We responded to the Scottish Parliament’s Criminal Justice Committee inquiry on AI-generated image abuse and deepfakes.
1. What are the key issues relating to AI-generated intimate image abuse and deepfakes?
The creation of AI generated deep fake abuse and sharing of intimate images is a serious violation of children and young people’s rights and has significant implications not only for their safety from harm but also their wider human rights on and offline.
Public authorities in Scotland have duties to uphold children’s rights in domestic law under the UNCRC Incorporation (Scotland) Act when carrying out a ‘relevant function’. However Scottish and UK Governments also have an obligation under international law to uphold these rights in all areas through the creation and implementation of law and policy at a national level.
Child sexual abuse material (CSAM) which includes the creation and sharing of images and deepfakes using AI, engages multiple children’s rights under the UNCRC, in particular:
- Children’s right to be free from violence, abuse and exploitation, including sexual abuse under Article 19 and children’s right to be protected from sexual exploitation and abuse under Article 34
- Under the UNCRC, children have a right to be protected from sexual abuse and exploitation, including the sharing of intimate images and creation of AI-generated child sexual abuse materials(CSAM).[1] It is a breach of a state’s obligations under the Convention to fail to adopt appropriate legislative and policy measures to prevent child sexual abuse, including in this form.
- Children’s right to privacy under Article 16
- The creation and sharing of AI-generated deepfake child abuse and intimate images involves a significant invasion of children and young people’s privacy – the use of personal data, including images of their faces, is a key part of AI deepfake abuse. Protections for children’s privacy rights must include requiring companies to institute privacy by design in development of their services and to support children and young people and their parents and carers in removing material which violates this right.[2]
- Children’s right to free expression under Article 13
- Children have a right to express their ideas and views and access information on a range of subjects. Social media and other digital tools and websites are a key space for them to do this. The threat of deepfake abuse can make children and young people, particularly girls, less comfortable accessing online spaces and sharing things within them, for fear of being targeted. This chilling effect undermines their access to important spaces for free expression and access to information.
- Children’s right to be free from discrimination in the enjoyment of their rights under Article 2
- AI generated and deepfake CSAM disproportionately targets girls, leading to a discriminatory violation of their rights. Measures to address the issue must take account of its discriminatory impact and seek to address this, including by targeting discriminatory norms and behaviours and providing support for girls and young women subject to this type of abuse.
2. How effectively do current laws, policies, enforcement arrangements and support services respond to issues associated with AI-generated intimate image abuse and deepfakes?
The creation and sharing of AI-generated intimate image abuse and deepfakes depicting children will often already be classified as an offence under section 52 of the Civic Government (Scotland) Act 1982, under the term ‘pseudo photograph’. We have previously commented on this in response to the Scottish Government’s consultation on Improving Protections in the Justice System for Women and Girls.[3]
The Crime and Policing Act 2026 criminalises the development of tools which may be used for the generation of AI deepfakes or generation of CSAM (so-called “nudification apps” as well as other tools which can be used to generate CSAM images) in England and provides for amendments to existing Scottish legislation under section 74 to do the same. Section 76 of the Act provides for the criminalisation of instructions relating to the creating of CSAM deepfakes or AI generated images in Scotland. In our response to a Scottish Government consultation this year, we raised concern that the wording of Scottish proposals to address this issue may create loopholes in the law.[4] We recommend Scottish Ministers consider commencing provisions of the Crime and Policing Act to ensure the creation of these types of tools is criminalised in Scotland as quickly as possible.
Companies providing user to user or search services are also under an obligation to prevent sharing under the Online Safety Act. Ofcom’s most recent update to its Illegal Content and Protection of Children Codes of Practice has strengthened some protections in this area, allowing for the banning of users and increased use of hash matching to detect CSAM including AI-generated images.[5] Effective enforcement of measures at the platform regulation level is essential to tackling this form of abuse. Given the scale of the issue, with research showing numbers of this type of abuse increasing, enforcement is clearly not yet effective.[6]
In general, in relation to AI and online safety, and associated criminal abuse, there is a need for children and their parents and carers to have access to support and education on dealing with the risks of some of these tools. Parents and carers need to have the information needed to support young people in accessing new technologies safely, and children and young people should feel able to raise concerns with trusted adults who understand the technology they are talking about and can support them where they are at risk. However, it is important that the burden of ensuring children’s online safety is not left to children themselves and their families – much more must be done at a legislative, regulatory and criminal justice level to support children’s safety.
