This Government is deeply concerned by the devastating impact online fraud can have on victims, both financially and emotionally, including that which is artificial intelligence (AI)-enabled.

That is why, under the Online Safety Act (OSA), the Government has taken action to ensure online platforms and services are required to implement robust preventative measures to stop fraudulent content from reaching potential victims. AI content is treated no differently to other types of content under the OSA – if it is fraudulent in nature, platforms are required to prevent it from reaching users and take swift action if it does appear.

From mid-2027 the largest in-scope companies will also be required to take greater steps to prevent fraudulent adverts from appearing, including those that are AI-enabled.

The 30-month review period for those granted protection reflects the Government’s view that protection should be provided for as long as it is needed. As conditions in countries of origin can change over time, it is appropriate to review whether an individual continues to require international protection before further permission is granted.

Individuals with an ongoing protection need will have their status renewed, while those who no longer require protection will be expected to leave the UK unless they qualify for leave on another basis. Safe return reviews have long been a feature of the UK's protection system. Protection needs can change over time, and it is therefore right that an individual's circumstances are reviewed before further leave or settlement is granted. The Government remains committed to providing protection to those who need it, in line with the UK's international obligations.

Settlement for those granted protection has never been automatic. Individuals have always needed to meet the requirements in place at the point they become eligible to apply, including demonstrating that they continue to require protection where relevant. The principle that protection should be provided for as long as it is needed, and reviewed where circumstances change, is a longstanding feature of the UK's protection system.

In addition, in future those on core protection will be able to switch to the protection work and study route, which will provide greater certainty about their future in the UK. This will also allow them to earn down the time it takes to obtain settlement and open up refugee family reunion rights. Individuals who switch to this route will no longer be subject to safe return reviews, recognising their contribution through work or study.

The Prime Minister announced on 8 June that the UK will be the first country in the world where it will be impossible for any child in the UK to take share or view nude images.

We will work with tech companies for three months to ensure that comprehensive nudity detection is implemented at operating system level, with adults able to switch it off by proving they are over-18 through age assurance. If industry does not meet our high expectations, then we will legislate.

Nudity detection happens entirely on‑device: no data leaves the device; there is no reporting to industry, government or law enforcement. Apple and Google have both produced public documentation explaining that their existing nudity detection technology happens entirely on device and neither company has access to the imagery,

Our engagement with industry over three months is with the objective of enhancing this existing technology to apply comprehensively and effectively across the device. Efficacy, proportionality and user privacy will of course continue to be central to those discussions.

The Prime Minister announced on 8 June that the UK will be the first country in the world where it will be impossible for any child in the UK to take share or view nude images. We will work with tech companies for three months to ensure that comprehensive nudity detection is implemented at operating system level, with adults able to switch it off by proving they are over-18 through age assurance. If industry does not meet our high expectations, then we will legislate.

The primary aim of this intervention is to disrupt grooming and sextortion pathways, to prevent exploitation of children as early as possible; and as an additional safety intervention to ensure that children cannot access pornographic material.

It is important that each user of a smartphone uses the device with their own operating system account. Such an account should be set up with nudity detection default-on, and only deactivated through age assurance to prove the user is over-18. If a device is shared, parents/carers should make sure that default-on nudity detection protections are not deactivated if the device is likely to be used by a child. Parents/carers should also be mindful if their phone – which may have nudity detection deactivated – is picked up by a child. Risks such as this, amongst others, will rightly form part of the government's three-month engagement with tech companies to work through technical solutions to meet the government's objective.

While this intervention is specifically bounded to nudity detection which can be deactivated through age assurance to prove the user is over-18, the government will consider the benefits and risks of wider potential use cases for these technological innovations.

Device-level age assurance could have further applicability to other online age restrictions. The Prime Minister announced restrictions on social media services for under-16s, on 15 June - Social media to be banned for under-16s in landmark government move to give kids their childhood back - GOV.UK.

