Consultation outcome

Summary of evidence, methodology, and organisations who responded to the consultation: June 2026

Updated 15 June 2026

1. Introduction

This document summarises responses to the ‘Growing up in the online world: a national conversation’ full consultation questionnaire, and relevant email submissions. Only the portion of the consultation relevant to the announcement on 15 June 2026 is summarised here. The remainder of the consultation results will be published alongside relevant announcements in July.

2. Background

The Department for Science, Innovation and Technology (DSIT) consulted on further measures to prepare children for the future in an age of rapid technological change. This included potential age restrictions on social media and other services such as gaming sites and AI chatbots, restrictions on addictive design features and risky functionalities, and better support for parents and families. The consultation ran from 10:30am on 2 March 2026 to 11:59pm on 26 May 2026.

Respondents could submit their views through the ‘full consultation’ questionnaire, which included all questions, or separate, focused consultation questionnaires for children and young people up to 21, as well as their parents and carers. The parents’ and children’s consultations were hosted by our delivery partner Savanta.

In addition to the parent and children consultations, DSIT commissioned Savanta to conduct a nationally representative panel survey of parents and children, asking the same questions. We also accepted responses via email, including many pro-forma responses from a campaign.

Across the range of options, we received 116,211 responses, including 24,534 to the full consultation questionnaire, 279 unique emails, and 33,141 campaign emails. This subset of results is summarised in this document.

There were a further 39,116 responses to the parent’s consultation, 5,011 responses to the parent’s panel survey, 5,113 responses to the children and young people’s consultation, and 9,017 responses to the children and young people’s panel survey. The results from these have been published alongside this document.

3. Methodology

The consultation included a combination of closed (“multiple choice”) and open (free text) questions, as well as email responses, which didn’t follow the questionnaire format.

For the closed questions, we have presented the data tables from Smart Survey, the questionnaire platform the department uses. Because of the length of the questionnaire, the consultation asked very few demographic questions. However, we have detailed the responses by parents^{[footnote 1]}, for comparison with the equivalent questions in the parents’ consultation.

To analyse open questions in any consultation, the responses to each question are assessed, and common themes in responses are identified. Then the number of responses that include each theme is calculated.

Historically, these two processes have been done by large project teams of civil servants, working for months to read every response several times to collate a summary of all responses. However, this process can be completed more quickly and consistently by AI, with human oversight.

This consultation was analysed with Consult, an AI tool developed by the government specifically to analyse responses to public consultations. It analyses thousands of responses any government consultation might receive in hours, before presenting policy makers with interactive dashboards to explore responses directly.

Consult follows the general approach outlined above to analyse responses. After data cleaning and upload, Consult identifies potential common themes in responses. These themes are reviewed by officials who compare each theme to responses to confirm that they are representative. Officials edited any themes that were not accurate and discarded themes that were not relevant or were repetitive. When the themes were finalised, Consult counted the number of responses that mentioned each theme and reported the results for officials to analyse.

In the case of this consultation, a change to the standard Consult process was agreed. An interim run of the Consult process was completed around the mid-point of the consultation period, using the first c. 12,000 responses. Themes identified in this run informed the analysis of the responses in the full final data set.

To mitigate against any major changes in the nature of responses officials used:

• Analytical safeguards within the Consult tool, including flags for post-interim responses and evidence-rich submissions, to identify missed or emerging themes.
• Direct reviews for new and evidence-rich responses that could not be mapped to an existing theme.
• Analysis to demonstrate that the interim dataset is consistent with the final dataset.
• Comparison with another consultation which used an interim and final run.

The 279 email responses to the consultation were individually reviewed by officials because these responses typically did not follow the questionnaire format, arrived towards the end of the consultation period, were evidence-rich, and were submitted by key stakeholders. Officials analysed each submission and created a short summary of key themes and issues identified. These summaries can be found at the end of each section of questions and summarise views from stakeholders across all of the questions in each section.

These email responses often originate from civil society organisations, businesses and other stakeholders with a unique and important perspective, for example the impact on children with a disability, or a platform with an innovative service. Therefore, in some cases the insight of a single respondent is highlighted in the summary, to ensure that perspective is considered in policy-making.

A further 33,141 pro forma emails organised by a campaign were received. The proforma text for this response has been included as Annex A.

A list of organisations represented by respondents is included as Annex B.

3.1 Describing the prevalence of themes in free text

Consultations are not representative polls and are designed to elicit the range of insights on the policy, rather than assess the prevalence of opinions. Response analysis involves some subjective judgement.

Therefore, the summaries of responses to each of the open questions below, use these descriptive terms to indicate the prevalence of themes that were raised in responses:

Descriptor	Percentage of responses
‘very few’	0-5%
‘few’	6%-10%
‘a small number’	11%-25%
‘some’	26%-45%
‘around half’	46%-55%
‘many’	56%-75%
‘most’	76%-90%
‘a large majority’	91%-95%
‘an overwhelming majority’	96%-99%
‘all’	100%

In some cases, support for a theme may appear lower than expected. That may be because few respondents raised that issue specifically while most made a general statement about risks, for example, or it may occur where most respondents answered the question with a numerical minimum age, for instance, and only a few respondents included additional context to be summarised as a theme.

3.2 Describing the prevalence of themes in free text

Consultations are not representative polls and are designed to elicit the range of insights on the policy, rather than assess the prevalence of opinions. Response analysis involves some subjective judgement.

Therefore, the summaries of responses to each of the open questions below, use these descriptive terms to indicate the prevalence of themes that were raised in responses:

Descriptor	Percentage of responses
‘very few’	0-5%
‘few’	6%-10%
‘a small number’	11%-25%
‘some’	26%-45%
‘around half’	46%-55%
‘many’	56%-75%
‘most’	76%-90%
‘a large majority’	91%-95%
‘an overwhelming majority’	96%-99%
‘all’	100%

In some cases, support for a theme may appear lower than expected. That may be because few respondents raised that issue specifically while most made a general statement about risks, for example, or it may occur where most respondents answered the question with a numerical minimum age, for instance, and only a few respondents included additional context to be summarised as a theme.

3.3 Caveats

• Consultation respondents are a self-selecting sample, characterised by higher motivation and knowledge of how to engage with government. Responses are not from a nationally representative sample of the public. For example, we know that parents and children from lower socio-economic backgrounds are underrepresented in responses to the consultation. That is why a nationally representative panel survey of parents and children was also conducted by Savanta. The results of this panel survey have been published as part of this response.

• Due to the length of the survey, respondents were asked which Chapters of the consultation they wanted to answer the questions for. Further question routing (to ensure respondents only see relevant questions) was used in a very few cases, and a note has been made in the summary where and how it was used. Also, beyond some initial scoping questions, no question was compulsory, therefore not every question will have been responded to. The total count of responses and skips has been reported under each question.

• Percentages have been rounded to the nearest whole percent. Percentages may not sum to 100% due to rounding.

4. Benefits and risks of children using social media

4.1 Consultation results – question 1

What are the benefits of social media use, and being online, for children?

Summary of findings

Many of the respondents to this question reported social connection, inclusion, and wellbeing as a benefit, highlighting the role of online platforms in maintaining friendships and reducing isolation. Some respondents identified other benefits including educational access, while a small number emphasised the importance of developing media literacy and digital skills, along with preparation for adulthood and social connection. A small number referenced exposure to diverse perspectives, and a few respondents highlighted engagement with the news.

A few respondents pointed to more specific benefits, including organisation, collaboration, and event access and support, and a very few indicated that social media provides connections for neurodivergent and/or disabled children. A very few respondents also noted that being online is an integral part of modern childhood.

A notable proportion of respondents raised concerns in response to this question. A small number of respondents reported very little or no benefits of children being on social media, and a few others highlighted the importance of parental oversight and/or teacher oversight when children are online. Additionally, very few respondents suggested alternative ways to achieve online benefits without relying on social media.

Theme Name	Theme Description	Response Share
Social connection, inclusion, and wellbeing	Online platforms help children maintain friendships, build supportive communities, reduce isolation, and provide access to emotional support and mental health resources, especially for marginalised or vulnerable children.	Many
Educational access	Online access provides children with educational resources, supports academic development and creates an opportunity to form social connections.	Some
Very little or no benefits of children being on social media	Social media is often seen as less beneficial or even harmful compared to other online resources, with minimal benefits that are outweighed by risks or are available through other means.	A small number
Exposure to diverse perspectives	Being online exposes children to different cultures, viewpoints, and global issues, broadening their understanding, empathy, and critical thinking.	A small number
Media literacy, digital skills, preparation for adulthood and social connection	Online engagement helps children develop essential media literacy, technical, and communication skills, preparing them for adulthood and future opportunities.	A small number
Engagement with the news	Online platforms enable children to access news, express opinions, participate in civic discourse and activism, and develop as informed citizens.	Few
Parental oversight and/or teacher oversight key	The benefits of online engagement for children are maximised when use is safe, supervised, and guided, with parents playing a key role in moderating activity and ensuring security. A majority of these responses said there is no benefit to children being on social media.	Few
Organisation, collaboration, and event access	Online tools help children organise activities, collaborate on group projects, and engage with local communities or clubs based on their interests.	A few
Support and connections for neurodivergent and/or disabled children	Online resources and tools assist children with special educational needs or disabilities by providing support and connections.	Very few
Online life part of modern childhood	Being online is an inevitable and integral part of modern childhood, with online and offline lives deeply intertwined.	Very few
Children’s use of the online world for entertainment purposes	References to children going online for entertainment purposes.	Very few
Alternative ways to achieve online benefits	The advantages attributed to social media can be achieved through other means such as messaging, calls, or supervised internet use, without relying on social media platforms.	Very few
Connecting with peers and community	Mixed views on social media benefits (mostly negative) but some view it to provide opportunity to connect with peers and community.	Very few
Inclusion	Children may experience social pressure and feel left out if they are not online when their peers are, potentially leading to exclusion.	Very few
Concerns about algorithmic reinforcement and addictive design	Concerns regarding the impact of algorithms and addictive features.	Very few
Other	The response discussed an issue not covered by the listed themes.	Very few
No reason given	These responses did not provide substantive answers to the question	Very few

4.2 Consultation results – question 2

What are the harms or risks of social media use, and being online, for children?

Summary of findings

Around half of the respondents to this question reported that the harms and risks of social media use and being online include the increased exposure to harmful and age-inappropriate content. Some highlighted addiction, excessive screen time, and displacement of healthy activities, exposure to harms from algorithmic manipulation, and negative mental health and emotional wellbeing impacts. Some of the respondents also identified an increased risk of cyberbullying and online harassment and highlighted the increased risk of online grooming and exploitation.

Others emphasised risks relating to data and systems. A small number raised concerns about sharing personal data and information and targeted advertising, and while a small number highlighted the role of parental responsibility and digital literacy education in mitigating harms. A few referenced a lack of effective regulation and platform accountability, and a few also noted physical health risks and sleep disruption from online use.

A few respondents identified additional risks, including the facilitation of criminal behaviour and financial exploitation, and a very few identified uncertainty about long-term effects and emerging risks.

Theme Name	Theme Description	Response Share
Increased exposure to harmful and age-inappropriate content	Children are at significant risk of encountering violent, sexual, extremist, or otherwise harmful material online, which can negatively impact their psychological, emotional, and moral development.	Around half
Addiction, excessive screen time, and displacement of healthy activities	Online platforms are designed to be addictive, leading to excessive screen time, sleep disruption, and displacement of healthier activities and real-world social skills, negatively impacting attention, physical health, and academic performance.	Some
Increased risk of online grooming and exploitation	Children are vulnerable to being contacted, groomed, or exploited by predatory adults or strangers online, leading to risks such as sexual exploitation, abuse, and trafficking.	Some
Increased risk of cyberbullying and online harassment	Children face persistent bullying, harassment, exclusion, and abuse online, which can be relentless and lead to severe emotional distress and mental health consequences.	Some
Exposure to harms from algorithmic manipulation	Children are vulnerable to harmful content, misinformation, extremist content, and manipulative material online, often amplified by algorithms that influence beliefs and behaviours for engagement and profit.	Some
Negative mental health and emotional wellbeing impacts	Social media use is linked to increased anxiety, depression, low self-esteem, body image issues, loneliness and isolation, and even self-harm or suicidal ideation in children.	Some
Sharing personal data and information and targeted advertising	Children’s personal data may be collected, misused, or exposed online, increasing vulnerability, exploitation, and long-term reputational risks. Children’s data is exploited by targeted advertising.	A small number
Role of parental responsibility and digital literacy education	Parental supervision, digital literacy education, and guidance are crucial for mitigating online risks for children.	A small number
Physical health risks and sleep disruption from online use	Prolonged screen time and online engagement contribute to sedentary lifestyles, sleep deprivation, eye strain, obesity, and other negative physical health outcomes in children.	Few
Facilitation of criminal behaviour and financial exploitation	Social media can facilitate criminal activities such as scams, fraud, gambling, and financial exploitation, with children being targeted or involved in these risks.	Few
Lack of effective regulation and platform accountability	Social media platforms lack sufficient regulation, moderation, and accountability to protect children from harm, with calls for stricter safeguards and oversight.	Few
Uncertainty about long-term effects and emerging risks	There is concern about unknown or emerging long-term impacts of social media use on children’s development and wellbeing, including risks from new technologies like AI and deepfakes.	Very few
Concerns about government overreach and overregulation	Excessive government intervention may infringe on parental responsibility or be unwarranted compared to offline risks, potentially backfiring.	Very few
Environmental impact of increased device use by children	Increased use of digital devices by children contributes to e-waste and environmental concerns.	Very few
Other	The response discussed an issue not covered by the listed themes.	Very few
No reason given	These responses did not provide substantive answers to the question.	Very few

4.3 Consultation results – question 3

Do you think the benefits of children using social media, and being online, outweigh the risks, or the other way around?

Do you think the benefits of children using social media, and being online, outweigh the risks, or the other way around?	All responses (%)	All responses (total)	Of which parents (%)	Of which parents (total)
Benefits strongly outweigh the risks	16%	3,372	7%	683
Benefits somewhat outweigh the risks	10%	2,044	4%	382
Benefits and risks are roughly equal	9%	1,770	5%	498
Risks somewhat outweigh the benefits	11%	2,314	10%	958
Risks strongly outweigh the benefits	52%	10,684	72%	6,692
Don’t know / Prefer not to answer	2%	354	1%	72
Answered		20,538		9,285
Skipped		3,996		776

4.4 Email summaries on benefits and risks of children using social media

Summary of responses:

• Many stakeholders, across both civil society and industry, agreed being online can bring significant benefits for children, particularly in supporting social connection, access to information, creativity, learning and self-expression. At the same time, many respondents emphasised that these benefits sit alongside substantial harms and risks.

