Written questions by Wrigley.

Innovation and Technology, what assessment she has made of the potential impact of digital exclusion on (a) elderly and (b) disabled people who do not have a reliable internet connection.

Those who don’t have access to, or cannot use digital technologies, including older people and disabled people, are likely to pay more for certain transactions, have worse health outcomes, and find it more difficult to manage their finances. Knock-on effects of digital exclusion can include difficulty accessing essential services and isolation from their community.That’s why in February, we published the Digital Inclusion Action Plan which outlines the first five actions we are taking over the next year on digital inclusion. One of these actions was to launch the £9.5mn Digital Inclusion Innovation Fund to support and expand local community initiatives to get people online, which we did in August. All funded projects will support at least one of the focus demographic groups, which include older and disabled people.

Innovation and Technology, what discussions she has had with business representatives on ensuring that elderly people are not prevented from using businesses that choose to move to digital-only services.

The government does not generally intervene in how businesses choose to conduct their activities or offer their products and services. However, the Government expects all businesses to treat all consumers fairly.Having said that, everyone has a role to play in realising a shared vision for a digitally inclusive UK to ensure everyone, including the elderly, can participate in our modern digital society. That is why we launched the Digital Inclusion Action Plan in February, which sets out the first actions we are taking over the next year to boost digital inclusion.

Innovation and Technology, when she plans to respond to the correspondence from the hon. Member for Newton Abbot, dated 2 July 2025, on behalf of his constituent L Coombes.

The Department places significant importance on providing timely responses to correspondence. We sincerely apologise for the delay in this instance and assure the Member that a full response will be provided as soon as possible.

Innovation and Technology, what guidance his Department provides to departments using suppliers who operate (a) Foundry, (b) Gotham and (c) other cross-sector data platforms on the risk of cross-domain data use.

All departments must adhere to the UK Data Protection legislation to manage risk and protect personal data when they are using cross-sector data platforms provided by suppliers. All departments are controllers of the personal data they hold and are individually responsible for demonstrating compliance with the data protection principles, and take appropriate technical and organisational mitigations in line with the UK GDPR to reduce risk. Under the same legislation, all departments are required to appoint a data protection officer (DPO), who must be an adequately resourced expert in data protection to assess compliance of data platforms, provide advice regarding Data Protection Impact Assessments (DPIAs) to identify and mitigate risks.The Government Digital Service has published the principles for securing personal data in government services, which includes a principle that outlines the actions that departments need to take for best compliance with the data protection legislation when they use platforms provided by third-party suppliers. The actions departments are expected to take include commercial agreements with robust terms and conditions for protection of data, appropriately assigned liabilities, risk assessments, audits, monitoring and oversight of compliance with processing terms and conditions as well as seeking further assurances of compliance from platform suppliers.The Government Security Group and the Government Digital Service have developed the Security by Design Policy, which covers the management of third-party product security risks.Government Digital Marketplace is a resource for departments and the wider public sector organisations to find reliable and secure technology for their digital projects. Departments are provided with guidance that covers features, security arrangements, standards, certifications and compliance information of supplier platforms, which fall under existing government frameworks.

Innovation and Technology, what assessment he has made of the potential impact of his Department's white paper entitled A pro-innovation approach to AI regulation, published in March 2023, CP 815, on proprietary models trained on NHS data.

As set out in the AI Opportunities Action Plan, most AI systems should be regulated at the point of use and our expert regulators are best placed to do so. Departments are working proactively with regulators to provide strategic direction and support them on their AI capability needs.

Innovation and Technology, whether he has received recent correspondence from Palantir.

The Secretary of State has received recent correspondence from Palantir and will be responding in due course.

Innovation and Technology, if he will bring forward legislative proposals to guarantee that individuals have a right to opt out of digital ID schemes.

The Government has no plans to introduce legislation on the right to opt out of using digital identities. Nor has the Government taken any steps to make their use mandatory.The digital verification services provisions in the Data (Use and Access) Act 2025 are designed to give people choice. They help both people and organisations to realise the benefits of secure digital identities with greater confidence, if they choose to use them.The Government also plans to introduce digital versions of existing government-issued documents through the GOV.UK Wallet, a secure digital service for storing government-issued credentials on users’ smartphones. However, people will still be able to prove their identity using physical documents if they prefer.