The government committed to learn lessons from Australia’s experience by introducing more highly effective age assurance (HEAA) measures to support compliance, making it far harder for children to bypass safeguards; and that Ofcom will conduct a rapid study on what is effective age assurance for verifying whether someone is over 16.

Device-level age assurance can be more effective, convenient and privacy preserving for the user – including making it harder for users to circumvent restrictions via VPNs as can be done with platform level age checks. The government will continue to actively consider the extent to which device-level age assurance could be used to restrict access to age-gated services, such as social media and VPN services. The government’s announcement on 15 June 2026 did not include any explicit proposal to restrict or ban VPNs for under-16s, or universal screen time proposals.

We welcome measures that tech companies have already brought forward in this area – such as nudity detection already implemented by both Apple and Google. Apple introduced device-level age assurance in its iOS26.4 operating system update in March 2026, and expanded Communication Safety features to intervene when gore or violent content is detected on child accounts. These are constructive and welcome developments, which will protect children from other types of harmful content as well as nudity. Key to the targeting of such interventions is the differentiation of a child user of a device, which can be achieved through device-level age assurance to activate other safety interventions. The government will continue to consider options to go further in this respect.

Regarding the use, and restricted use, of devices while children are at school, Department for Education guidance (updated February 2026) is explicit that all schools should be mobile phone free environments by default, and that schools prohibit the use of mobile phones throughout the school day. Furthermore, the government continually horizon scans to appraise new technological innovations which may enable better safeguarding of children online, and device usage in school. The government engages with industry regularly to stay abreast of promising products which could be applied to better safeguarding children.

The Prime Minister announced on 8 June that the UK will be the first country in the world where it will be impossible for any child in the UK to take share or view nude images. We will work with tech companies for three months to ensure that comprehensive nudity detection is implemented at operating system level, with adults able to switch it off by proving they are over-18 through age assurance. If industry does not meet our high expectations, then we will legislate.

The primary aim of this intervention is to disrupt grooming and sextortion pathways, to prevent exploitation of children as early as possible; and as an additional safety intervention to ensure that children cannot access pornographic material.

It is important that each user of a smartphone uses the device with their own operating system account. Such an account should be set up with nudity detection default-on, and only deactivated through age assurance to prove the user is over-18. If a device is shared, parents/carers should make sure that default-on nudity detection protections are not deactivated if the device is likely to be used by a child. Parents/carers should also be mindful if their phone – which may have nudity detection deactivated – is picked up by a child. Risks such as this, amongst others, will rightly form part of the government's three-month engagement with tech companies to work through technical solutions to meet the government's objective.

While this intervention is specifically bounded to nudity detection which can be deactivated through age assurance to prove the user is over-18, the government will consider the benefits and risks of wider potential use cases for these technological innovations.

Device-level age assurance could have further applicability to other online age restrictions. The Prime Minister announced restrictions on social media services for under-16s, on 15 June - Social media to be banned for under-16s in landmark government move to give kids their childhood back - GOV.UK.

The government committed to learn lessons from Australia’s experience by introducing more highly effective age assurance (HEAA) measures to support compliance, making it far harder for children to bypass safeguards; and that Ofcom will conduct a rapid study on what is effective age assurance for verifying whether someone is over 16.

Device-level age assurance can be more effective, convenient and privacy preserving for the user – including making it harder for users to circumvent restrictions via VPNs as can be done with platform level age checks. The government will continue to actively consider the extent to which device-level age assurance could be used to restrict access to age-gated services, such as social media and VPN services. The government’s announcement on 15 June 2026 did not include any explicit proposal to restrict or ban VPNs for under-16s, or universal screen time proposals.