• Respondents suggested that both benefits and harms are unevenly distributed. Several respondents therefore implied that children’s online experiences are shaped by age, vulnerability and personal circumstances, and that a more nuanced approach is needed than treating all children or all online risks as the same.

• Some respondents argued that the relationship between children’s online activity and outcomes for wellbeing is complex and depends on the context and individual child. This was a common position among industry respondents, who cautioned against assuming that online engagement is inherently harmful or that the same activity will have the same effect in every setting. Other stakeholders noted that evidence on causal links between harm and online behaviours remains developing and contested.

• Overall, respondents agreed that children’s online experiences involve a mix of benefits, harms and wider risks. Across the responses, there was support for the principle that children should be able to access the various benefits of being online without being exposed to avoidable harms.

4.5 Benefits

Online Connection

Civil society and industry stakeholders agreed that being online and using social media can support social connection, access to information, creativity, learning, and self-expression. Stakeholders highlighted how being online and on social media can help children connect with family and friends, find community, and peer support.

Industry respondents placed greater emphasis on the positive role of digital services in supporting connection, participation and development where services are designed with safeguards, age-appropriate defaults and parental tools.

Digital literacy

Industry respondents emphasised the opportunities digital services can provide, highlighting access to digital skills development and preparing children for further education and work. One stakeholder argued that being online benefits children’s digital literacy and digital skills and that this is essential for their future employment and civic participation.

News and Information

Some MPs who responded to the consultation noted that young people at their constituency events raised that they rely on social media to access the news and current affairs. One respondent noted the importance of young people’s access to lawful journalistic and democratic content.

Education

Education was a key benefit for civil society and industry stakeholders, noting social media and access to the internet can provide opportunities for learning and development, including for children in marginalised groups such as ethnic minorities and those with special education needs and disabilities (SEND).

Industry respondents emphasised the opportunities digital services can provide, highlighting access to educational resources, digital skills development, creativity and community.

Stakeholders noted that educational online tools provide children with meaningful screentime and aid their education around language skills. For instance, Welsh young people reported that social media and internet access was important to them for Welsh-language learning.

One stakeholder acknowledged the benefit of online platforms for high-value educational and social resources and that restrictions would impact access to those in need of support.

Safeguarding and Support

One respondent stressed that children must not lose access to the benefits of being online which includes vital safeguarding support via the likes of Childline.

Stakeholders also noted LGBTQ+ and neurodivergent children have specific support benefits from access to being online. These children were reported to use social media and the internet for mental health support, finding community, online connectivity including socialising and making friends, and finding resources relating to their lived experiences.

One stakeholder highlighted the benefits of being online for children with disabilities. They noted that Deaf and Deafblind young people rely on online spaces for access to BSL culture and peer support.

Welsh children reported that social media and apps like WhatsApp were important for rural connectivity, in particular maintaining contact with their parents and carers and feeling safe in rural areas.

Medical Information

A health-related stakeholder flagged social media and digital tools are frequently used by children and young people for medical and mental health advice, and peer-support networks. They noted that any future policies must take into consideration children with health conditions and their rights to access to healthcare, peer support, and accessibility tools.

Opportunities and Activities

One youth organisation fed back that 100% of their members use social media to find opportunities, activities, and support. Another stakeholder noted the positive role social media plays in informing young people about careers and employment opportunities, particularly in sectors such as construction that face significant skills shortages.

One stakeholder distinguished between unsafe services intended for adults and well-designed, child-safe services, which they argue can drive creativity and connection for children. Gaming industry respondents also argued that online gaming has developmental and social benefits for children.

• Similarly, another stakeholder argued its service has overwhelmingly positive wellbeing benefits for UK teenagers, and that they should be able to access cultural, creative, and educational content.

Risks and Harms

• At the same time, many respondents emphasised that these benefits sit alongside substantial harms and risks.

4.6 Harmful content

• Respondents highlighted children’s exposure to harmful content and unhealthy patterns of use, as well as wider impacts on sleep, self-esteem, and concentration.

• One stakeholder noted concerns from both youth workers and young people about the impact of online harms, violent content, misogyny, exploitation, cyberbullying, addictive design, and unsafe digital environments on children and young people’s wellbeing, relationships, and safety.

• Stakeholders also suggested that harms can be gendered, with girls more likely to encounter content promoting unrealistic body types and boys more likely to encounter explicit content. One civil society respondent flagged online harms to children are not a technological issue rather an extension of societal harms like misogyny and gendered violence.

Interactions with strangers

• An MP roundtable for pupils in the North of England noted that 71% of participants have received unsolicited messages from strangers online.

Harmful and addictive designs

• Stakeholders referred to harms from recommender algorithms where personalised content can expose children to wellbeing harms, including harmful content. One stakeholder noted that many persuasive or manipulative design features depend on the use of personal information.

• Civil society stakeholders argued that persuasive platform design choices can make harmful content more frequent or more difficult for children to avoid online.

Data and Privacy Risks

• Young people highlighted that they have concerns over their own digital footprint, including where parents have been the ones to share data about their own children. They also raised concerns about sharing ‘bad’ or misinformation. This was echoed by other civil society stakeholders too.

Bullying

• Reports raised concerns about group-chat trolling, bullying, and in particular risks for SEND children.

Hatred and online radicalisation risks

• Civil society stakeholders raised concerns that young people online are susceptible to far-right rhetoric and online radicalisation. In particular civil society stakeholders were worried about algorithmically driven radicalisation, specifically far-right ‘engagement bait’ that targets young people on social media platforms.

• A range of civil society stakeholders also raised concerns on the relationship between social media usage and grooming, and exposure to violence and extremism.

Mental and physical health risks

• Concerns were raised about the mental health impacts on children from social media usage, as well as addiction-like behaviours. One respondent noted that health harms — mental, physical and social — are being seen daily across every specialty of medicine. Similar concerns were raised by mental health clinicians.

• Some stakeholders highlighted that marginalised groups such as neurodiverse and ethnic minority children may be more likely both to experience benefits and experience harm online.

• Stakeholders specifically addressed risks linked to self-harm, suicide, eating disorder and depression content, citing survey evidence that over the course of one week more than a third of 13–17-year-olds in the UK surveyed had seen at least one type of suicide, self-harm, depression or eating disorder material. One stakeholder presented evidence that social media was directly implicated in children presenting to hospital with self-harm, suicidal ideation, eating disorders and mental health crises. They also distinguished between ‘being online’, where they felt the benefits outweighed the risks, and ‘using social media’ where the risks outweighed the benefits.

5. Restricting social media services by age

In this section, we consulted on whether there should be a legal requirement for social media services to have a minimum age of access, including the extent to which respondents agree that this should be set at 16, and whether a lower threshold (such as 13, 14 or 15) would be more appropriate.

We also asked about the potential impacts of raising the minimum age above 13, including effects on children’s safety and wellbeing, parents and carers, users’ privacy and data, and business costs, revenue, and innovation.

Consultation results – question 4

Would you support a legal requirement for social media services to have a minimum age of access?

Would you support a legal requirement for social media services to have a minimum age of access?	All responses (%)	All responses (total)	Of which parents (%)	Of which parents (total)
Yes	71%	15,552	89%	8,443
No	26%	5,671	10%	979
Don’t know/ Prefer not to answer	3%	605	1%	77
		21,828		9,499
		2,706		562
Answered		21,828		9,499
Skipped		2,706		562

5.1 Consultation results – question 5

To what extent do you agree or disagree with the following statement: “Social media services should have a minimum age of access of at least 16 and should not be accessible to any children under that age?”

(This question was only asked of those who responded “Yes” or “Don’t know” to the question above)

To what extent do you agree or disagree with the following statement:

“Social media services should have a minimum age of access of at least 16 and should not be accessible to any children under that age”	All responses (%)	All responses (total)	Of which parents (%)	Of which parents (total)
	Strongly agree	78%	12,209	88%	7,415
	Somewhat agree	12%	1,897	8%	671
	Neither agree nor disagree	1%	204	1%	40
	Somewhat disagree	4%	627	2%	159
	Strongly disagree	4%	640	2%	156
	Don’t know / Prefer not to answer	0%	26	0%	3
	Answered		15,603		8,444
	Skipped		8,931		1,617

5.2 Consultation results – question 6

Would you support a legal requirement for social media services to have a minimum age of access lower than 16? If so, at what age would you set it?

(This question was asked with slight variations for those who responded “Strongly agree” or “Somewhat agree” to the question above, as opposed to all other responses)

Would you support a legal requirement for social media services to have a minimum age of access lower than 16? If so, at what age would you set it? (“strongly agree”, “somewhat agree” responses only)	All responses (%)	All responses (total)	Of which parents (%)	Of which parents (total)
Yes – 13	3%	458	2%	163
Yes – 14	4%	542	3%	203
Yes – 15	5%	703	4%	338
No – not lower than 16	81%	11,622	87%	7,023
Don’t know / Prefer not to answer	1%	165	1%	46
Other (please specify)	6%	823	4%	347
Answered		14,313		8,120
Skipped		10,221		1,941

Would you support a legal requirement for social media services to have a minimum age of access lower than 16? If so, at what age would you set it? (all other responses)	All responses (%)	All responses (total)	Of which parents (%)	Of which parents (total)
Yes – 13	53%	673	44%	140
Yes – 14	18%	223	21%	66
Yes – 15	4%	46	5%	16
Don’t know / Prefer not to answer	3%	37	2%	6
Other (please specify)	23%	287	27%	87
Answered		1,266		315
Skipped		23,268		9,746

5.3 Consultation results – question 7

What do you think the impacts would be of having a minimum age requirement higher than 13 for social media services?  For example, impacts on the safety and wellbeing of children, or the impact for parents and carers, as well as other users. You could also comment on the impact on all users’ privacy and data or on business costs, revenue, and innovation.

Summary of findings:

The most frequently cited theme among respondents was that an age requirement would improve child safety and wellbeing, referenced by around half of those who responded to this question. Respondents identified a range of anticipated benefits, including greater protection from online harms, reduced exposure to inappropriate content, lower risk of addiction and negative mental health impacts, and support for healthier child development. A small number of respondents advocated for a more balanced approach centred on parental responsibility, emphasising that measures such as greater parental involvement, improved digital education, safer platform design and gradual introductions to online services should be considered as preferable or necessary complements to simply raising the age limit.

A small number of respondents also raised privacy, data security, and civil liberties concerns. These responses were focused on how enforcing minimum age requirements would impact data collection, loss of anonymity, enable surveillance and potential government or corporate overreach, affecting all users. A small number also highlighted enforcement challenges and circumvention risks, stating that age limits can be difficult to enforce and that children may bypass restrictions or migrate to less regulated and more risky platforms because of minimum age requirements.

A small number of respondents said that restricting social media for younger users may hinder their digital skill development, limit access to educational and support resource and could increase social isolation, particularly for marginalised children. However, a small number of others considered that raising the minimum age would help establish clearer societal expectations around children’s use of social media, affording children greater opportunity to develop before facing online pressures. Additionally, a few respondents suggested that a minimum age requirement for social media would naturally make children spend more time engaged on other activities, such as outdoor pursuit or time with friends and family.

A small number of responses also referenced business, economic and innovation impacts however, the sentiments were mixed. Respondents mentioned that enforcing an age limit could be difficult for companies to enforce, driving them out of the UK market. However, other responses highlighted that there should be greater responsibility on these tech companies and that child safety should be a priority over profit.

Theme Name	Theme Description	Response Share
Improved child safety and wellbeing	Raising the minimum age was seen as protecting children from online harms, cyberbullying, inappropriate content, addiction, and negative mental health impacts, supporting healthier development.	Around half
Privacy, data security, and civil liberties concerns	Enforcing higher minimum age requirements raised concerns about intrusive age verification, data collection, surveillance, loss of anonymity, and potential government or corporate overreach affecting all users.	A small number
Preference for parental responsibility and balanced approaches	Parental involvement, digital education, safer platform design, and gradual introduction are considered preferable or necessary complements to simply raising the age limit.	A small number
Enforcement challenges and circumvention risks	Higher age limits are difficult to enforce, as children may bypass restrictions or migrate to less regulated platforms, potentially undermining the policy and exposing them to greater risks.	A small number
Business, economic, and innovation impacts	Stricter age requirements could reduce the youth user base and revenue for social media companies, increase compliance costs, harm smaller platforms, and potentially stifle innovation and competition.	A small number
Negative impact on digital literacy and access	Restricting social media for younger users may hinder their digital skill development, limit access to educational and support resources, and increase social isolation, particularly for marginalised children.	A small number
Establishing societal norms and delaying exposure	A higher minimum age would help set societal expectations about appropriate social media use for children, normalise delayed use, and allow children to develop before facing online pressures.	A small number
Displacement to other activities	If access were restricted, children would naturally spend more time on other activities, particularly outdoors, with friends or with family.	Few
Increased risk of social exclusion and stigma	Raising the minimum age for social media use could lead to greater feelings of isolation, exclusion, and inequality among children, especially those from vulnerable or minority groups who rely on online communities for support.	Few
Unclear or minimal impact of age limit change	There is uncertainty or belief that raising the age limit would have little effect, as children may still access social media or outcomes depend on enforcement and complementary measures.	Few
Support for higher minimum age and legal consistency	Advocates proposed setting the minimum age at 16 or higher to align with other legal thresholds and better safeguard children, while others debated the appropriateness of the current age limit.	Very few
Infringement on children’s rights and autonomy	Higher age limits may restrict children’s rights to information, community, and autonomy.	Very few
Need for child-friendly alternatives and activities	If access is restricted, there was support for developing safer, age-appropriate social media platforms and for providing more offline activities, community support, and safe spaces for young people.	Very few
Exceptions for essential services	Certain applications, such as those supporting victims of abuse or managing addictions, should remain accessible to young people even if general social media access is restricted.	Very few
Support for banning social media for all ages	A minority view held that social media is so harmful it should be banned for everyone, not just those under a certain age.	Very few
Current age limit not based on scientific evidence	The existing minimum age of 13 was viewed as arbitrary and not grounded in scientific understanding of child development or readiness for social media.	Very few
Environmental and physical health considerations	Reducing young users could decrease data usage and energy consumption, while intensive age verification systems may have negative environmental impacts or reduce exposure to electromagnetic radiation.	Very few
Other	The response discussed an issue not covered by the listed themes.	Very few
No reason given	The response did not provide a substantive answer to the question.	Very few

5.4 Email responses on restricting social media services by age

Summary of responses:

Support for a minimum age of access

• There is broad support across civil society respondents for some form of legal requirement for social media services to have a minimum age of access. A range of civil society organisations supported some form of minimum age requirement, though they differed significantly on what age that should be and to which services it should apply.

Levels of support for what that age should be

• Among those who support a minimum age, there is no consensus on the appropriate threshold. Several civil society respondents strongly supported 16 as the minimum age of access. Other respondents argued for extending this to 17 or 18, particularly in relation to services with risky features.