Innovation and Technology, pursuant to the Answer of 13 June 2025 to Question 57480 on Broadband: Newton Abbot, what his planned timetable is for supplying (a) data speeds of over 30Mbps to the remaining 2% of (i) homes and (ii) businesses and (b) gigabit capable broadband to the remaining 17% of (A) homes and (B) businesses.

The rollout of superfast and gigabit-capable broadband is primarily driven by the private sector, with government only intervening where commercial development will not suffice. The government plans for 99% of UK homes and businesses to have access to gigabit-capable broadband by 2032. In Newton Abbot, Openreach is delivering a Project Gigabit contract to premises that are unlikely to be reached by suppliers’ commercial rollout. This contract will disproportionately target premises currently without superfast speeds (>= 30Mbps).The first connections under this contract are expected to be made early next year, and the contract is currently expected to run until 2030.In the meantime, we expect superfast and gigabit-capable coverage to continue to steadily increase in Newton Abbot as a result of suppliers’ commercial rollout.

Innovation and Technology, if she will make an assessment of the potential impact of the delay to the gigabit broadband target to 2032 in Newton Abbot constituency.

Openreach is delivering a Project Gigabit contract in the Newton Abbot constituency, that is expected to run until 2030. The first connections are expected to be made early next year.

Innovation and Technology, if he will take legislative steps to require AI systems used by (a) government and (b) industry to disclose where their data searches are from.

Many existing rules apply to AI systems. When personal data is processed by AI systems, data protection law requires transparency from developers about how they collect and use personal data, and outline information in a privacy notice. The Online Safety Act regulates AI services which allow users to share content or search live websites to provide search results. They must assess and mitigate risks to users of illegal content and content harmful to children.The Government will introduce AI legislation, delivering on its manifesto commitment and preparing for AI’s opportunities and risks. It will consult on proposals later this year

Innovation and Technology, if he will bring forward legislative proposals to regulate who is responsible in cases of defamation involving an AI model.

A range of existing rules already apply to AI systems, such as data protection, competition, equality legislation and other forms of sector regulation. The UK has well established defamation laws, and our data protection laws also apply to personal data processed in the context of AI. We continue to monitor developments in the common law in this area.The government is clear in its ambition to bring forward AI legislation which delivers on our manifesto commitment, making sure our statute book is ready for the age of AI. We will consult publicly on our wider legislative proposals later this year.

Innovation and Technology, if he has had recent discussions with the Information Commissioner’s Office on the potential risks of using foreign-owned companies to manage cloud-based data systems in the public sector.

The Secretary of State has not specifically discussed foreign-owned cloud-based data systems with the Information Commissioner’s Office (ICO). The ICO has discussed the compliance of cloud system rules on international transfers with government officials. The ICO will produce and update guidance during the implementation of the Data (Use and Access) Act, including around the use of processors and sub-processors which may be relevant to cloud based systems.

Innovation and Technology, if he will make an assessment of the potential merits of introducing regulation to ensure consumers are aware if the legal advice they have sought is AI generated.

The Government believes that the vast majority of AI systems should be regulated at the point of use and that our existing expert regulators are best placed to do this. Oversight of this technology is distributed across several sectoral regulators and departments. Through delivery of the AI Opportunities Action Plan, we are making sure regulation is well-designed and implemented, to fuel fast, wide and safe development and adoption of AI.

Innovation and Technology, if he will allocate funding to research on the (a) reliability and (b) auditability of non-deterministic models of AI used in national security issues.

AI offers the most powerful lever we have for national renewal; it underpins the Government’s plan to drive economic growth, transform public services, and boost living standards for working people across the country.During the recent Spending Review, the Government showed its commitment to delivering the AI Opportunities Action Plan (AIOAP), with the Chancellor confirming over £2 billion to implement the Plan, and up to £750m for a new Edinburgh supercomputer, which will support our understanding of this emerging technology.More widely, the government is investing record levels of investment in R&D, growing to £22.6 billion per year by 2029/30. Through UKRI, we are supporting science and innovation across the UK to better deliver on the government’s priorities, including national security.