We welcome measures that tech companies have already brought forward in this area – such as nudity detection already implemented by both Apple and Google. Apple introduced device-level age assurance in its iOS26.4 operating system update in March 2026, and expanded Communication Safety features to intervene when gore or violent content is detected on child accounts. These are constructive and welcome developments, which will protect children from other types of harmful content as well as nudity. Key to the targeting of such interventions is the differentiation of a child user of a device, which can be achieved through device-level age assurance to activate other safety interventions. The government will continue to consider options to go further in this respect.

Regarding the use, and restricted use, of devices while children are at school, Department for Education guidance (updated February 2026) is explicit that all schools should be mobile phone free environments by default, and that schools prohibit the use of mobile phones throughout the school day. Furthermore, the government continually horizon scans to appraise new technological innovations which may enable better safeguarding of children online, and device usage in school. The government engages with industry regularly to stay abreast of promising products which could be applied to better safeguarding children.

The Prime Minister announced on 8 June that the UK will be the first country in the world where it will be impossible for any child in the UK to take share or view nude images. We will work with tech companies for three months to ensure that comprehensive nudity detection is implemented at operating system level, with adults able to switch it off by proving they are over-18 through age assurance. If industry does not meet our high expectations, then we will legislate.

The primary aim of this intervention is to disrupt grooming and sextortion pathways, to prevent exploitation of children as early as possible; and as an additional safety intervention to ensure that children cannot access pornographic material.

It is important that each user of a smartphone uses the device with their own operating system account. Such an account should be set up with nudity detection default-on, and only deactivated through age assurance to prove the user is over-18. If a device is shared, parents/carers should make sure that default-on nudity detection protections are not deactivated if the device is likely to be used by a child. Parents/carers should also be mindful if their phone – which may have nudity detection deactivated – is picked up by a child. Risks such as this, amongst others, will rightly form part of the government's three-month engagement with tech companies to work through technical solutions to meet the government's objective.

While this intervention is specifically bounded to nudity detection which can be deactivated through age assurance to prove the user is over-18, the government will consider the benefits and risks of wider potential use cases for these technological innovations.

Device-level age assurance could have further applicability to other online age restrictions. The Prime Minister announced restrictions on social media services for under-16s, on 15 June - Social media to be banned for under-16s in landmark government move to give kids their childhood back - GOV.UK.

The government committed to learn lessons from Australia’s experience by introducing more highly effective age assurance (HEAA) measures to support compliance, making it far harder for children to bypass safeguards; and that Ofcom will conduct a rapid study on what is effective age assurance for verifying whether someone is over 16.

Device-level age assurance can be more effective, convenient and privacy preserving for the user – including making it harder for users to circumvent restrictions via VPNs as can be done with platform level age checks. The government will continue to actively consider the extent to which device-level age assurance could be used to restrict access to age-gated services, such as social media and VPN services. The government’s announcement on 15 June 2026 did not include any explicit proposal to restrict or ban VPNs for under-16s, or universal screen time proposals.

We welcome measures that tech companies have already brought forward in this area – such as nudity detection already implemented by both Apple and Google. Apple introduced device-level age assurance in its iOS26.4 operating system update in March 2026, and expanded Communication Safety features to intervene when gore or violent content is detected on child accounts. These are constructive and welcome developments, which will protect children from other types of harmful content as well as nudity. Key to the targeting of such interventions is the differentiation of a child user of a device, which can be achieved through device-level age assurance to activate other safety interventions. The government will continue to consider options to go further in this respect.

Regarding the use, and restricted use, of devices while children are at school, Department for Education guidance (updated February 2026) is explicit that all schools should be mobile phone free environments by default, and that schools prohibit the use of mobile phones throughout the school day. Furthermore, the government continually horizon scans to appraise new technological innovations which may enable better safeguarding of children online, and device usage in school. The government engages with industry regularly to stay abreast of promising products which could be applied to better safeguarding children.

The Prime Minister announced on 8 June that the UK will be the first country in the world where it will be impossible for any child in the UK to take share or view nude images. We will work with tech companies for three months to ensure that comprehensive nudity detection is implemented at operating system level, with adults able to switch it off by proving they are over-18 through age assurance. If industry does not meet our high expectations, then we will legislate.