• Conversely, other civil society respondents favoured a minimum age of 13 (generally the minimum age that services themselves currently specify) but with stronger enforcement than required by the OSA. One respondent stated that a minimum age at 16 would be better than the status quo, but advocated for risk-based minimum ages where expert-led guidance determines whether services are accessible at 13, 16, or adult-only. Another respondent advocated prohibiting personalised services for under-13s with graduated protections up to 18.

Views on a blanket minimum age of access

• Many respondents, particularly from industry, opposed a blanket minimum age of access, instead arguing for either a risk-based or age-differentiated access. Several of those who opposed cited risks of unintended consequences, lack of evidence, and a preference for alternative approaches.

• This view is shared by several civil society and academic respondents. Many respondents expressed concern or opposition, favouring risk-based or graduated approaches over blanket bans. Other respondents rejected an Australian-style ban, citing early evidence of ineffectiveness. Several respondents also warned that children may be less likely to report harms if accessing services is perceived as prohibited. One respondent noted that they do not support a stand-alone minimum age as a primary policy solution and argued it must be secondary to stronger platform duties around age-appropriate design.

Graduated or staged approach

• Many respondents advocate for graduated, risk-based frameworks rather than a single threshold. An industry respondent favoured parental oversight for 13–15-year-olds with autonomy at 16, whilst other industry respondents supported age-differentiated access with progressively expanded features. A civil society respondent proposed expert-led safety standards determining whether services are accessible at 13, 16 or adult-only. An academic respondent suggested graded age-rating approaches similar to film and game classifications.

Risks to vulnerable groups and displacement

• Multiple organisations highlighted risks to LGBTQ+ young people and those with special educational needs, health conditions, or neurodivergence, for whom online spaces provide identity, peer support, and community. There was widespread concern that bans would push children onto lesser-regulated services with weaker safeguards.

Children’s rights

• Several respondents argued that overly broad restrictions risked violating children’s rights. A number of civil society organisations argued age-based bans risk infringing children’s fundamental and digital rights, such as rights to expression, information, and association. One respondent also argued that, as 16- and 17-year-olds will be enfranchised by the Representation of the People Bill, their access to journalistic content, news publisher content, content of democratic importance must be legally guaranteed.

Cliff edge and digital literacy

• A recurring concern was the “cliff edge” effect of a binary age threshold. Many respondents warned that children reaching 16 without prior exposure would lack the digital literacy and resilience to navigate online risks. Some respondents also emphasised the educational and developmental benefits of supervised, age-appropriate access during early adolescence.

Scope and definitional challenges

• Another theme across responses was the difficulty of defining “social media” and concern about an overly broad scope. Multiple industry respondents argued specific types of services should be excluded, such as e-commerce, vertical search, messaging, gaming and creative platforms, and music streaming. A civil society respondent noted the risks that a narrow definition misses high-risk services while a broad one captures lower-risk ones, whilst others argued scope should extend beyond social media to include gaming, messaging and AI chatbots.

Safety by design and platform accountability

• There was strong cross-sector support for placing greater responsibility on platforms and stronger safety-by-design duties. One respondent called for amendments to the Online Safety Act (OSA) focused on stronger risk mitigation and removal of the safe harbour provision. Another respondent argued platforms are responsible for designing their services on the basis that children are present and require protection. Several respondents across civil society and industry favoured design-level interventions over access restrictions. One respondent warned that a shift to access-based regulation risks letting platforms “off the hook” for safeguarding whilst another noted that the current OSA framework is already being implemented and policy should build on existing industry efforts.

Parental controls and age assurance

• Several respondents highlighted the role of parents and the importance of robust age assurance. Several industry respondents advocated for age verification at app store or OS level rather than platform level and emphasised parental responsibility and existing parental control systems. A civil society respondent stressed that without effective age assurance, any ban risks providing false assurance while children continue to access services in practice.

6. Restricting access to services based on features and functionalities

In this section of the consultation, we sought views on features and functionalities that may be particularly risky for children. We asked about features such as livestreaming, sending nude images and videos, disappearing content, connecting or talking to strangers, and location sharing.

The consultation asked for views on whether children should be restricted from accessing these features and functionalities and, if so, at what age. The consultation also considered the impacts of any potential restrictions.

6.1 Consultation response – question 12

Some online services allow their users to engage with the following functionalities. Do you think these functionalities should be age restricted so that children below a certain age cannot engage with them? (Please select all that apply)

• Live streaming
• Ability to send nude images or videos
• Disappearing content
• Location sharing
• Connecting or talking to strangers
• None of the above
• Other (please specify)
• Don’t know/ Prefer not to answer

Some online services allow their users to engage with the following functionalities. Do you think these functionalities should be age restricted so that children below a certain age cannot engage with them? (Please select all that apply)	All responses		Of which parents…
	Response Percent	Response Total	Response Percent	Response Total
Live streaming	66%	14,222	84%	7,931
Ability to send nude images or videos	79%	17,072	92%	8,654
Disappearing content	66%	14,188	85%	8,021
Location sharing	64%	13,808	77%	7,297
Connecting or talking to strangers	71%	15,252	89%	8,387
Don’t know/ Prefer not to answer	2%	510	1%	120
None of the above	13%	2,708	5%	483
Other (please specify):	12%	2,501	8%	773
Answered		21,505		9,451
Skipped		3,029		610

6.2 Consultation response – question 13

Based on your previous answers, please specify your preferred minimum age for each of the functionalities below:

• Live streaming
• Ability to send nude images or videos
• Disappearing content
• Location sharing
• Connecting or talking to strangers
• None of the above
• Other (please specify)

Feature	Responses	Most Supported Age	2nd Most Supported Age	Top theme from free-text comments
Livestreaming	13,374	16 (59%)	18 (33%)	Increases risk of harm to children online.
Nude images and videos	15,370	18 (70%)	16 (17%)	Sending nude images or videos should be completely banned for all ages.
Disappearing content	12,952	16 (51%)	18 (41%)	Should be completely banned due to safety concerns.
Location sharing	12,708	16 (47%)	18 (41%)	Sharing with parents or guardians should be exempt from any ban.
Talking to strangers	14,175	18 (51%)	16 (38%)	Minimum age should be 18, and children should never connect with strangers online.

Note: For more detailed statistics on supported ages, please see the respective feature summaries below. Most respondents to this question provided a numerical age response only. A smaller number also provided additional free-text commentary for each feature (ranging from 3% to 8% of total respondents, depending on functionality). Descriptors used in both the age-only and free-text summaries reflect the proportion of all respondents that raised specific themes. The number of free-text respondents is indicated for each feature.

Livestreaming

13,374 respondents provided age-only responses for the age at which live streaming should be restricted. Many said live streaming should be restricted for those under 16, some said under 18, and a very few respondents suggested other ages, including 13, 14, 15, and 21.

378 individuals provided free-text answers. A very few respondents included additional context, indicating they believed that livestreaming increases the risk of harm to children online, citing concerns including online abuse and exploitation, privacy risks and exposure to inappropriate or harmful content on streams. These respondents highlighted that given difficulties moderating this type of content, the risks of livestreaming outweigh the benefits. A very few respondents said that the content and additional context should be considered as factors when setting a minimum age for livestreaming. A very few respondents also said that parental controls could complement restrictions, for example, by enabling younger children to access this functionality with parental consent.

Minimum Age to Send Nude Images and Videos

15,370 respondents provided age-only responses for the minimum age for the ability to send nude images or videos. Many said the ability to send nude images or videos should be restricted for those under 18, a small number said under 16, a few said under 21 and a very few suggested 25. A very few respondents suggested a complete prohibition of sending nude images or videos for all ages.

1,288 respondents provided free-text responses. A very few highlighted risks of harm, such as exploitation or emotional harm, that the ability to send nude images or videos present for children.

Additional justifications to minimum age selections were also provided. A very few said the minimum age should be 18, to align with existing laws regarding adult content and very few said a minimum age of 16, in line with the age of consent.

Disappearing Content

12,952 respondents provided an age-only response on the minimum age for disappearing content. Around half supported a minimum age of 16, while some supported 18. A very few respondents suggested other ages, including 15, 17, and 21.

641 respondents provided additional free-text commentary. A very few respondents believed that disappearing content should be completely banned due to safety concerns. A very few stated that disappearing content increases risks of harm to children, such as bullying and exploitation, noting that it can be harder to identify and tackle harmful content that is immediately deleted. A very few expressed specific support for a minimum age of 18 or older, while a very few supported 16. A very few expressed uncertainty about what is meant by ‘disappearing content’ and what would be captured by any restrictions.

Location Sharing

12,708 respondents provided age-only responses for the age at which location sharing should be restricted. Some said location sharing should be restricted for those under 16, some said under 18, and few said another age between 13, 14, 15 and 21.

689 respondents provided free-text responses. A very few respondents included additional context, indicating that any age restriction on location sharing should be accompanied with an exemption, specifically to allow children to share their location with trusted contacts, such as parents or guardians.

Very few said that location sharing poses privacy and safety risks for children and that it can facilitate harm such as stalking or bullying. A very few respondents said that there should be a complete ban for all ages, due to privacy and safety concerns. A very few referenced “Any age”, “All ages” or similar terms; however, given the framing, it was not possible to determine whether these respondents intended to advocate for a total ban on the functionality or for no age restriction at all.

Talking to strangers

14,175 respondents provided age-only responses for the age at which connecting or talking to strangers online should be restricted. Around half said the ability to talk to strangers should be restricted for those under 18, some said under 16, very few suggested other minimum ages including 13, 14, 15, 17 and 21.

537 respondents provided free-text responses. A very few respondents said that the minimum age should be set at 18 with added context that children should never be able to connect with strangers online. A very few highlighted that communicating with strangers online poses risks to children such as grooming, exploitation, and scams.

Very few suggested that consideration should be given to the content or context of a service when establishing a minimum age for communicating with strangers. A very few advocated for a parental consent requirement, whereby children would be permitted to connect with strangers online but only with explicit parental approval. It was noted this could provide flexibility in allowing younger children to access this functionality.

Very few referenced “Any age”, “All ages” or similar terms; however, given the framing, it was not possible to determine whether these respondents intended to advocate for a total ban on the functionality or for no age restriction at all.

6.3 Consultation response – question 14

To what extent do you agree or disagree with the following statement: “Restricting children’s access to these features/ functionalities, would provide for a safer online experience for children”. Features/functionalities include live streaming, the ability to send nude images or videos, disappearing content, location sharing and connecting or talking to strangers.

To what extent do you agree or disagree with the following statement: “Restricting children’s access to these features/ functionalities, would provide for a safer online experience for children”. Features/functionalities include live streaming, the ability to send nude images or videos, disappearing content, location sharing and connecting or talking to strangers.	All responses (%)	All responses (total)	Of which parents (%)	Of which parents (total)
Strongly agree	63%	13,531	82%	7,727
Somewhat agree	12%	2,647	8%	768
Neither agree nor disagree	4%	863	2%	153
Somewhat disagree	4%	836	1%	111
Strongly disagree	15%	3,277	6%	603
Don’t know/ Prefer not to answer	2%	427	1%	98
Answered		21,581		9,460
Skipped		2,953		601

6.4 Consultation response – question 15

What do you think the impacts would be if some online services were required to introduce age restrictions on specific features and functionalities? For example, impacts on the safety and wellbeing of children, or the impact for parents and carers, as well as other users. You could also comment on the impact on all users’ privacy and data or on business costs, revenue, and innovation.

Summary of findings:

Some of the respondents reported that age restrictions on specific features and functions would lead to enhanced child safety and wellbeing by reducing exposure to harms such as harmful content or bullying. However, a small number raised concerns related to the effectiveness and enforcement of age restrictions, expressing views that children will find ways to circumvent the restrictions. A small number of respondents highlighted that age restrictions could reduce children’s access to beneficial uses of these features, such as social connection, and push them into riskier, unregulated parts of the internet. The potential impact on adults’ access to these functionalities was also highlighted.

A small number also noted concerns with data collection or breaches and user privacy. A few advocated for a more proportionate and targeted approach to restrictions which considers the risk level of a feature. These respondents considered that interventions should be targeted at high-risk features, although it was noted that some risky features can have benefits, for example location sharing.

A small number of respondents considered that age restrictions on specific features and functionalities would provide positive reassurance for parents, helping to reduce the burden currently placed on parents and carers in ensuring their children have healthy online lives.

In contrast, a few respondents reported that primary responsibility for children’s access to risky functionalities should lie with parents, focusing on oversight and the use of parental controls. A very few suggested that better education for both parents and children would be more effective for improving wellbeing than banning high-risk features and functionalities. These respondents also cited education as key to preparing children to navigate the online world.

Impacts on businesses, especially small businesses, and concerns regarding services choosing to leave the UK market were also highlighted by a small number of respondents as a potential consequence of restrictions on risky functionalities. A very few thought that the impact would drive positive innovation in safety features and hold platforms responsible for providing a safe environment.

