AI Opportunities Action Plan
United Kingdom · Strategy · Department for Science, Innovation & Technology
Our summary
The UK government's strategy to accelerate adoption and development of artificial intelligence to promote economic growth, create jobs, and enhance public services. It addresses three main goals: building AI foundations through computing infrastructure and talent; deploying AI across the public and private sectors; and positioning the UK as a leader in frontier AI development rather than merely a user of foreign technology.
Written by the Index in our own words. It is not legal advice; the official text is the authority.
Inside this policy
Every section of this policy in order. Our notes summarise each one in plain language and do not quote the official text, which is the authority.
Drafted by the Index An automated summary of the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.
19 sections
Jump to a chapter
Foreword by the Secretary of State for Science, Innovation and Technology
The UK must lead in shaping how artificial intelligence advances and is applied, rather than passively accepting developments from other countries. Government will work with leading companies, academics, and entrepreneurs to guide AI progress according to principles of shared prosperity, improved public services, and expanded work opportunities for citizens.
The opportunity
The plan aims to position AI as a key lever for economic growth and national missions, building on the UK's strengths in research, startups, frontier companies and AI safety governance. Government action should focus on investing in AI foundations, driving cross-economy adoption, developing national champions across the AI stack, using government procurement as a market-shaping lever, attracting global talent and capital, and building on UK strengths in AI applications and emerging research areas like AI for science and robotics. The UK should make a strategic commitment to AI development despite uncertainty about future progress.
1 Lay the foundations to enable AI
4 sections
This section covers four areas essential for AI development: building secure and sustainable data centres; making government and private datasets available to developers; training and retaining AI scientists and engineers; and establishing regulatory frameworks that enable safe AI innovation.
1.1 Building sufficient, secure and sustainable AI Infrastructure
Computational infrastructure should comprise sovereign, domestic and international compute; government should establish a 10-year AI infrastructure plan within 6 months with sustained investment, expand the AI Research Resource by at least 20 times by 2030 into mission-oriented clusters avoiding vendor lock-in, and appoint mission-focused programme directors to allocate sovereign compute. AI Growth Zones should streamline data centre development with at least one zone nominated quickly, whilst government mitigates sustainability and security risks, supports innovative hardware, and establishes international partnerships with like-minded countries.
1.2 Unlocking data assets in the public and private sector
Government should responsibly unlock public and private datasets to fuel AI development and innovation. Specific actions include identifying at least 5 high-impact public datasets for AI researchers and innovators; collecting new data in strategically important areas; publishing guidelines for releasing government datasets; combining compute with proprietary data access; building public sector data infrastructure; and developing privacy-preserving versions of sensitive datasets using synthetic data techniques.
1.3 Training, attracting and retaining the next generation of AI scientists and founders
The UK must prepare a large number of new AI specialists within 5 years by measuring skills deficits, expanding university AI graduate capacity through industry partnerships, and increasing participation from underrepresented groups particularly women (22% of the AI workforce) via targeted initiatives. Government should expand education pathways through further education, apprenticeships, and employer-led upskilling; launch a flagship scholarship programme for up to 100 students; ensure the lifelong skills programme addresses AI labour market changes; and rapidly attract elite AI researchers through proactive recruitment. Government should establish internal headhunting capability, explore immigration pathways for graduates from leading institutions, address visa barriers, and expand Turing AI Fellowship funding by creating 15 new Pioneer Fellowships and committing funding for 25 additional Acceleration and World-Leading fellowships over 3 years.
1.4 Enabling safe and trusted AI development and adoption through regulation, safety and assurance
The UK should maintain a light-touch regulatory stance using well-designed rules and evaluation tools to enable safe AI adoption without suppressing innovation, particularly protecting vulnerable populations. Government must fund and expand the AI Safety Institute to evaluate advanced models and clarify frontier model regulation, modernize text and data mining rules to match EU standards, and ensure sector regulators have funding and capacity. Regulators must prioritize understanding AI impacts, use tools like regulatory sandboxes to accelerate AI in high-growth sectors, and publish annual reports on innovation enablement with transparent metrics; if evidence shows insufficient innovation promotion, government should consider establishing a central body with higher risk tolerance to issue pilot sandbox licences overriding sector regulations.
2 Change lives by embracing AI
4 sections
Outlines four subsections covering how government and the private sector should adopt AI to fulfil government missions, using a structured approach to test and scale solutions, reinforcing each other's roles through procurement and infrastructure, and removing barriers to private sector adoption.
2.1 AI Adoption is core to delivering the government’s missions
Widespread deployment of dependable, secure AI systems is essential for the government to achieve its five priority missions through enhanced service delivery and economic productivity. Government should strengthen foundational areas such as data infrastructure and skills while leveraging its position as a significant purchaser of AI systems and its ability to encourage businesses to adopt AI.