The primary aim of this intervention is to disrupt grooming and sextortion pathways, to prevent exploitation of children as early as possible; and as an additional safety intervention to ensure that children cannot access pornographic material.

It is important that each user of a smartphone uses the device with their own operating system account. Such an account should be set up with nudity detection default-on, and only deactivated through age assurance to prove the user is over-18. If a device is shared, parents/carers should make sure that default-on nudity detection protections are not deactivated if the device is likely to be used by a child. Parents/carers should also be mindful if their phone – which may have nudity detection deactivated – is picked up by a child. Risks such as this, amongst others, will rightly form part of the government's three-month engagement with tech companies to work through technical solutions to meet the government's objective.

While this intervention is specifically bounded to nudity detection which can be deactivated through age assurance to prove the user is over-18, the government will consider the benefits and risks of wider potential use cases for these technological innovations.

Device-level age assurance could have further applicability to other online age restrictions. The Prime Minister announced restrictions on social media services for under-16s, on 15 June - Social media to be banned for under-16s in landmark government move to give kids their childhood back - GOV.UK.

The government committed to learn lessons from Australia’s experience by introducing more highly effective age assurance (HEAA) measures to support compliance, making it far harder for children to bypass safeguards; and that Ofcom will conduct a rapid study on what is effective age assurance for verifying whether someone is over 16.

Device-level age assurance can be more effective, convenient and privacy preserving for the user – including making it harder for users to circumvent restrictions via VPNs as can be done with platform level age checks. The government will continue to actively consider the extent to which device-level age assurance could be used to restrict access to age-gated services, such as social media and VPN services. The government’s announcement on 15 June 2026 did not include any explicit proposal to restrict or ban VPNs for under-16s, or universal screen time proposals.

We welcome measures that tech companies have already brought forward in this area – such as nudity detection already implemented by both Apple and Google. Apple introduced device-level age assurance in its iOS26.4 operating system update in March 2026, and expanded Communication Safety features to intervene when gore or violent content is detected on child accounts. These are constructive and welcome developments, which will protect children from other types of harmful content as well as nudity. Key to the targeting of such interventions is the differentiation of a child user of a device, which can be achieved through device-level age assurance to activate other safety interventions. The government will continue to consider options to go further in this respect.

Regarding the use, and restricted use, of devices while children are at school, Department for Education guidance (updated February 2026) is explicit that all schools should be mobile phone free environments by default, and that schools prohibit the use of mobile phones throughout the school day. Furthermore, the government continually horizon scans to appraise new technological innovations which may enable better safeguarding of children online, and device usage in school. The government engages with industry regularly to stay abreast of promising products which could be applied to better safeguarding children.

Section 17 of the Children Act 1989 establishes the general duty of local authorities in England to safeguard and promote the welfare of children within their area who are in need and, so far as is consistent with that duty, to promote the upbringing of such children by their families.

I can confirm that the Home Office is working with the Department for Education alongside other government departments and the NRPF network to produce guidance to assist local authorities in their duties to those with No Recourse to Public Funds as part of the Child Poverty Strategy.

The duty Section 17 of the Children Act 1989 duty does not apply to the Home Office. Instead, Section 55 of the Borders, Citizenship and Immigration Act 2009 places a duty on the Home Office to make arrangements for ensuring that immigration, asylum, nationality and general customs functions are discharged having regard to the need to safeguard and promote the welfare of children in the UK.

The statutory guidance, called ‘Every Child Matters’, issued under Section 55 (3), sets out the key arrangements for safeguarding and promoting the welfare of children as they apply both generally to public bodies who deal with children in the UK (Part 1) and specifically to the Home Office (Part 2).