Theme Name	Theme Description	Response Share
Enhanced child safety and wellbeing	Age restrictions on specific online features were seen as significantly improving child safety and wellbeing by reducing exposure to harmful content, exploitation, and cyberbullying.	Some
Concern about the wider use of age assurance to enforce minimum age restrictions on risky functionalities	There were concerns about the risks wider use of age assurance, required to enforce minimum age restrictions on risky functionalities, may pose. Notably in relation to user privacy, increased data collection and the risk of data breaches.	A small number
Effectiveness and enforcement challenges	Age restrictions are difficult to enforce, and children will find ways to circumvent them.	A small number
Positive societal and parental outcomes	Age restrictions would enhance children’s wellbeing by encouraging healthy offline activities, which would also improve social cohesion. Restrictions would also provide positive reassurance for parents and reduce the burden on them to provide a healthy online life for their children.	A small number
Disruption to business	Age restrictions would raise compliance costs for businesses, especially smaller businesses, cause them to lose revenue through reduced number of users, and cause some services to leave the UK market at disadvantage to UK users. These concerns are balanced against the benefit in protecting children.	A small number
Preventing access to safe and beneficial features for both children and adults	Age restrictions will push children into riskier, unregulated parts of the internet, prevent them from accessing the benefits of features such as reducing isolation and access to support, and will prevent legitimate adult users from accessing these features too due to privacy concerns.	A small number
Parents should be responsible for their children’s access to risky functionalities	Instead of government intervention on risky functions, it was suggested that primary responsibility for children’s access to risky functionalities should lie with parents, focusing on oversight and the use of parental controls.	Few
Proportionate, targeted restrictions preferred	Age restrictions should be targeted at high-risk features to protect children from these risks whilst allowing both children and adults to enjoy the benefits of low-risk online features. Some concerns were raised about the benefits of some ‘high-risk’ features that children would be prevented from accessing, such as parents tracking their children’s location for safety.	Few
Concerns surrounding the impact of restrictions on freedom of expression	Age restrictions, and adults’ reluctance to undergo age verification, will restrict children and adults’ freedom of expression.	Few
Educating parents and children as a more effective method of improving wellbeing than banning high-risk features	Educating parents and child to increase their understanding of how to avoid risks online will enhance children’s wellbeing whilst avoiding issues with bans, such as enforcement and privacy concerns. Education will also prepare children for negotiating the online world once they are adults.	Very few
Encouraging innovation in platform safety features	Restricting high-risk features would hold platforms responsible for providing a safe environment and would drive positive innovation in safety features and age assurance.	Very few
A blanket ban for children to access social media is the most effective solution	A complete ban on social media for children is the most effective solution to improving children’s wellbeing, rather than targeted age restrictions.	Very few
Other regulated industries should be used as an example for age-restricted access to social media	Other regulated industries, such as sale of alcohol or tobacco, can be an example for how to approach children’s access to social media. This can be both a justification for age-restricted features, or an argument for alternative approaches like blanket bans.	Very few
Other	Sub-themes extracted from ‘other’ included the concern that the list of features and functionalities are too broad in the question - specifically, there are much stronger feelings towards restricting the ‘Ability to send nude images or videos’ whereas there are pros/cons listed regarding ‘Location Sharing’ and ‘Live Streaming’ restrictions. Other sub-themes included a greater emphasis on tacking perpetrators of online crime over regulation, and that restriction should be the role of the tech companies/ service provider.	Very few
No reason given	These responses did not provide substantive answers to the question.	Very few

6.5 Email responses on restricting access to services based on features and functionalities

Summary of responses:

Support for restrictions on risky features and functionalities

• Overall, there was broad support for restrictions on risky features and functionalities. There was recognition that while there are some benefits, such as social connection and creativity, these features can increase the risk of harm to children and young people. However, there were different views on which specific features should be restricted and to what extent. Some caution was expressed about ensuring that any approach government takes can be kept up to date as new risks or features emerge.

Support for a graduated approach

• Across industry and civil society organisations there was support for offering graduated, age-appropriate access to risky features and functionalities instead of blanket restrictions, to balance harm reduction with benefits of these features. Many organisations noted that any restrictions on risky functionalities should be risk-based, proportionate, and grounded in robust evidence. For instance, by restricting a functionality for younger children but providing access, with safeguards, for older children. This was described as more proportionate than blanket restrictions, helping to minimise some unintended consequences such as cliff edges and displacement, and recognising the different developmental needs of children as they grow up.

Other high-risk functionalities

• In addition to the features that the consultation sought views on, a number of respondents raised other features and functionalities they argued were high-risk. This included end-to-end encryption, and on-platform virtual spending and gifting.

Functionalities in context

• Industry stakeholders highlighted that the risk of harm posed by specific features and functionalities can vary based on context and design, with some favouring no restrictions whatsoever. It was noted in responses that functionalities are not necessarily inherently risky, with blanket feature restrictions therefore seen as overly broad and blunt. In recognition of this, the preference of some respondents was for restrictions to be delivered through the existing regulatory framework provided by the Online Safety Act, such as through the use of risk assessments. However, this was not the case for all industry organisations – with some expressing that they did not support restrictions for any functionalities.

Voluntary protections

• Some industry organisations highlighted existing voluntary protections that have already been implemented such as default settings on children’s accounts, parental controls, and age-based restrictions. These were sometimes suggested as tools the government could use to deliver graduated and age-appropriate access to specific risky features.

Impacts on industry

• A range of potential impacts on industry were also raised, including costs associated with additional compliance. Associated costs included increased requirements to deploy age assurance on restricted functionalities or technical updates to a platform to support restrictions. It was noted that these costs would be particularly burdensome for SMEs. A small number of services also highlighted the risk that these costs may lead to services deciding to implement service-level restrictions for children or remove their services from the UK altogether.

Age assurance

• Some respondents highlighted risks around the use of age assurance to enforce restrictions on functionalities and implications for privacy and increased data collection. It was also noted that this could increase friction for adult users as they would also be required to engage with any new age checks.

Innovation

• Some industry organisations highlighted concerns that restrictions may limit the ability or incentives for services to innovate. However, one service noted that a features-led approach may be more likely to support the innovation of age-appropriate design to enable services to provide low risk offerings to young people.

Impact on gaming services

• A number of gaming industry respondents raised concern about an overly broad scope and argued their services should be excluded. One respondent argued that features such as stranger pairing are fundamentally different in video gaming than in social media and carry lower risk. Another respondent noted gaming is subject to the 2026 PEGI reforms, which rate unrestricted online communication PEGI 18 and loot boxes PEGI 16, and argued that this was the right model for targeted, proportionate, feature-level regulation.

Support for restrictions

• Many civil society and child safety organisations expressed support for restricting risky features and functionalities to u16s and, in some cases, to u18s. There was also strong support from a campaign of 33,000 responses to the consultation. One respondent agreed with restricting high-risk functionalities and proposed a tiered approach for 13–15-year-olds and 16–17-year-olds. Another respondent supported age restrictions on risky functionalities at age 16 but noted that video communication features must be treated differently to protect Deaf BSL users who rely on video calls as their primary communication medium.

Harms enabled by risky functionalities

• Many responses suggested that risky functionalities directly enable and amplify various harms to children, including grooming, child sexual abuse, coercion, sextortion, and bullying, and that restricting them would materially reduce children’s exposure to harm.

Functionalities raised

• The most common support was for restrictions on communicating with strangers, disappearing content and the ability to send nude images or videos. There were more mixed views on whether to restrict livestreaming and location sharing. Some respondents suggested additional risky functionalities that should be subject to restrictions, including end-to-end encrypted messaging and financial features such as gifting or spending. Another respondent also noted that algorithmic content feeds pushed content from strangers to children. Organisations argued that restrictions on risky functionalities should be enforced by strong age assurance measures. One respondent advocated for features that enable children to communicate with strangers to be disabled as a default app-level setting and also noted that adults, with the exception of family, should not be watching children livestreaming.

Support for restrictions alongside wider measures

• While supportive of restrictions, many organisations also noted that restrictions would be most effective when coupled with wider changes such as ensuring services are safer by design and enforcing minimum age requirements. One respondent argued that restrictions are most effective when paired with plain-English explanations of the harm patterns they address, on the basis that children are more likely to respect rules when they understand how a feature can be used for grooming, coercion, or exploitation.

7. Chatbots and AI

In this section, we consulted on AI chatbots, including the benefits to children of using AI, as well as the features of chatbots that represent a risk to child users. We asked respondents for their views on whether AI chatbots should have age restrictions, either for access to chatbots generally, or for particular features that represent a risk to children.

7.1 Consultation response – question 26

What are the benefits to children of using AI chatbots? For example, this might include as a search function, for educational purposes, for creativity.

Summary of findings:

Some respondents indicated that they perceived no benefits from children’s use of AI chatbots, citing concerns that the associated risks outweigh any potential advantages. A few respondents expressed the view that existing tools or human interaction are preferable alternatives. A few respondents raised concerns about exposure to misinformation or harmful content, a few raised concerns about overreliance and reduced critical thinking, and a very few raised concerns about broader negative impacts on cognitive development. A very few responses raised risks to academic integrity, a very few raised risks to privacy, data protection, and ethical concerns, and a very few raised risks to wider environmental and societal impacts of AI-driven chatbots.

Conversely, some respondents argued that AI chatbots can support personalised learning and accessibility, with AI chatbots identified as supporting homework, explaining complex topics, and assisting children with additional needs. A small number of respondents also highlighted benefits for creativity and curiosity, a few highlighted benefits to digital literacy and future skills, a very few highlighted benefits of emotional support and a very few of entertainment and engagement.

A few respondents emphasised that any benefits are conditional on supervision and safeguards, while a very few expressed uncertainty and a need for further research. A very few provided no reasoning, and additional concerns were captured in a very few ‘other’ responses, including more acute safeguarding risks and views on regulation and parental responsibility.

Theme Name	Theme Description	Response Share
AI chatbots support personalised learning and accessibility	AI chatbots can assist children with learning, homework, and understanding complex topics by providing quick, accessible, and personalised explanations and support, including for those with additional or special needs, disabilities, neurodiversity, language barriers, or social anxieties.	Some
No or very minimal perceived benefits from AI chatbots	There are minimal or no benefits to children using AI chatbots, as existing resources are seen as sufficient or the risks outweigh any potential advantages.	Some
AI chatbots foster creativity and curiosity	AI chatbots can encourage creativity, imagination, curiosity, and self-expression in children by helping them generate ideas, develop stories, and explore new interests in a supportive environment.	A small number
AI chatbot benefits are conditional on supervision and safeguards	The advantages of AI chatbots for children depend on appropriate adult supervision, robust safety measures, age-appropriate design, and clear guidelines to ensure safe and effective use.	Few
AI chatbots expose children to misinformation and harmful content	AI chatbots may provide inaccurate, misleading, biased, or age-inappropriate information, exposing children to risks if not properly regulated and supervised.	Few
AI chatbots enhance digital literacy and future skills	Using AI chatbots helps children develop digital literacy, critical thinking, technological awareness, and AI literacy, preparing them for future education, careers, and responsible technology use.	Few
AI chatbots increase risk of overreliance and reduced critical thinking	AI chatbots may hinder children’s critical thinking, problem-solving, independence, and creativity by encouraging overreliance and reducing original thought.	Few
Children should use search engines or talk to humans instead of chatbots	Human interaction and typical educational methods are considered superior to AI chatbots for children’s learning, creativity, and social development.	Few
AI chatbots provide emotional and social support	AI chatbots can offer a safe, non-judgemental space for children to express emotions, seek advice, receive companionship, and support mental wellbeing, especially for those who are anxious, isolated, or reluctant to talk to others.	Very few
AI chatbots are detrimental to children’s cognitive development	AI chatbots impede children’s cognitive development and reduces overall intelligence, social, and life skills.	Very few
Uncertainty about AI chatbot benefits and need for further research	The benefits of AI chatbots for children remain uncertain, with calls for more research and regulation before any advantages can be established.	Very few
AI chatbots contribute to environmental and societal harms	AI chatbots may have negative environmental impacts, increase resource consumption, reinforce corporate control, and cause broader societal harms when used by children.	Very few
AI chatbots offer entertainment and engagement	AI chatbots can provide fun, entertainment, and engaging interactive experiences for children, including opportunities for family bonding and motivation.	Very few
AI chatbots raise privacy, data security, and ethical concerns	AI chatbots raise issues around children’s data privacy, security, surveillance, and ethical use, including potential misuse, bias, and environmental impacts.	Very few
AI chatbots undermine academic integrity through misuse and cheating	AI chatbots may enable children to cheat on homework or assignments, misuse the technology, or bypass restrictions, undermining academic integrity and genuine learning.	Very few
Other	Sub-themes extracted from ‘other’ include concerns around AI stimulating maladaptive behaviours in children, specifically around self-harming and suicide. Arguments to regulate ‘Big Tech’ and not the public. Other participants cited government overreach, arguing that parents should be making these decisions.	Very few
No reason given	These responses did not provide substantive answers to the question.	Very few

7.2 Consultation response – question 27

Which AI chatbot features are most risky for children? (Please select all that apply)

• The realism of interactions, including realism of content generated
• The personalisation of interactions
• How they mimic relationships (friendship)
• How they mimic relationships (romantic)
• How they mimic empathy
• Flattering language
• Features to encourage more questions/ requests (e.g. asking questions back)
• The ability to recall interactions across sessions
• The type of content generated – a) video, b) text, c) audio, d) image
• Allowing children to have accounts
• Hallucination or false, misleading responses
• Ability to engage in and generate mature content (e.g. sexual / romantic roleplay)
• Other (please specify)
• None of the above/AI chatbot features are not risky for children
• Don’t know/ Prefer not to answer

Response	All responses (%)	All responses (total)	Of which parents (%)	Of which parents (total)
The realism of interactions, including realism of content generated	74%	15,561	81%	7,456
The personalisation of interactions	69%	14,464	74%	6,888
How they mimic relationships (friendship)	80%	16,864	86%	7,927
How they mimic relationships (romantic)	82%	17,141	86%	7,990
How they mimic empathy	70%	14,707	74%	6,859
Flattering language	72%	15,170	77%	7,143
Features to encourage more questions/ requests (e.g. asking questions back)	59%	12,466	65%	6,046
The ability to recall interactions across sessions	54%	11,331	59%	5,488
The type of content generated – a) video, b) text, c) audio, d) image	63%	13,173	68%	6,308
Allowing children to have accounts	62%	12,987	71%	6,599
Hallucination or false, misleading responses	82%	17,270	85%	7,877
Ability to engage in and generate mature content (e.g. sexual / romantic roleplay)	80%	16,860	87%	8,071
Don’t know/ Prefer not to answer	3%	701	2%	195
None of the above/AI chatbot features are not risky for children	4%	800	3%	264
Other (please specify):	9%	1,985	6%	555
Answered		21,013		9,253
Skipped		3,521		808

7.3 Consultation response – question 28

Which functionalities of AI chatbots should minimum age restrictions apply to?

Summary of responses:

Around half of respondents supported applying minimum age requirements across all AI chatbot functionalities, reflecting a precautionary stance rooted in concerns about children’s safety, exposure to harmful content, and lack of maturity. A very few went further, advocating for bans on AI chatbots altogether due to perceived wider societal and environmental harms.

Within more targeted responses, specific high-risk functionalities were identified. A few identified sexual or romantic content, particularly where chatbots simulate relationships or roleplay interactions, and a few identified anthropomorphic features, or where AI mimics human emotions or behaviours. A few respondents highlighted generative capabilities, such as an AI chatbot’s ability to produce images, audio, video, or specifically deepfakes. A very few cited concerns around personalisation, persistent memory, and data collection, which indicated respondents’ unease about privacy, profiling, and the potential for tailored or persuasive interactions. A very few also emphasised the importance of chatbots clearly identifying themselves as non-human, while a very few focused on company accountability for enforcing safeguards.

Alongside these views, some respondents supported more selective or conditional access. For example, a few argued that educational, factual, or welfare-related uses should remain accessible for children, particularly in supervised or school settings, indicating support for differentiated access rather than blanket restrictions. Conversely, a few opposed additional measures outright or a few preferred alternative approaches such as parental controls and digital literacy, while a very few expressed doubts about effectiveness and enforceability of age restrictions.

