often bringing about drastic and lasting changes. Many report serious health impacts—including anxiety, depressive disorders, post-traumatic stress disorder (PTSD) and suicidality—and a severely diminished sense of safety, with consequences for their relationships, social lives and participation in public life.  Individuals who previously consensually shared intimate images and sex workers experience victim-blaming at a higher rate than other victim-survivors of NCII abuse. 

Often, victim-survivors are forced to confront the abuse repeatedly, due, in part, to legal processes and the permanence of images and videos in digital spaces. Furthermore, acts of sextortion, as well as the reputational harm caused by the sharing of intimate images, can lead to significant financial losses, including job loss or broader economic insecurity. 

In addition to the direct impacts of NCII abuse on victim-survivors, the perceived threat of NCII abuse alone also functions as a tool for intimidation that fosters self-censorship. This dynamic contributes to a  chilling effect, whereby individuals at disproportionate risk to experience abuse withdraw from online and offline public spaces to avoid harm. As a result, progress towards gender equality is inhibited and existing disparities in political representation for already underrepresented groups can become more entrenched. 

The mitigation of NCII abuse 

NCII abuse is rooted in the same systemic inequality as other forms of GBV. Therefore, effectively mitigating this type of abuse requires coordinated interventions that recognise it as part of a broader continuum of gendered violence across all areas of life. A holistic response should combine measures such as victim-survivor-centred laws and enforcement mechanisms, accessible reporting mechanisms, trauma-informed support services and initiatives like consent education. However, the response also requires addressing restrictive gender norms and stigmatisation such as ‘victim-blaming’ narratives in society and confronting the gendered structural inequalities that make some groups disproportionately vulnerable to marginalisation. Crucially, mitigating all forms of GBV should prioritise reducing the burden on victim-survivors whilst holding perpetrators and those who facilitate and enable harm accountable.  

By providing the infrastructural space in which this abuse can persist, online platforms play a central role in producing, circulating and sustaining NCII. Despite this, technology companies’ responses often remain largely reactive and inconsistent in practice. While some providers include explicit prohibitions on NCII within their terms and conditions and provide dedicated reporting pathways in line with certain jurisdictions’ legal requirements (such as Meta, X and Microsoft), this approach is not universal. Even where formal takedown mechanisms exist, research indicates that requests are not always processed effectively and harmful content often remains accessible on platforms. Smaller or decentralised platforms may lack specific NCII provisions altogether, instead relying only on broader categories like privacy violations. Industry Codes of Practices can serve as an important starting point to strengthen cross-industry efforts to tackle NCII, but these agreements often only cover parts of the wider NCII ecosystem and are typically voluntary. Other proactive approaches to preventing GBV in digital environments include concepts such as ‘Safety by Design’, ‘Design Justice’ and ‘trauma-informed principles’, which seek to ensure the integration of safeguards into systems from the outset and emphasise the meaningful inclusion of victim-survivor perspectives in shaping technological development and governance. However, the amount of NCII widely available online demonstrates that neither large social media nor technology companies involved in producing and disseminating NCII have currently adopted these approaches in a meaningful manner.  

Several jurisdictions have established broader online safety regulations that impose legally binding obligations on platform providers, that are also relevant to NCII abuse. For example, the EU’s Digital Services Act (DSA) mandates reporting mechanisms and requires platforms to act on illegal content and further recognises GBV as a systemic risk. Other examples of broader online safety regulations relevant to NCII include the UK’s Online Safety Act , the US’ Take It Down Act and Australia’s Online Safety Act. Additionally, the UK’s Crime and Policing Act and amendments to the EU’s AI Act intend to hold accountable providers of AI systems that produce synthetic NCII. Although these frameworks introduce formal accountability obligations for service providers, they face implementation challenges. For example, early research on the DSA indicates significant shortcomings in compliance and reporting practices, limitations in enforcement capacity and an overreliance on platform discretion.  

Beyond platform accountability frameworks, key legal reform issues include definitions of ‘consent’ and ‘intimate’ imagery and whether to criminalise the creation of NCII alongside its dissemination. For example, the UK has taken steps to criminalise requesting, creating and disseminating NCII, whereas the EU Directive 2024/1385 mandates EU member states to criminalise sharing NCII by 2027 but not its creation.   

In Germany, a new draft law looks to streamline and close regulatory gaps with a scope that goes beyond criminalising sexualised abuse to also include the use of deepfakes in the context of disinformation more broadly. Denmark is exploring innovative legislative measures that would expand this scope further by giving people copyright control over their own face, voice and body. Denmark’s approach could tackle NCII as an infringement-based rights claim, rather than a privacy or criminal-law issue only, which would offer a wider scope than most other existing laws on NCII. This shift would allow victim-survivors to seek redress under already well-established notice and takedown schemes.  

Notably, existing laws in many jurisdictions already address NCII partly through provisions related to privacy, harassment and data protection. However, these laws are often considered fragmented and insufficiently enforced. Whilst addressing shortcomings in legal frameworks and policy responses is an important step in mitigating NCII, these measures must be effectively implemented and enforced in practice to achieve meaningful change for victim-survivors.