New evidence on online harms affecting children
The Youth Justice Board released an evidence pack in June 2026 synthesising current research, statistics, and policy developments on online harms affecting children. The report particularly warns that image-based online sexual harms are intensifying and that existing offences may criminalise children even when they are themselves victims of online harm.
Details
Children are increasingly exposed to a range of online harms, including cyberbullying, online sexual harms, online radicalisation, and criminal and sexual exploitation. Platform design features—including algorithms, infinite scrolling, direct messaging, and popularity metrics—are deliberately employed to influence users’ behaviours and children are particularly vulnerable. Online harms frequently overlap and disproportionately affect girls, particularly in relation to intimate image sharing and AI-generated sexual imagery. For instance:
- 70% of children aged 13-17 reported encountering violent content on social media in the past 12 months, up from 55% in 2022.
- 62% of children who engaged in real-world violence had participated in online conversations about harming others.
- In the year ending March 2025, 54% of all proven sexual offences committed by children aged 10-17 related to indecent or extreme pornographic images or videos. Most concerned the creation of indecent images of children (45%), and 975 were committed by boys.
- Online child’s sexual abuse and exploitation accounted for at least 42% of recorded CSAE offences in 2024, up from 35% in 2023, with AI-generated imagery driving the increase of the creation of such unlawful content.
Producing, possessing, or distributing ‘indecent’ images of a child is an offence in England and Wales under the Protection of Children Act 1978, the Sexual Offences Act 2003, with more offence categories added under the recent Online Safety Act 2023. Notably, these offences apply to children themselves, placing those who consensually create or share images at risk of criminalisation. ‘Outcome 21’ allows a non-criminal disposal absent aggravating features, though the report notes it has sometimes been misapplied to serious offences without referral to Youth Justice Services.
The report notes that traditional protective factors, such as general parental supervision, are less effective in a rapidly evolving digital landscape. However, regulatory developments are on the horizon, including: (i) a Crime and Policing Bill amendment requiring platforms to remove non-consensual intimate images within 48 hours; (ii) a planned ban on social media for under 16s from Spring 2027, with device controls to restrict explicit content; (iii) an Ofcom investigation into ‘X’ (formerly Twitter) over AI-generated child sexual abuse material; and (iv) forthcoming Youth Justice Board research on youth justice responses to online sexual offences, expected in 2027.
Commentary
The digital environment is not merely a space where offline harms are replicated. It is a distinct site of risk in which platform architecture actively facilitates harm to children. In particular, legislation can leave children drawn into image-based offending frequently as both victims and perpetrators. For practitioners, the priorities are practical. Children’s online activity should be a routine line of enquiry in every assessment and instruction, not an afterthought, and safeguarding, rather than sanction, should drive the response. As the report itself cautions, ‘no single measure can keep children safe’. The coming reforms will only benefit children if implemented with children’s welfare, rather than prosecution, at the centre.