Theme Name	Theme Description	Response Share
All AI chatbots should be age restricted	Minimum age requirements should apply to all AI chatbots regardless of functionalities, with high thresholds or complete bans for children to protect against risks such as safety concerns, lack of maturity, and exposure to harmful content.	Around half
Support for restricting sexual or romantic content	To impose age restrictions where AI services produce sexualised or romantic content, including roleplaying relationships.	Few
Support for restricting anthropomorphism	To impose age restrictions where AI services mimic human interactions or present themselves as having emotions, thoughts, or behave in a human-like manner.	Few
Allow educational and factual AI use for children	Access to AI chatbots should be permitted for educational, factual, or welfare-related purposes, particularly in supervised or school environments, while restricting other functionalities for children.	Few
Support for age restrictions on generative AI media	Age restrictions should be applied to AI chatbots that generate images, audio, video, or deepfakes, particularly when these outputs could be explicit, violent, misleading, or facilitate abuse.	Few
General opposition to new age restrictions	New age restrictions are unnecessary, or unjustified, as existing laws, provider-level controls, or platform responsibility are considered sufficient, or age restrictions are seen as impractical.	Few
All AI chatbots should be banned due to wider risks	The use of AI chatbots is believed to contribute to environmental harm and broader societal issues, with some advocating for strong measures such as shutting down companies to mitigate these risks.	Very few
Parental controls and education preferred over age restrictions	Parental controls, supervision, and digital literacy education are considered more effective or appropriate than government-imposed age restrictions for managing children’s access to AI chatbots.	Very few
Support for restricting personalisation	To impose age restrictions on functionalities that personalise responses to children, for example by saying their name, speak in a personal tone, or change output to mirror behaviour.	Very few
Doubts about age restriction effectiveness	There is scepticism regarding whether age restrictions and age verification measures can be effectively enforced and whether they adequately protect privacy, with suggestions that alternative approaches like education or more nuanced controls may be preferable.	Very few
Uncertainty about appropriate AI chatbot restrictions	There is uncertainty or lack of knowledge regarding which AI chatbot functionalities should have age restrictions, indicating the complexity of determining suitable safeguards and the need for clearer guidance.	Very few
Support for restricting persistent memory	To impose age restrictions where AI services maintain information of children between chatbots responses or conversations.	Very few
Support for restricting any aspect of AI which collects children’s data	To impose age restrictions where AI services collect the data of children, for example name, date of birth, or children’s chat inputs.	Very few
Support for tiered age restrictions	Advocated for implementing consistent, tiered, or feature-specific age restrictions for AI chatbots, suggesting categorisation or graduated frameworks to tailor protections based on risk and user maturity.	Very few
Consistent regulation with media age ratings	AI chatbot functionalities should be regulated in accordance with existing content laws and media age ratings to ensure consistency and age-appropriate access.	Very few
Chatbots should have mandatory self-identification	Chatbots must always clearly disclose that they are not human, with explicit warnings to prevent user confusion and manipulation, particularly protecting minors.	Very few
Company accountability for AI chatbot safety	Companies operating AI chatbots should be held responsible for enforcing age restrictions, safeguarding children, and addressing any harm caused by their services.	Very few
Other	Sub-themes surrounding harms caused by the design of the AI, rather than the accessibility of the AI. Concerns that child development is a spectrum and a blanket ban undermines this. Respondents noted that features reinforcing unhelpful thinking patterns for children, or reducing critical thinking should be addressed, rather than a blanket ban. Finally, fears that AI parasocial relationships will reduce childhood appetite for more complex, in-person interactions were raised.	Very few
No reason given	These responses did not provide substantive answers to the question.	Few

7.4 Consultation response – question 29

Should AI chatbots have minimum age restrictions?

Response	All responses (%)	All responses (total)	Of which parents (%)	Of which parents (total)
Yes – minimum age requirements for AI chatbots	35%	7,393	42%	3,908
Yes – restrict access to certain features and functionalities	11%	2,357	13%	1,154
Yes – both minimum age requirements and restricting access to certain features and functionalities	30%	6,332	35%	3,234
No	19%	3,947	8%	775
Don’t know/ Prefer not to answer	4%	903	2%	190
Answered		20,932		9,261
Skipped		3,602		800

7.5 Consultation response – question 30

What do you think the impact would be of introducing age restrictions on AI chatbots or certain features and functions? For example, impacts on the safety and wellbeing of children, or the impact for parents and carers, as well as other users. You could also comment on the impact on all users’ privacy and data or on business costs, revenue, and innovation.

Summary of findings:

Respondents identified a broad mix of benefits, risks, and trade-offs associated with introducing age restrictions on AI chatbots. The most prominent theme, which was raised by some respondents, was perceived improved child safety and wellbeing, where limited access to AI-driven chatbots during childhood was seen to support healthier development and to empower parents and educators. A small number of respondents also highlighted that age restrictions on AI chatbots could promote children’s critical thinking, reduce their overreliance on AI, and a very few highlighted that they could reduce academic dishonesty. A very few noted that restrictions could incentivise safer, child-friendly design and greater industry responsibility.

A small number of respondents also cited concerns about privacy and data risks related to age verification methods, as well as a few raising concerns about limited effectiveness due to circumvention and a few noted that impacts may be minimal or unclear.

A few participants also raised economic and innovation burdens, parental responsibility and education as preferable safeguards, and potential negative effects on education and digital inclusion, alongside a very few who raised broader freedom and government overreach concerns.

Theme Name	Theme Description	Response Share
Improved or enhanced child safety and wellbeing	Age restrictions were seen as improving children’s safety and wellbeing by reducing exposure to harmful content, supporting healthy development, and empowering parents and educators.	Some
Promotion of critical thinking and real-world skills	Restricting AI chatbot access is believed to foster children’s independent thinking, creativity, critical thinking, and real-world social and emotional skills by reducing overreliance on AI.	A small number
Privacy, data security, and surveillance risks - including for adults	Enforcing age restrictions would require intrusive age verification, increasing risks of privacy breaches, data misuse, surveillance, and loss of personal freedom.	A small number
Ineffectiveness and circumvention risks	Age restrictions are likely to be easily bypassed by children, making the measures ineffective and possibly pushing them toward less safe or unregulated alternatives.	Few
Minimal or unclear impact of age restrictions	Some believe age restrictions would have little, unclear, or no meaningful effect due to the evolving nature of AI and children’s adaptability.	Few
Increased business and innovation burden	Age restrictions on AI chatbots would raise compliance costs, create technical and regulatory challenges, and potentially hinder innovation and competitiveness, particularly for smaller UK companies.	Few
Preference for parental responsibility and education	Parental involvement, education, and digital literacy are considered more effective than government-imposed age restrictions for managing children’s use of AI chatbots.	Few
Negative impact on education and digital inclusion	Age restrictions could limit children’s access to educational resources, hinder learning and digital literacy, and worsen educational and social inequalities, especially for vulnerable groups.	Few
General opposition to AI chatbots and age restrictions	A very few advocated for a complete ban on AI chatbots or rejected the necessity of age restrictions, questioning their value in society.	Very few
Government overreach and impact on freedoms	Age restrictions may represent excessive government control, erode personal freedoms, interfere with family autonomy, and burden law-abiding users.	Very few
Unintended negative consequences and digital exclusion	Age restrictions could cause digital exclusion, loss of access to support services, or drive users to unregulated platforms, increasing inequality.	Very few
Support for targeted, proportionate regulation	Clear, proportionate, and targeted rules for AI chatbots or high-risk features were preferred to ensure safe and responsible use, rather than blanket bans.	Very few
Incentivising safer, child-friendly AI and industry responsibility	Age restrictions could encourage the development of safer, age-appropriate AI technologies and promote greater industry responsibility for user safety.	Very few
Reduction in academic dishonesty	Age restrictions are expected to decrease plagiarism, cheating, and misuse of AI for academic dishonesty among children, supporting genuine learning and academic integrity.	Very few
Reduction in AI’s environmental impact	Age restrictions may reduce AI’s environmental footprint by lowering energy and resource consumption, but could also increase infrastructure demands.	Very few
Need for further research and balanced regulation	More research, ongoing monitoring, and nuanced regulation are necessary to protect children effectively without imposing unnecessary restrictions.	Very few
Other	Concerns surrounding economic sanctions or repercussions from nations where these companies are based, but feelings that this would be ‘worth it’. Other sentiments captured feelings of reconnecting with the physical world and real people.	Very few
No reason given	These responses did not provide substantive answers to the question.	Few

7.6 Email responses on chatbots and AI

Summary of responses:

Benefits of chatbots

• There was a widely held view across industry and civil society that AI chatbots offer significant benefits to children.
• Many respondents highlighted the educational benefits of chatbots. Some organisations noted that chatbots can make education more accessible for students with special educational needs or disabilities. Industry stakeholders flagged chatbots can help with learning and a civil society stakeholder responded that some students may feel more comfortable asking a chatbot for assistance compared to an adult.
• Several responses also highlighted that chatbots can be beneficial to children’s digital literacy. Submissions outlined how AI chatbots can help develop digital literacy, encourage an interest in STEM subjects, and prepare children for the use of chatbots in the world of work.
• Some civil society respondents noted that chatbots can offer children emotional support. Stakeholders noted that chatbots can help vulnerable children feel less lonely and provide support during difficult periods.
• Industry respondents also highlighted the benefits of chatbots for creativity and play. Respondents said chatbots can support storytelling, writing, idea generation, and hobbies, and can also be used as part of gaming and play.

Harms of chatbots

• Respondents also raised concerns around risks and harms to children from AI chatbots.

Harmful or age-inappropriate content

• Responses from civil society and industry raised concerns that chatbots can expose children to harmful or age-inappropriate content, including self-harm and suicide material, violent content, and harmful dieting advice. Stakeholders said chatbots can provide inaccurate, misleading, or hallucinatory responses, which children may trust too readily. This is risky when chatbots are used for advice on topics around health, safeguarding, relationships, or other sensitive issues. Stakeholders also raised risks around AI generated misinformation and deepfake content.

Grooming and child sexual abuse material

• Respondents also raised concerns around grooming and child sexual abuse material (CSAM). Romantic or sexual roleplay and grooming-like dynamics were flagged by stakeholders. One stakeholder reported that 50% of 59 companion chatbot platforms it tested displayed sexualised content on their landing page with no age gate. Concerns around chatbots creating CSAM were raised by many responses.

Features and functionalities

• Many respondents also noted features & functionalities of chatbots they argued are harmful to children. Both industry and civil society respondents raised concerns about emotional and psychological harms associated with some features offered by chatbots: for instance, which mimicked empathy, friendship, or romance. Civil society groups went further and noted issues with children developing emotional reliance on chatbots as a result of some features, including sycophantic behaviour, persistent memory and features designed to prolong engagement. These risks were argued to be especially prominent for vulnerable children. Some respondents highlighted how the embedding of chatbots within mainstream messaging platforms introduces new risk into services that are already widely used by children.

Privacy and data protection risks

• Respondents also raised privacy and data protection risks. Stakeholders warned children may share sensitive personal information with chatbots without understanding how that data is retained. Civil society stakeholders also heard concerns from children and young people about AI chatbot collection and retention of personal data, and the uncertainty around its future impacts.

Proposed mitigations

• Industry and civil society stakeholders proposed a range of mitigations to address harms to children. These are listed below.

• Minimum age requirements: A wide range of non-industry stakeholders supported risk-based minimum age requirements for access to chatbots. Some stakeholders advocated for a complete ban for under-13s on higher-risk services such as companionship chatbots, with older children being able to access services with protections in place. There were also calls for companion chatbots to not be accessible for under-18s with some advocating for a chatbot ban for under-16s.

• Restricting high-risk functionalities: Industry and civil society argued regulation should focus on higher-risk functionalities. This included the simulation of emotions, persistent memory, romantic or sexual roleplay, and features that encourage prolonged engagement.

• Restricting sensitive content: Both civil society and industry stakeholders supported restrictions on chatbots discussing sensitive topics. These responses included suggestions such as mental health, self-harm, and suicide, as well as sexual content.

• Transparency and signposting: Civil society stakeholders and industry proposed chatbots include clear reminders to users that they are interacting with AI. Some stakeholders called for chatbots to be required to generate mandatory wellbeing notices when users have been using chatbots for long periods. Stakeholders supported chatbots pointing children to appropriate services when they raised sensitive topics. One stakeholder called for watermarks to be placed on AI generated content to help users understand what content is AI generated.

• Safety by design: Civil society stakeholders called for chatbots to be subject to a broader regulatory regime requiring risk assessments and mitigations for identified harms. Stakeholders called for this to be achieved by bringing chatbots comprehensively into scope of the OSA. Industry stakeholders supported a risk based regulatory framework, noting this approach would provide certainty for businesses whilst also ensuring children and young people were protected regardless of any new technological developments or features that could emerge in the future. Alignment with existing AI governance and regulation frameworks

• One response called for the UK to implement proposals that align with The Council of Europe Framework Convention on Artificial Intelligence and Human Rights, Democracy 8 and the Rule of Law. The Convention requires countries that have ratified the framework (the UK has signed it) to ensure that AI systems align with principles that – among other things – require systems to operate safely and predictably. In addition, another stakeholder argued that all chatbots should adhere to the Department for Education AI Product Safety Standards for AI tools provided in education.

Further proposals

• Additionally, respondents proposed a range of further proposals to mitigate harms to children from chatbots. One stakeholder argued that the government should direct the ICO to extend its Children’s Code review to cover companion AI platforms. Some stakeholders also called for GenAI tools to be brought fully into scope of the OSA.

• One stakeholder argued that educating children on how chatbots function should be a key safeguard. They argued this would empower children to treat AI outputs as probabilistic and fallible rather than authoritative and to critically evaluate advice from AI chatbots.

• Another stakeholder called for independent pre-deployment safety testing and ongoing monitoring of AI models as a systemic requirement, and a further stakeholder proposed establishing a cross-sector AI regulator.

Exemptions and implications for businesses

• Industry and civil society stakeholders raised the implications for businesses of measures in this space. Industry stakeholders argued regulation should be tightly targeted at higher-risk chatbot services, rather than applying to all AI-enabled products or conversational tools. Industry respondents also warned broad legal or governmental definitions of chatbots or blanket regulations applied to all types of chatbots, rather than specific high-risk chatbots, could discourage or hinder chatbot innovation and testing.

• Stakeholders also suggested there could be market impacts in the gaming sector if regulations are too broad. They noted that restrictions affecting common in-game communications features could make some products less viable in the UK market.

8. Improving age assurance

In this section, we consulted on the effectiveness of age assurance (AA) systems in enforcing digital age restrictions. We asked about the impacts of increasing age checks online, including for adults, noting privacy and business costs. We also consulted on whether the public thinks that all users should have to use age assurance if it supports child online safety. We asked about how AA could be made more effective and workable, and what should be considered when reviewing the effectiveness of AA technologies.

8.1 Consultation response – question 31

To what extent do you agree with the following statement: “Adults should complete age checks more often, if it means children are safer online”?