We have raised concern, in response to UK Government plans to restrict access to social media for under-16s, that blanket bans which are not easily enforceable can leave young people still able to access spaces that are less regulated, or push them to riskier apps and areas of the internet.[7] When young people are not ‘supposed’ to be on social media but can still access it, it can make them feel less able to seek help when they are at risk or exposed to harmful content, including AI-generated CSAM.
In terms of safeguarding, training on the risks of some of these emerging technologies is essential for all those working with children. It’s also important to consider rising exposure to normalised misogyny online for young people and improve education in this area, in particular with regards to this type of crime. Where such tools become increasingly available and normalised online, children and young people themselves can be involved in the perpetration of these crimes, and tackling this requires engaging with young people as well as awareness of these newer types of harmful behaviours in those working with young people.
4. Are there any examples, evidence, research, international approaches or emerging developments that you would like to draw attention to?
The Children’s Commissioner for England published a report[8] in April 2025 based on focus group discussions with children and young people in England and analysis of the availability of tools and apps. It highlights the risks to children posed by AI deepfakes, and their views and experiences of the technology. It emphasises children and young people’s concerns relating to the technology and the way it impacts their online and offline behaviour and experiences, with young girls in particular feeling at risk online, including from their peers. While some of the recommendations have been addressed through legal changes, such as the banning of ‘nudification’ apps in England and changes to Ofcom’s codes, some challenges at the platform level including the importance of screening for risks at design stage and the need for greater AI regulation are still to be addressed.
At the international level, recent guidance on protecting children’s rights in the context of AI can be found in the Joint Statement on Artificial Intelligence and the Rights of the Child, co-signed by numerous international bodies including the UN Office for the High Commissioner on Human Rights, UN Special Procedures and UN Human Rights Treaty Bodies.[9] It highlights the need for a child rights-based approach to the design, development, deployment and governance of AI and sets out measures states should take to protect children’s safety in this context.
At the UK level, the role of the Online Safety Act and wider AI regulation in keeping children safe from these types of crimes, by tackling them at the platform level, must continue to be prioritised. Too often harms to children online, including AI-generated CSAM, are materialising at a faster pace than regulatory and criminal justice efforts. Legislation and enforcement in this area must keep pace with the serious harms children are at risk of online.
[1] UNCRC Committee, General comment No. 13 (2011) The right of the child to freedom from all forms of violence, para 31(b)
[2] See UNCRC Committee, General comment No. 25 (2021) on children’s rights in relation to the digital environment, para 70
[3] CYPCS, Response to Improving Protections in the Justice System for Women and Girls Consultation, August 2026
[4] CYPCS, Response to Improving Protections in the Justice System for Women and Girls Consultation, August 2026
[5] Ofcom, Statement: Online Safety – Additional Safety Measures, November 2026
[6] See Internet Watch Foundation, Rise in AI child abuse imagery fuels call for urgent EU action, October 2026 and Internet Watch Foundation, Harm without limits: AI child sexual abuse material through the eyes of our Analysts, 2026
[7] CYPCS, Response to the UK Government’s ‘Growing up in the online world: a national consultation’, May 2026
[8] Children’s Commissioner for England, Children, nudification tools and sexually explicit deepfakes , April 2025
[9] ITU, CRC, UNICEF, ILO, IPU, UNICRI, UNODA, OHCHR, the SRSG-CAAC, the SRSG-VAC, the UN Special Rapporteur on the sale, sexual exploitation and sexual abuse of children, and Safe Online, Joint Statement on Artificial Intelligence and the Rights of the Child, January 2026