Innovation and Technology, if he will publish guidance on the safe use of non-deterministic AI in public sector decision-making systems.

The AI playbook clearly describes the non-deterministic nature of probabilistic, AI-based predictive analysis and automated decision-making, while prescribing ten key principles for the safe and responsible use of AI. For example, it highlights the non-deterministic and probabilistic nature of AI models—specifically large language models (LLMs)—which can yield inaccurate or hallucinated outputs. Accordingly, it stresses legal and ethical duties, requiring departments to register AI-powered algorithms via the Algorithmic Transparency Recording Standard. It also recommends and emphasises meaningful human oversight, including monitoring, assurance, and user feedback mechanisms, with a risk-aware approach from inception through the entire life cycle of the services.

Innovation and Technology, if he will take legislative steps to introduce a regulatory framework to set standards for (a) testing and (b) verification of non-deterministic AI systems.

It is right that most AI systems are regulated at point of use by our existing expert regulators, who are best placed to do this. In response to the AI Action Plan, Government committed to building regulators’ capabilities to do this.Additionally, the Government is developing legislative proposals which will allow us to securely realise the enormous benefits of the most powerful AI systems. We will launch a public consultation, and engage with a full range of stakeholders, on our legislative proposals in due course.

Innovation and Technology, whether his Department has made an assessment of the potential impact on operational trust when non‑deterministic AI models produce different outputs from identical inputs.

The government recognises the importance of operational trust in AI systems. DSIT is building confidence in and driving adoption of AI by supporting a growing, competitive, and dynamic AI assurance ecosystem.By providing ways to measure, evaluate, and communicate the trustworthiness of AI systems, AI assurance can increase confidence in AI systems, support AI adoption, and boost economic growth.We have also placed a renewed focus on skills, recently announcing a joint commitment with industry to provide 7.5 million workers with fundamental AI skills. Equipping workers with these skills will also help build operational trust in AI systems, building understanding as to when a model is working as intended, even where different outputs are produced from identical inputs.

Innovation and Technology, whether civil servants receive (a) training and (b) guidance on managing the risks of non-deterministic outputs in AI-assisted decision-making.

DSIT has provided civil servants with both training and guidance on managing the risks of AI, including non-deterministic models such as LLMs. Training includes over 70 free courses on AI fundamentals, ethics, machine learning and generative AI, available through Civil Service Learning and the Government Campus. These emphasise critical evaluation of AI outputs and awareness of risks like bias and hallucination. The AI Playbook for the UK Government outlines principles for managing risks, accuracy, bias, hallucinations, and security threats. Civil servants are advised to apply human oversight and robust testing to ensure responsible use of AI and always adhere to the Model of Responsible Innovation.

Innovation and Technology, whether his Department has issued guidance to government Departments on the (a) procurement and (b) use of non-deterministic AI models in public service delivery.

DSIT has issued detailed guidance via the AI Playbook for the UK Government on both (a) procurement and (b) use of AI, including forms of non-deterministic AI such as Generative AI.The Buying AI section outlines best practices on engaging commercial teams, specifying requirements, and aligning procurement with ethics and regulation.The Playbook stresses that such AI systems produce probabilistic, non-deterministic outputs and require continuous monitoring, ethical safeguards, and clear understanding of their limitations. It provides guidance on managing the risks of generative AI—including hallucination, bias, and misinformation—while emphasising human oversight, data protection, and responsible deployment.

Innovation and Technology, what mechanisms are in place to ensure (a) reproducibility and (b) accountability when AI is used in cross-government digital services.

The AI Playbook discusses accountability requirements for the use of AI in the public sector, emphasising human validation for high-risk decisions and user-reporting channels for oversight. Transparency is mandated through the Algorithmic Transparency Recording Standard (ATRS), requiring public disclosure of algorithms in decision-making, with contestability mechanisms for appeals and redress.The Playbook and Magenta Book stress considering reproducibility and scalability from project inception, supported by rigorous evaluation and quality assurance. The AI Community fosters collaboration across government to prevent duplicated efforts, while the AI Knowledge Hub’ collection of AI use cases supports reproducibility and scalability of projects through shared learning and best practices.