To what extent do you agree or disagree with the following statement: “Adults should complete age checks more often, if it means children are safer online”	All responses (%)	All responses (total)	Of which parents (%)	Of which parents (total)
Strongly agree	53%	11,645	74%	6,963
Somewhat agree	10%	2,069	10%	982
Neither agree nor disagree	3%	665	2%	219
Somewhat disagree	3%	635	2%	161
Strongly disagree	30%	6,513	11%	1,011
Don’t know/ Prefer not to answer	2%	329	1%	105
Answered		21,856		9,441
Skipped		2,678		620

8.2 Consultation response – question 32

What should be considered to make minimum age restrictions effective and workable? This could mean either age restrictions for access to whole services, or for specific risky or ‘addictive’ features or functionalities.

Summary of findings:

Some respondents noted that, for minimum age verification methods to be workable, they should be underpinned by systems that are reliable, secure and privacy-preserving, minimising data collection and ensuring that marginalised groups are not excluded.

Some respondents raised significant concerns about the effectiveness of age verification, questioning whether such measures can be sustained given children’s ability to circumvent them. Concerns were also raised about the potential negative impact on users’ rights to data privacy, including the transfer of personal data to companies based overseas, and about the risk of government overreach.

A small number of respondents emphasised that any age verification framework should include robust measures to prevent circumvention.

A small number of respondents raised how education and online literacy will be equally or more important in keeping children safe than enforcing online age verification. Respondents noted that tools include the proper use of existing parental controls to restrict device usage and these respondents tended to call for parental responsibility for children’s internet use to ensure compliance. A very few advocated for hardware-level enforced restrictions for children rather than online age verification.

In relation to enforcement, a few called for platforms to be made legally more accountable for failings to protect children online. A small number of respondents underlined how any age verification system must be evidenced and targeted to the highest risk features to children on platforms.

A very few raised the issue of how this is being handled globally, and a very few raised the need for technology experts to regularly review policy.

Theme Name	Theme Description	Response Share
Concerns about effectiveness and unintended consequences of age restrictions	There are significant concerns about the effectiveness of age restrictions, ease of circumvention, privacy risks, government overreach, and potential negative outcomes.	Some
Age verification must be robust, privacy-preserving, and accessible	Effective minimum age restrictions require reliable, secure, and privacy-preserving age verification systems that minimize data collection and avoid excluding marginalized groups.	Some
Age restrictions should be for whole services and limit circumvention	Age restriction systems should be for whole services and should be designed to prevent circumvention, while supporting accessibility and innovation.	A small number
Minimum age thresholds should be clear, proportionate, and risk-based	Age restrictions must be based on evidence, targeted to high-risk features to avoid unintended consequences.	A small number
Education and digital literacy are essential for compliance	Comprehensive education and awareness campaigns for parents, children, and the community about online risks and age restrictions support safe online behaviours.	A small number
Platforms must be legally accountable for enforcing age restrictions	Strong legal and regulatory frameworks, robust enforcement, and meaningful penalties are necessary to ensure platforms uphold effective age restrictions.	Few
Policies must be regularly reviewed and adapted with expert input	Age restriction policies and enforcement mechanisms should be updated and audited in consultation with experts to address technological changes and emerging risks.	Very few
International cooperation is helpful for effective enforcement	International cooperation, harmonised standards, and consideration of global legal frameworks are helpful to address cross-border challenges in age restriction enforcement.	Very few
Other	These responses centre around themes of parental responsibility and alternatives to age verifying adults such as device-based controls and permissions in hardware for children. Other sub-themes included government overreach, and tech companies’ interests and profits should not be prioritised over safety.	Very few
No reason given	These responses did not provide substantive answers to the question.	Few

8.3 Consultation response – question 33

What do you think the impacts might be from requiring age assurance across a greater number of online platforms? For example, impacts on the safety and wellbeing of children, or the impact for parents and carers, as well as other users. You could also comment on the impact on all users’ privacy and data or on business costs, revenue, and innovation.

Summary of findings:

The largest emerging theme from this question was related to the impacts of age assurance requirements on privacy and civil liberties. Some responses highlighted the potential link between requiring age assurance and increasing how much and how often personal data is collected, to privacy breaches, cyber-attacks, and the potential for this personal data to be exploited by malicious actors.

A small number of respondents raised that verification can be undermined by children using methods to circumvent it, alongside the possibility of displacing them to unregulated platforms. Similarly, a very few questioned how age verification would realistically be enforced and managed, given the limitations and vulnerabilities of existing methods.

A small number of respondents raised the possibility of businesses no longer operating in the UK as a probable impact and a very few noted the potential for platforms to receive reduced revenues.

Some respondents positively linked requiring more platforms to age assure to benefits to children’s online safety, although many of these responses also mentioned issues of friction and data privacy. Additionally, a few referenced changes in the role and responsibilities of parents if less children use online platforms.

A small number of respondents thought that more age assurance may lead to digital exclusion for marginalised adults and children being unable to access online resources and communities for issues such disability, mental health, domestic abuse, or for LGBTQ issues, alongside increased friction and reduced convenience for adults.

Theme Name	Theme Description	Response Share
Heightened privacy and civil liberties risks	Mandatory age assurance is expected to increase personal data collection, risk privacy breaches, erode anonymity, enable surveillance, and threaten civil liberties, especially for vulnerable users.	Some
Improved child safety and parental confidence	Expanding age assurance is anticipated to enhance children’s safety and wellbeing by reducing exposure to harmful content.	Some
Increased digital exclusion, friction and user inconvenience	Age assurance requirements are likely to create barriers for adults, low-income, and vulnerable users who lack identification or digital resources, leading to exclusion and reduced platform usage.	A small number
Higher business costs and reduced market competition	Age assurance is expected to increase compliance costs and administrative burdens for businesses, especially smaller platforms, potentially stifling innovation, reducing competition, and causing market withdrawal.	A small number
Risk of circumvention undermining effectiveness or displacement to less regulated platforms	Age assurance measures may be easily bypassed, particularly by children, reducing their effectiveness or potentially driving users to less regulated platforms.	A small number
Parents would need to take more responsibility	Parents would need to take more responsibility and have more oversight over children’s online use.	Few
Minimal or uncertain impact expected	Some anticipated that age assurance will have little or uncertain impact, especially if implementation is well-targeted and users adapt over time.	Very few
Implementation and enforcement difficulties	Age assurance faces significant technical, logistical, and legal challenges, including limitations of current technologies.	Very few
Loss of revenue for services	Age assurance may lead to less revenue for platforms.	Very few
Children should retain platform access	Some believe children should have unrestricted access to online platforms and that age assurance restrictions are unnecessary.	Very few
Children’s data better protected	Age assurance proposals may decrease the amount of data collected from children by platforms, lowering risks associated with data harvesting and privacy breaches for minors.	Very few
Trial period and ongoing evaluation needed	A trial period and continuous assessment are recommended to evaluate the effectiveness and impacts of age assurance before full implementation.	Very few
Other	Themes include censorship and freedom impacts, government overreach, discussion of how children will experience different childhoods, and how technology companies will have less influence, and should take more responsibility. A loss of civil liberties and anonymity online, including the impact on protests was also raised.	Very few
No reason given	These responses did not provide substantive answers to the question.	Very few

8.4 Consultation response – question 34

How, if at all, could age assurance be made more effective?

Summary of findings:

Responses suggest no single solution dominates, with a mix of measures suggested. A small number emphasised parental responsibility as the best solution, highlighting the need for greater supervision, education, and stronger parental controls alongside technology.

Scepticism about effectiveness of age assurance remains notable with a small number of respondents citing concerns about circumvention and limited benefits, while a small number of respondents expressed outright opposition, with these responses often noting an aversion to digital IDs. Technical solutions received support, including a small number who supported using official documents, but a small number were also interested in prioritising privacy and data protection, suggesting a desire for secure, low intrusion systems.

Structural approaches also featured. A small number of respondents called for stronger regulation and platform accountability, and a few favoured government-led or national ID systems. A few suggested more layered approaches and advocated for combining verification methods. A few respondents also suggested other approaches, including inbuilt device or network-level controls and improved education and digital literacy.

Theme Name	Theme Description	Response Share
Emphasising parental responsibility and controls	Parents should play a central role in age assurance through supervision, education, robust parental controls, consent, and legal accountability, with technical measures as complementary.	A small number
Expressing scepticism about the effectiveness of age assurance	There is doubt regarding the effectiveness, enforceability, and technical feasibility of age assurance, with concerns about circumvention and limited safety benefits.	A small number
Using official documents for age verification	Effective age assurance requires the use of official documents (e.g. photo ID, passports, birth certificates).	A small number
Prioritising privacy and data protection	Age assurance systems must minimise data collection, reduce friction, and use privacy-preserving solutions to prevent surveillance, breaches, or misuse.	A small number
Opposing age assurance	Opposition reflects concerns that age assurance is unnecessary or undesirable, often linked to risks of government overreach or impacts on civil liberties.	A small number
Strengthening regulation and platform accountability	Effective age assurance requires robust regulation, standardisation, independent oversight, regular audits, and meaningful penalties to hold platforms and service providers accountable.	A small number
Delivering age assurance through a government led service, national ID or digital passport	Age assurance should be delivered through centralised government run solutions (e.g. national ID or digital passports) rather than third party providers, to reduce data breach risks and ensure high, standardised levels of assurance.	Few
Implementing multi-layered age verification methods	Improve effectiveness of age verification checks by deploying multi-layered age verification methods at once (e.g. photo ID and age estimation).	Few
Applying device, platform, and network-level controls	Age assurance can be strengthened by implementing controls at device, app store, operating system, router, or ISP/network level, with built-in restrictions and parental management to prevent circumvention and reduce data sharing.	Few
Enhancing education and digital literacy	Education and awareness campaigns for parents, children, and the public about online safety, digital literacy, and age restrictions are essential complements to technical solutions.	Few
Preventing circumvention and identity fraud	Measures such as blocking VPNs, detecting identity fraud, and ongoing verification are necessary to maintain the effectiveness of age assurance systems.	Few
Offering flexible and alternative verification methods	Providing less invasive, creative, or flexible verification options including self-declaration, credit card checks, parent-led or community-driven verification, and platform design changes can accommodate diverse users and avoid exclusion.	Very few
Raising age thresholds and implementing bans	Increasing age thresholds or restricting access for under-16s or under-18s, including outright bans on social media, devices, or apps for children, especially in schools, is advocated to enhance safety.	Very few
Leveraging expert input and continuous improvement	Effective age assurance requires ongoing research, evaluation, and adaptation as technologies and risks evolve, alongside the involvement of technical experts and industry specialists to design, assess, and refine robust, up-to-date solutions.	Very few
Adopting risk-based and layered approaches	Age assurance systems should combine multiple methods and focus stronger checks on high-risk services, while minimising friction and data collection for low-risk contexts.	Very few
Learning from international and cross-sector examples	Drawing on experiences from other countries and sectors like banking or passports can help identify best practices and avoid known pitfalls in age assurance.	Very few
Addressing accessibility and risk of exclusion	Strict age assurance requirements may unfairly exclude disadvantaged users or those without access to required documents or technology, risking digital exclusion and negatively impacting user experience.	Very few
Other	Sub-themes raised under ‘other’ include tech companies being held responsible, the importance of parental involvement, and promoting age assurance as a positive, rather than a negative restriction to change behaviours.	Very few
No reason given	These responses did not provide substantive answers to the question.	Few

8.5 Consultation response – question 35

What should be considered when assessing the effectiveness of age-verification and age-assurance technologies?

Summary of findings:

The most prevalent theme related to data protection and privacy, which some respondents cited as a key consideration when assessing the effectiveness of age verification and age assurance technologies. This included concerns about where personal information is stored and how privacy and security are maintained.

Some respondents referenced the risk of circumvention, with a number of these responses highlighting the relative ease with which existing verification measures can be bypassed.

A small number of respondents identified the need to consider alternative approaches to age verification and age assurance. Examples cited included parental consent mechanisms, allowing parents or guardians to determine age of access, and the use of parental controls.

In addition to responses identifying factors to consider when assessing the effectiveness of these technologies, a small number of respondents expressed opposition to the use of age verification and age assurance altogether. Within these responses, concerns were frequently raised about the risk of data leaks or personal data being compromised. A further theme referenced by respondents was the case for regulating content itself, rather than restricting access by age.

Theme Name	Theme Description	Response Share
Accurate and effective, with rigorous privacy, data protection, and security	Age-verification systems must accurate and effective, while rigorously protecting user privacy and data, minimising collection and retention, ensuring robust cybersecurity, complying with data protection laws, and safeguarding against misuse, breaches, and surveillance risks.	Some
Effective, accurate, and circumvention-resistant systems	Age-verification technologies should reliably prevent underage access, minimize errors, and be robust against circumvention and manipulation across all demographics.	Some
Opposition due to civil liberties and privacy concerns	Critics expressed scepticism or opposition to age-verification, citing concerns about necessity, effectiveness, privacy, surveillance, government overreach, and impacts on civil liberties and free speech.	A small number
Alternative and complementary approaches including parental involvement	Non-technological measures such as parental controls and responsibility, education, device-level controls, and broader safety strategies should be considered alongside or instead of technological age-verification, with parents playing an active role.	A small number
Prioritising child safety and harm reduction	The main objective is to protect children from harmful content and online risks, with effectiveness measured by reductions in harm, improvements in wellbeing, and a focus on children’s mental health.	Few
Proportionality and risk-based approach respecting civil liberties	Measures should be proportionate to the risk posed, ethically justified, avoid excessive intrusiveness, respect civil liberties and lawful adult access, and balance child protection with user rights.	Few
Ensuring usability, accessibility, and fairness	Age-verification technologies should be user-friendly, accessible to all including those with disabilities or lacking standard ID, avoid undue burdens or discrimination, and be independently tested for fairness and inclusivity.	Few
Transparency, oversight, and accountability required	Age-verification systems must operate with transparent processes, independent government-linked oversight, regular audits, and clear accountability to ensure compliance and public trust.	Few
Consideration of technical and alternative methods	The assessment should include biometrics, privacy-preserving solutions, and the feasibility of alternative or complementary verification systems.	Very few
Legal and regulatory compliance required	Age-verification systems must comply with relevant laws and regulations, including GDPR and platform accountability, and assign clear responsibility and liability for breaches or misuse.	Very few
Cost-effectiveness and economic feasibility	Implementation should be cost-effective, scalable, and feasible for all platforms, avoiding disproportionate burdens especially on small businesses and considering broader economic implications.	Very few
Continuous evaluation and adaptability needed	Age-verification systems should undergo ongoing monitoring, independent review, real-world testing, and evidence-based updates to remain effective and address emerging risks.	Very few
Stakeholder and expert engagement necessary	Design and assessment should involve input from a broad range of stakeholders including children, parents, privacy specialists, and independent experts to ensure practical and trusted solutions.	Very few
Support for national digital ID	Advocates endorsed the introduction of a national digital ID or ID card to facilitate secure and effective age verification.	Very few
Learning from international approaches and comparable regulatory models	Design and assessment of age verification technologies should draw on lessons from international implementation and past evidence, as well as comparable regulatory approaches (e.g. alcohol, smoking, and gambling), while accounting for differences in legal, cultural, and global digital contexts.	Very few
Other	Sub-themes encouraged focusing on content rather than age, highlighted the benefits of social media, the challenges of regulating AI, and the risks of outsourcing age assurance outside the UK.	Very few
No reason given	These responses did not provide substantive answers to the question.	Few

8.6 Email responses on improving age assurance

Summary of responses:

Proportionality

• Respondents recognised that AA was important for providing age-appropriate digital experiences but considered that it needs to be deployed alongside wider safety measures and is not a ‘silver bullet’ or standalone solution.