We want to encourage refugees to integrate more fully into the communities providing them sanctuary and we will encourage refugees to switch out of the Core Protection route wherever possible into a new, bespoke work and study route. In addition to the wider the socio-economic advantages, this will also enable them to access family reunion and settlement rights with new fees and conditions in accordance with the rules of that route.

We are committed to ensuring our policies are sustainable and do not place unnecessary burdens on the taxpayer, and this policy is no different. In terms of resources and processing, only those who do remain on Core Protection will have their protection needs regularly reviewed. Reviews will be targeted based on objective country information, and any other new information that comes to light, allowing us to make the most efficient use of resources. We will continue to monitor staffing levels and will deploy our workforce flexibly subject to business needs as we have done in the past.

The approach to reviewing status will be efficient and targeted; we will reassess where country conditions or personal circumstances have changed significantly.

To do this we will build on the efficiencies that have already seen record levels of initial decisions being made, including use of AI, such as our policy search tool or case summarisation which are already assisting asylum decision makers.

Under the Online Safety Act, all regulated services must implement proportionate safety measures across all spaces. Platform design cannot be used as an excuse to avoid detection and reporting obligations. The Crime and Policing Bill will further strengthen protections for children against computer-generated abuse by criminalising AI child sexual abuse material, banning nudify tools, outlawing AI paedophile manuals and introducing an AI model testing defence to prevent abuse at source.

To ask His Majesty’s Government what steps they are taking to protect children in private messaging spaces from the prevalence of computer-generated child sexual abuse material.

Does the Minister agree that much of the responsibility for making Section 121 of the Online Safety Act effective lies with Ofcom? Will he confirm that the Government expect Ofcom to be bold and robust in its application of the new law to online spaces in private messaging contexts, and to use its new powers quickly so that we in Parliament can recognise that we have passed an effective law to protect children from vile online abuse?

My Lords, given the recent experience with AI platforms such as Grok generating unacceptable non-consensual sexual images and the warnings from the Internet Watch Foundation, I welcome recent comments from the Secretary of State for DSIT about Grok. However, what discussions are senior Ministers having with technology companies directly to ensure that they understand their duties under the Online Safety Act and will comply?

My Lords, work is under way with the national centre for violence against women and girls to improve the police’s response to non-consensual intimate image abuse so that they can proactively report such content for removal and hashing. Does the Minister think the Government should mandate this system to ensure that it becomes more effective across the country?

My Lords, upload prevention technology is already being used by platforms in private messaging environments to detect harmful content such as malware, and, as we have just heard, experts such as the Internet Watch Foundation have confirmed that the same approach is feasible for detecting known child sexual abuse material, yet platforms continue to falsely claim that such scanning is impossible. What steps will the Government take specifically to ensure that Ofcom is testing and challenging these claims so that such arguments cannot be used to evade compliance?

Can Ofcom work to do this now or does it have to wait for the end of the consultation?

My Lords, a number of noble Lords have raised concerns about Grok and those unacceptable images. The Minister says that it is a duty of Ofcom. Is he aware whether Ofcom has actually raised the issue of these images with Grok directly yet?

My Lords, what possible confidence can we have in Ofcom to take action effectively, given its lamentable failure to stand up for customers against the mobile phone companies?

My Lords, the Minister is again experiencing what happened during the stages of the Crime and Policing Bill when we discussed this subject, because of the strength of feeling right across the House about this. What really shocked me in the briefing for this Question is the fact that while we know child sexual abuse material is increasing exponentially, largely through AI, 91% of that material is actually being produced by children themselves, astonishingly. Can I appeal to the Minister to work closely with his colleague, the noble Baroness, Lady Smith, on Report on the Children’s Wellbeing and Schools Bill, and to think about what we can do to try to face up to this unfortunate fact and do something about it?

My Lords, I welcome the Minister’s strength in his answers. He says that if Ofcom does not act to hold X to account for Grok’s illegal activities, the Government will. Can he expand on how the Government will act, if they need to?