• Respondents agreed that AA should be proportionate to the risk of harm. There was consensus that AA benefits need to be balanced with trade-offs, particularly privacy and user friction. Many stakeholders suggested that lower-risk services should be able to use lower-confidence and less invasive AA measures, whereas higher-confidence measures such as HEAA should be targeted at platforms or services with the highest levels of risk.

• Some organisations pointed to evidence suggesting that parents are supportive of adults doing more age checks if it leads to improved child safety online. Some gave a view that more age checks are justified if children are made safer. One stakeholder disputed the presumption that more checks will automatically produce better child-safety outcomes, and many individuals opposed mandatory age checks. Others stated that a focus on ensuring services are safe by design is preferable to relying on age checks.

Privacy-preserving

• Significant concerns were raised from civil society, industry, regulators and individuals about the privacy implications of AA and the potential for data breaches. Many responses referenced a widely publicised data breach as an example. Several organisations presented these risks as a trade-off for government to be aware of if increasing the use of age checks. There was an emphasis that AA use should be consistent with UK data protection legislation, should avoid excessive data collection, and be transparent in data handling practices.

• Respondents also highlighted the potential impact on children and young people’s data and privacy, as well as on adults.

Centralised age-assurance

• Stakeholders were generally supportive of device, operating system, or app store-level AA to complement platform-level AA. This would allow a user, or a parent or carer, to verify their age on their device or operating systems, and for an age token to be generated that could then be shared as the user accesses different apps and services on their device.

• Key cited benefits were that this would reduce friction and be easier to use, particularly for parents and carers, this was a view held by industry and civil society. Stakeholders also felt this would have privacy benefits, reducing the need to share personal information with multiple services, and enabling APIs and Zero-Knowledge Proofs (ZKPs, a cryptographic method that allows one party to prove age without providing underlying data). Industry noted that this would also benefit developers.

• There was the recognition that device or operating system-level AA could support wider child safety interventions, particularly device-level nudity detection.

• Some industry stakeholders noted there is not a consensus within the sector on where AA responsibility should sit, and some tech companies caution against device or operating system-level AA, instead recommending a system-wide approach. However, there was strong support for device, operating system, and app store-level AA from civil society and industry providers of online services.

• There was support from a range of respondents for interoperable AA solutions, including digital wallets, standardised APIs, and use of ZKP to reduce friction and compliance costs.

Accessibility, inclusion and future methods

• There was consensus that no single AA approach is sufficient for all users. Stakeholders noted that, for AA to be accessible, users should have multiple ways to verify their age, and requiring use of photo ID alone should be avoided, as photo ID is not universally accessible and could disadvantage users from lower socio-economic backgrounds. Some stakeholders also raised concerns about the accuracy and effectiveness of age assurance methods like facial age estimation for people from minority ethnic backgrounds and for users with disabilities.

• Multiple stakeholders noted the benefits of a layered approach to AA, building understanding of users’ ages through multiple data sources. In particular, industry highlighted use of age inference models to assess user behaviour over time, noting benefits including lower user friction, reduced personal data processing, and improved accessibility.

• Stakeholders stressed the need for regulation to remain future-proofed, to allow for evolving methods.

Clarity on expectations of AA

• Stakeholders called for clear standards for what constitutes effective AA, to support compliance by clarifying best-practice expectations, and to improve user trust and uptake of AA.

• Responses suggested that standards should cover accuracy, reliability, transparency, robustness and accessibility. Stakeholders felt that effectiveness should look at ongoing, system-wide deployment of AA in practice, rather than accuracy of specific tools in isolation.

• Some, particularly civil society, called for numerical thresholds of accuracy and certification schemes to demonstrate compliance with standards. There was support for further transparency on service’s use of AA to help understand effectiveness.

• Industry and civil society noted the importance of regulatory consistency, calling for future requirements to align with the OSA, alongside clarity on the roles and responsibilities of Ofcom and the ICO.

• Some called for clearer standards on whether users should be allowed multiple attempts to complete age checks.

Burden on services and industry engagement

• Stakeholders noted the cost on services of AA processes, particularly for smaller or low-risk companies, with some noting that this risks an increased dominance of big tech companies.

• Stakeholders emphasised the need for future requirements to be informed by evidence and industry engagement to ensure future requirements are technically feasible. Some highlighted Ofcom’s upcoming statutory report on age assurance as an important source of evidence.

9. Respondent breakdown

The full consultation collected only a small amount of personal information about respondents. Breakdowns are displayed below.

Are you answering as a private individual or on behalf of an organisation?	Response Percent	Response Total
Private individual	97%	23,102
On behalf of an organisation	3%	707
Answered		23,809
skipped		725

Please specify organisation type	Response Percent	Response Total
Business/trade association	13%	89
Civil society/third sector/community organisation	34%	236
Education setting	31%	211
Prefer not to answer	1%	10
Other (please specify):	21%	143
Answered		689
Skipped		23,845

Do you have a parental or caring responsibility for at least one child who will be 21 or younger by the 26th of May of this year (2026)?	Response Percent	Response Total
Yes	44%	10,061
No	48%	10,944
Prefer not to answer	9%	2,056
Answered		23,061
Skipped		1,473

As part of your current occupation, do you work with children aged 21 or younger in an education setting? (for example, as a teacher)	Response Percent	Response Total
Yes	68%	16,236
No	26%	6,142
Prefer Not To Answer	6%	1,349
Answered		23,727
Skipped		807

Response	Response Percent	Response Total
In the UK	97%	23,056
Outside the UK	1%	328
Prefer Not To Answer	1%	281
Answered		23,665
Skipped		869

10. Data quality

A consultation is intended to collect the range of views and insights into proposed government policy. It does not measure levels of public support for a policy option.

To maximise accessibility, the full consultation did not use any measures to prevent multiple responses. Instead, officials monitored responses for indications of coordinated responses, for example – inexplicably high numbers of responses with the same IP address, an unexplained spike in responses, or long and detailed responses completed in unduly short time periods.

No examples of coordinated inauthentic responses were identified. 93% of responses came from a unique IP address. 99% of responses came from IP addresses with only 1-3 responses. The peak number of responses from a single IP was 18, an outlier, and the largest examples related to schools.

There were no significant unexplained spikes in responses on any day, nor large numbers of responses completed unduly quickly.

11. Annex A - Campaign email response

Dear Secretary of State,

Please treat this email as my response to your consultation on growing up in the online world. I am responding as an individual.

I want to thank the Government for making a binding commitment to Parliament, and to parents and children across the country, that whatever else you may do in response to this consultation, you will now definitely introduce age or functionality based restrictions for under 16s, and that by no later than July you will set out your plan for making this happen.

This is exactly what I would like to see happen, as soon as possible. It is a relief to know that the Government is willing to stand beside the millions of parents who want to protect their children from daily exploitation by the big tech companies. Safe design must be the price paid by tech companies to access our children in the online world.

Your consultation paper suggests a range of features and functionalities - such as algorithmically-driven content - that social media, gaming and other similar platforms could be prohibited from using in services they offer to our children.

This also includes livestream broadcasts, location sharing, stranger-pairing, disappearing messages, infinite scrolling, autoplay, public tokens of affirmation, alerts and push notifications.

I strongly support including them all within scope of the measures that you have committed to introduce. You should also include penalties for non-engagement, location tracking, and AI features that pose risks of attachment or manipulation for children, as recommended by The Joint Statement of Principles for Furthering the Protection of Children Online signed by 41 leading children’s organisations including the NSPCC, the Centre for Young Lives and Parentkind, and 27 bereaved parents who have lost children to these platforms.

We heard your promise and we expect you to deliver on it, no ifs or buts: raise the age for harmful social media features to 16.

Yours sincerely,

12. Annex B - List of organisations represented by respondents

• 123 Org
• 5Rights Foundation
• Aberlour Children’s Charity
• Academy of Medical Royal Colleges
• Action for Children
• Action on Smoking and Health, the Alcohol Health Alliance and the Obesity Health Alliance
• Activision Blizzard
• Ada Lovelace Institute
• Adapt Inc. Research Labs
• Adolescence, Mental Health and the Developing Mind Initiative
• Adult Child Health and Environmental Support (ACHES)
• Advance
• Advanced Psychotherapy & CBT (trading name of Barnsley Hypnosis & Counselling)
• Advertising Association
• Age Check Certification Scheme
• Age Verification Providers Association
• AI Objectives Institute
• Akifu Toys
• Albany Primary School
• Alcohol Change UK
• Alcohol Focus Scotland
• Alcohol Health Alliance UK
• Alder Hey Children’s NHS Foundation Trust
• All Party Parliamentary Group for Digital Creators
• All Saints School
• Amazon UK (Amazon UK Services Ltd.)
• Ambitious About Autism
• Amersham School
• Amnesty International
• Analigner Partnership
• Animal Sentience Committee
• ANME Ltd
• Anna Freud
• Anthropic
• Anti-Bullying Alliance
• APPG on Wellbeing Economics
• Applied Comics Etc
• Ardleigh Green School
• Ards & North Down Policing & Community Safety Partnership
• ARGA Observatory
• Arnold View Primary School
• Arts for Labour
• ASCL
• ASSIST NI
• Association for Competitive Technology (ACT)
• Association for Young People’s Health (AYPH)
• Association of Directors of Children’s Services Ltd. (ADCS)
• Association of Lawyers for Children (ALC)
• Aughton Parish Council NE Ward
• Aurora Services Group Ltd
• Autism in Affinity Spaces Project (Queen Mary University of London)
• Ayrshire and Arran Health Board
• Backlash
• Bad Rabbit Security Limited
• Banstead Community Junior School
• Barnardo’s
• Basis Yorkshire
• Bassetts Farm Primary
• Baycroft School
• BCS, the Chartered Institute for IT
• Be Free Campaign
• Beat Eating Disorders
• Bedford Hindu Society and Community Trust
• Belfast City Youth Council
• Belmont Juniors
• Bentley Wood High School
• Berkshire Youth
• Bestway International Christian Centre
• Big Brother Watch
• Bite Back 2030
• Bluebell Primary School
• Bolder Academy
• Bolton Methodist Churches
• Bolton School
• Boney Hay Primary Academy
• Boundary Oak School
• BPS
• Brainwaves / Dementias Platform UK, Department of Psychiatry, University of Oxford
• Breck Foundation
• Brianna Ghey Legacy Project
• Bridgewater Primary School
• British Association for Behavioural & Cognitive Psychotherapies
• British Association of Social Workers (BASW)
• British Computer Society
• British Film Institute (BFI)
• British Medical Association (BMA)
• British Psychological Society
• British Screen Forum
• British Sign Language Advisory Board
• Brook
• BSI
• BT Group
• Buckingham School
• Buxton Sea Cadets
• CAMHS
• Canterbury College
• Capernwray Hall
• Cardiff Council / Cardiff University
• Cardiff Steiner School
• Cardiff Youth Service
• Carers Trust
• Castleford Academy
• Caterham School
• Catmose College
• CEASE (Centre to End All Sexual Exploitation)
• Center for Countering Digital Hate (CCDH)
• Centre for Information Policy Leadership (CIPL) at Hunton Andrews Kurth LLP
• Centre for Policy Research on Men and Boys
• Centre for Protecting Women Online, The Open University
• Centre for Research on Outdoor Play, University of Exeter
• Centre for Technology and the Body
• Centre for Young Lives
• Centre of Expertise on Child Sexual Abuse
• Chamber of Progress
• Channel 4
• Cheam High School
• Child Protection Committees Scotland
• Child Rights International Network (CRIN)
• Child Safeguarding Practice Review Panel
• Childlight - Global Child Safety Institute
• Childnet
• Children & Young People’s Mental Health Coalition
• Children and Young People’s Commissioner Scotland (CYPCS)
• Children and Young People’s Mental Health Coalition
• Children First (Protecting Scotland’s Children)
• Children in Northern Ireland (CiNI) in partnership with NSPCC Northern Ireland
• Children in Scotland
• Children in Wales
• Children North East
• Children’s Legal Centre Wales / Swansea University CYTREC
• Children’s Coalition for Online Safety
• Children’s Commissioner for England’s Office
• Children’s Commissioner for Wales (Comisiynydd Plant Cymru)
• Children’s Health Scotland
• Children’s Law Centre (CLC)
• Chilwell Croft Academy
• Christ Church C of E Primary School
• Christ Church, Winchester
• Christian Action Research and Education
• Clarendon Cottage Prep School
• Cledford Primary and Nursery Academy
• Cliftonville Primary School
• CLOSER - The home of longitudinal research (UCL Social Research Institute)
• Cloudflare
• CNI Scotland
• Co-Design Futures
• Coldfall Primary School
• Collective Shout
• Commission Into Countering Online Conspiracies in Schools
• Commissioner for Victims of Crime Northern Ireland (CVOCNI)
• Common Sense Media
• Compassion in Politics
• Computer and Communications Industry Association (CCIA)
• ConnectFutures C.I.C.
• Conscious Advertising Network and Dr Karen Middleton, University of Portsmouth
• Consumer Choice Center
• Coram
• Cornerhouse (Yorkshire)
• COSLA
• Council for Countering Online Disinformation
• Counselling Services Glasgow
• Coverstar
• Creative UK
• Cumbria County Council
• Curia AI
• Cyber Champions
• Cybersafe Scotland
• Cygnet Healthcare
• CYPF Consortium
• Danscentre
• Denton West End Primary
• Design Bioethics Lab, Neuroscience, Ethics, and Society Team, Department of Psychiatry, University of Oxford
• Diabetes UK
• Digishield Kids (Digishield Labs Ltd)
• Digital Futures for Children Centre, London School of Economics and Political Science
• Digital Intimacy Coalition
• Digital Mental Health Group (Orben Group), Mrc Cognition and Brain Sciences Unit, University of Cambridge
• Digital Policy Alliance
• Digital Poverty Alliance
• Dignity Labs Ltd
• Discord
• Distraction Community
• Ditch the Label
• Dixons Academies Trust
• Dorothy Barley Junior Academy
• Drs Hogan and Allen
• DuckDuckGo
• Dundee Youth Council
• Duolingo
• Durham University — Digital Literacies Network
• Early Years Digital Standards Action Group
• East Riding of Yorkshire Council
• Eastbury Community School
• Eastern Education Group
• eBay
• Echo on Limited
• Electronic Frontier Foundation
• Emanuel School
• Emmer Green Primary School
• End Violence Against Women Coalition (EVAW)
• Entain Plc
• EOS at Federated Hermes Limited
• Epic Games
• Era — the Digital Entertainment and Retail Association
• ES Teachers
• Eslington Primary School
• Esya Centre
• Evangelical Alliance
• Everycare Medway Swale Limited
• Faculty of Public Health
• Fair Play for Children
• Family Fund
• Family Help
• Family Law Bar Association (FLBA)
• Family Online Safety Institute (FOSI)
• First Steps ED
• Flippgen (Youth Town Hall)
• Flux Digital Policy
• Flying Start Nurseries (SW) Ltd
• Forces Children Scotland
• Formby High School
• Forth Valley College
• FOTI (Future of Technology Institute)
• Foundation for Information Policy Research (FIPR)
• Foundry Healthcare Lewes
• Fox Grove School
• FSM Ltd
• Full Fact
• Fullhurst Community College
• Fundamentally Children
• Future Views Today
• Gamer’s Voice (Gamers Voice Ltd) / Stop Killing Games UK
• Games Rating Authority
• Gateshead Local Authority
• GeoComply
• Georgian Gardens CP School
• GingerSnaps Home Ed Facts
• Girlguiding UK
• Global Encryption Coalition Steering Committee
• Global Partners Digital
• Gloucestershire County Council
• GNI
• Goldsmiths University of London
• Good Things Foundation
• Google
• Great Torrington School
• Greater London Authority (Mayor of London)
• Grove Park Primary School
• Gutchecks: Red Flags & Safety
• Gwent Public Health Team, Aneurin Bevan University Health Board
• Hampshire County Council
• Happy Minds Kids Academy
• Harlette Capital Ltd
• Harrietsham CE Primary School
• Harrow Monitoring Group
• Health Action Research Group
• Health Professionals for Safer Screens
• Healthy Minds at Work
• Henry Jackson Society
• Heron Way Primary School
• Hextable Primary
• Highland Council
• Highworth Grammar School
• Hillesley Primary School
• HM Prison and Probation Service
• Holly Lodge High School
• Holmer Green Senior School
• Holtspur School
• Hope Not Hate
• Hope Service
• Horseworks
• Hounslow Local Authority
• House of Culture Brighton C.I.C.
• Huddersfield Town Foundation
• Hull and East Riding Prevention Concordat for Better Mental Health
• Humber Teaching Foundation NHS Trust
• Hummersknott Academy
• Hypereum Ltd
• Ideas Foundation
• IEEE Standards Association (IEEE SA)
• Independent Review of Pornography (Baroness Bertin Review)
• Independent Schools Council
• Ineqe Safeguarding Group
• Information Commissioner’s Office (ICO)
• Information Technology and Innovation Foundation (ITIF)
• Inspired Thinking Group
• Institute of Alcohol Studies
• Institution of Engineering and Technology (IET)
• Intercultural Roots for Public Health
• International Governance of AI programme
• Internet Advertising Bureau (IAB UK)
• Internet Infrastructure Coalition
• Internet Matters
• Internet Society
• Internet Watch Foundation (IWF)
• Into Film
• IntoUniversity
• Iris Anticipa Ltd
• Isle of Wight NHS Youth Forum
• ISOC UK
• ISPA (Internet Service Providers’ Association)
• Istat
• Itchen Sixth Form College
• Just Treatment
• K1ND.World Limited
• Kape Technologies
• Keeping Safe Online (KSO) Youth Group (Welsh Government digital resilience programme)
• Keynsham Town Council
• k-ID
• King
• King Edward VI CEVC School
• King’s Leadership Academy Wavertree
• Kooth
• Lancaster and Morecambe College
• Langley Park School for Girls
• Leeds Arts University
• LEVEL Centre
• LGBT Youth Scotland
• Liaura Limited
• Life Church Southampton
• Life Skills Education Charity
• Lifelab and the Nxtgen Academy, University of Southampton
• Lifelink
• Lifton Community Academy
• Lime Tree Nursery
• LinkedIn
• Longholme Methodist Church Toddler Group
• Lucy Institute for Data and Society, University of Notre Dame
• Luna
• Lytham Highschool
• Lytham St Annes High School
• Mablins Lane CP School
• Mad Youth Organise - Just Treatment
• Maesgwyn Special School
• Manchester Metropolitan University
• Marie Collins Foundation
• Marine Society and Sea Cadets
• Marshlands Primary School
• Match Group
• Meadow Park School
• Melbourn Primary School
• Mental Health Foundation
• Mental Health Innovations
• Mental Health V.
• MeridianFow AI
• Messing Primary School
• Meta Platforms Inc and WhatsApp Inc
• Metrotone Studios
• Mhor Collective
• Microsoft
• Middle Tech Coalition (MTC)
• Milton Keynes City Council
• Mind
• Minderoo Centre for Technology and Democracy at the University of Cambridge
• Misogyny Policy Project
• Miss Estruch Biology Ltd
• Mohn Centre for Children’s Health and Wellbeing, Imperial College.
• Molly Rose Foundation
• Montessori Society UK
• Mothers Union
• Movember Foundation
• Mozilla
• Mumsnet
• Muslim Council of Britain
• NAHT
• Naima JPS
• NaLuna Studios
• NASS - National Association of Special Schools
• NASUWT - The Teachers’ Union
• National AIDS Trust and Chiva (Children’s HIV Association)
• National Child Mortality Database (NCMD) Programme, University of Bristol
• National Children’s Bureau (NCB)
• National Crime Agency (NCA)
• National Education Union (NEU)
• National Governance Association
• National Literacy Trust
• National Youth Agency (NYA)
• Neighbourhood Watch Network
• Nessie In Ed CIC
• Neurodiverse Insights
• Neuro-Informed
• Newcastle City Council - Public Health
• NEWFoCaS
• Newlands Primary School, Normanton
• Nexus
• Nexus NI
• NHS Ayrshire and Arran
• NHS Gloucestershire Foundation Trust
• NHS Greater Glasgow and Clyde
• NHSL Health Improvement
• Nord Security (NordVPN)
• Norfolk Local Authority
• North Down and Ard’s Women’s Aid
• Northern Ireland Human Rights Commission
• Not Yet Media Ltd
• Nottingham University hospitals NHS trust
• Npc
• NSPCC (National Society for the Prevention of Cruelty to Children)
• Nuffield Foundation
• Oak Tree Trust
• Office of the Cheshire Police and Crime Commissioner
• Office of the Police and Crime Commissioner for Devon, Cornwall and the Isles of Scilly
• Oldham Youth Service
• One Collective Power
• Oneness Bilingual School
• Online Dating and Discovery Association
• Online Objectification Project Team
• Online Responsibility Network (ORN) / Trust Alliance Group
• Online Safety Act Network
• Open Rights Group
• OpenAI
• OSITA (Online Safety Tech Industry Association)
• Our AI Collective CIC
• Our Lady of Lourdes Catholic Multi Academy Trust
• Oxleas Foundation NHS trust
• Pact (Producers Alliance for Cinema and Television)
• Paget High School
• Param Shakti Peeth of UK
• Parent Zone
• Parent/Carer Coffee Morning Group
• Parenting Across Scotland
• Parentkind
• Parents for A Better Internet
• Park Mead Primary School
• Pathways and Perspectives
• Paxman Academy
• Peers for Gambling Reform (PGR)
• People Vs Big Tech
• Peppy Agency
• Persona Identities
• PHA
• Philpots Manor School
• Pinterest
• Pirate Party UK
• Plan International UK
• Platfform
• Police Service of Northern Ireland
• Premier Christian Communications
• Primary School in Fife
• Priory (CAMHs clinicians)
• PRIVO
• Pro Performance Clinics
• Promo Cymru
• Prospect Training Services
• Prosper Care
• Proton AG
• PSHE Association
• Public Health Agency
• Public Health Led Pan Ayrshire AA ICYP Programme Board
• Qoria
• Queen Elizabeth’s Grammar School
• Queen Mary Legal Advice Centre
• Queensbury School
• Quinly Ltd
• Radbrook Primary School
• Rainbow Trust Children’s Charity
• Raise
• RCPCH &Us (Royal College of Paediatrics and Child Health — children, young people and family network)
• Reconnect
• Reddit
• Reimagining RSE
• Renfrewshire Child Protection Committee
• Research Project Exploring the Survey for Children and Young People
• Reset Tech
• Resurgam Community Development Trust
• Revealing Reality
• Rika Digital
• Rima Banerjee & Co. – Ai Compliance Specialists
• RMBC
• Roblox
• Rokesly Junior School
• Royal College of General Practitioners
• Royal College of Nursing
• Royal College of Paediatrics and Child Health (RCPCH)
• Royal College of Psychiatrists
• Royal National Institute of Blind People (RNIB)
• Royal Society for Public Health
• Royal Statistical Society (RSS)
• Rugby Free Secondary School
• Rushey Mead Academy
• Rye Youth Zone
• Safe on Social Media
• Safe Path Jordan
• Safelives
• Safer Young Lives Research Centre at University of Bedfordshire
• SafeScreens (a campaign of UsForThem)
• Safetymode Tech Ltd
• Samaritans
• Samsung Electronics UK
• Save the Children UK
• Schroders
• Scottish Children’s Reporter Administration (SCRA)
• Scottish Council for Voluntary Organisations
• Scottish Directors of Public Health Group
• Scottish Sexual Health Promotion Specialists (SSHPS)
• Scottish Youth Parliament (SYP)
• Scouts
• Sea Cadets
• Secex Limited
• Securehaven
• Send Youth Group
• Serc
• Services for Education
• Set@16.Org
• Settlebeck School
• Sex Education Forum
• SHAAP (Scottish Health Action on Alcohol Problems)
• Sheffield Children’s NHS Foundation Trust
• Sheffield Digital Justice
• SHERA Research group
• Shetland Islands Council
• Shireland CAT
• Shout Out UK
• Sir Graham Balfour School
• Skipton and Ripon Constituency Labour Party
• Skyscanner
• Smartphone Free Childhood
• Smartphone Free Schools Rating
• Snap Inc.
• Social Change
• Social Media Victims Law Center / Levin Law
• SocietyInside
• Somerset Local Authority
• South and Maudsley NHS Foundation Trust
• South Eastern Regional College
• South England Conference of Seventh Day Adventists
• South Street Community Primary
• South West Grid for Learning (SWGfL)
• Sowerby Primary Academy
• SPEAK UK
• Spilsbury Holdings Limited t/a Aztec Labs
• Spotify
• St Anne’s CE Lydgate Primary School
• St Barnabas Church, Emmer Green and Caversham Park
• St Christophers
• St Ignatius College
• St James CE Primary School
• St John’s CofE (Aided) Primary School
• St Joseph’s Catholic Primary School
• St Leonard’s Catholic School
• St Luke’s CE Primary School
• St Patrick’s Church - Woolston - Southampton
• St Philip Howard Primary
• St Wilfrid’s Catholic School
• St. Annes Catholic Primary School
• Stamplo Limited
• Stirchley Primary School
• Strava
• Surrey County Council
• Sway.ly
• Targeted Youth Work
• Tattershall Primary School
• Teacher Tapp
• techUK
• TeenAegis
• Tellmi Ltd
• Tencent
• Tender Education Arts
• Thamesview School
• The Authority Gap Consultancy
• The Behavioural Insights Team (BIT)
• The Ben Kinsella Trust
• The Big House Ireland
• The Bolter & Chainsword
• The Brainsafe Standard
• The British and Irish Law Education Technology Association (BILETA)
• The Chartered Institute of Building (CIOB)
• The Children’s Media Foundation (CMF)
• The Children’s Society
• The Citadel
• The Compass Centre
• The Douay Martyrs Catholic Secondary School
• The FitzWimarc School
• The Fostering Network
• The Free Speech Union
• The Graft Project
• The Grove School
• The Guardian Foundation
• The Indigo Childcare Group
• The Katie Trust
• The Keystone Academy
• The LEGO Group
• The Next Century Foundation
• The Reduce Digital Distraction Project
• The Salvation Army (United Kingdom and Ireland Territory)
• The Stable School (SEMH)
• The University of Manchester (#So.Me, #BeeWell, SMILE Research Centre)
• The Week Junior
• The Yellow Wood Project CIC
• The Zone Youth Leeds 17
• Three Rivers Academy
• TikTok
• Tircea CIC
• Together Association
• Tor Bridge High
• Trinity Church of England School
• Triple P UK and Ireland
• TrustDating LTD
• Trustpilot
• TTYY School
• Twitch
• UK Brain Health Foundation
• Ukie (Association for UK Interactive Entertainment)
• UNICEF
• United Schools Federation
• University of Bristol
• University of Kent
• University of Stirling
• University of the West of England
• Unpugged Coalition
• Untamed
• Upper School
• Ustwo Games
• Uzube
• Vale of Glamorgan
• Valley Primary School and Nursery
• Verifymy Limited
• Victoria Place Day Nursery
• Virgin Media O2
• Virgin Money Foundation
• Viridis Federation
• Visible Ministries- Naked Truth Project
• VodafoneThree
• Voice 21
• Voyage Travel Marketing
• VPN Trust Initiative (VTI) / Internet Infrastructure Coalition (i2Coalition)
• Wainscott Primary and Nursery School
• Walsall Council
• Warwickshire County Council
• Wavelength
• Wayland Academy Secondary School
• We Fix UK Now Community Support & Solutions CIC
• West Dunbartonshire Council - Education, Learning and Attainment
• West End Primary School
• West Herts College
• West Midlands Hindu Forum
• West Sussex County Council
• Whiston Preschool
• White Ash School
• White, Turing & Lovelace LLC
• Wickersley Partnership Trust
• Wikimedia Foundation
• Wildbrain
• Windsor Academy Trust
• Wirral Hospitals’ School
• With Kids
• Wokingham Borough Council (Community Safety)
• Women’s Aid Federation Northern Ireland
• Women’s Resource and Development Agency (WRDA)
• Woodlands (CIO)
• Workers Party of Britain
• Working Group on Gaming and Regulation
• World Cancer Research Fund
• Worthing High School
• Wyvern Primary School
• X
• xAI
• Yate Town Council
• Ygam
• YGG Blaendulais
• Yondr
• Yoti
• Youngminds
• Youth Sport Trust
• Youthlink Scotland
• Youthworks Consulting
• Zero Tolerance
• Zoom Communications, Inc.
• Zoom Strategies

1. Respondents who indicated that they had “parental or caring responsibility for at least one child who will be 21 or younger by the 26th of May” ↩