Adoption missions
2.2 Adopt a “Scan > Pilot > Scale” approach in government
Advocates that government employ a flexible methodology to identify AI opportunities, test them in pilots, and then scale successes across the public sector. Requires rethinking procurement to support the domestic AI ecosystem while moving quickly to learn from implementation.
Scan
Pilot
Scale
2.3 Enable public and private sectors to reinforce each other
Requires government to procure intelligently from AI suppliers with contract terms setting quality and performance standards, expose government data and systems through application programming interfaces to enable third-party integration, and establish a shared knowledge hub for best practices accessible to both sectors. Calls for identifying early adoption opportunities within 3 months.
2.4 Address private-sector-user-adoption barriers
Recognises that AI adoption could add £400 billion to the UK economy by 2030 and calls for using the Industrial Strategy to identify high-impact use cases, appointing sector champions in key industries, and targeting support for small and medium-sized enterprises and regional areas to ensure widespread adoption and address productivity disparities.
3 Secure our future with homegrown AI
The UK must develop leading frontier artificial intelligence companies through active state intervention, as control over increasingly capable systems will shape economic and national security outcomes, given computing power for frontier models has grown 10,000-fold since 2019 with similar growth expected by 2029. Government should establish UK Sovereign Artificial Intelligence to work with commercial partners to develop frontier capability, capture economic value, and foster globally competitive champions through ventures, startup support, or new company creation, prioritizing artificial intelligence for science given its economic and security value. UK Sovereign Artificial Intelligence should offer frontier companies and startups investment, computing infrastructure sites through Artificial Intelligence Growth Zones, curated government data and research access, support for attracting specialists and leaders, and national security engagement, securing economic returns and UK governance influence over advanced artificial intelligence in exchange.
Conclusion
This section outlines the government's commitment to implementing the Action Plan through long-term vision combined with immediate action, requiring investment in infrastructure and talent. The government must collaborate with industry, accept some risk in uncertain conditions, and maintain senior leadership focus to deliver transformational benefits for economic growth and public welfare across the UK.
Is this page useful?
Controls
What this policy requires, forbids, grants or penalises, one control at a time and in our own words. Each control links to the official text, which is the authority. This is not legal advice.
Drafted by the Index Drafted automatically from the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.
Themes covered 14 of 35
Choose a theme to see only its controls. About the themes
Risk and scope 1 of 4
- Risk classification, not covered
- Prohibited practices, not covered
- Impact assessment, not covered
Data and privacy 2 of 4
- Data subject rights and consent, not covered
- Cross-border data transfer, not covered
Transparency and records 1 of 4
- Labelling of synthetic content, not covered
- Technical documentation, not covered
- Record keeping and logging, not covered
Technical safeguards 3 of 4
- Human oversight, not covered
Assurance and lifecycle 0 of 4
- Conformity assessment and certification, not covered
- Registration and notification, not covered
- Incident and breach reporting, not covered
- Post-market monitoring and corrective action, not covered
Rights and protected groups 0 of 4
- Automated decision-making rights, not covered
- Children and vulnerable groups, not covered
- Biometric identification and surveillance, not covered
- Whistleblowing, complaints and redress, not covered
Roles and organisation 6 of 6
Enforcement and sectors 1 of 5
- Sector rules: health, not covered
- Sector rules: finance and insurance, not covered
- Sector rules: employment, not covered
- Sector rules: education, not covered
The full list of controls could not be loaded, so only the first 38 are shown.
1.1 2 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government should secure access to sufficient computing power for AI through publicly owned sovereign capacity, privately owned domestic capacity, and international partnerships.
- ObligationApplies to: Public authorityDrafted by the Index
Government should allocate sovereign AI compute to enable research, support academics and startups in training models, and maintain access to computing for critical services during market disruptions.
1.2 5 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government should explore using synthetic data generation to create privacy-preserving versions of highly sensitive datasets.
- ObligationApplies to: Public authorityDrafted by the Index
Government should strategically direct what new data is collected, particularly in areas that build on existing UK strengths such as health and disease research.
- ObligationApplies to: Public authorityDrafted by the Index
Government should create and issue standards and recommendations for releasing government datasets suitable for AI use, including effective data structures and distribution approaches.
- ObligationApplies to: Public authorityDrafted by the Index
Government should link compute resources to proprietary datasets as part of an incentive package for researchers and new companies establishing themselves in the UK.
- ObligationApplies to: Public authorityDrafted by the Index
Government should construct public sector data collection infrastructure and fund the development of new valuable datasets serving public sector and academic needs.
1.3 3 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government must assess the current size of the gap between supply and demand for AI professionals.
- ObligationApplies to: Public authorityDrafted by the Index
Government should support universities to develop new AI courses co-designed with industry partners and expand their teaching and recruitment capacity.
- ObligationApplies to: Public authorityDrafted by the Index
Government should increase diversity in the AI talent pipeline, including by increasing the proportion of women working in AI and data science.
1.4 3 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government should maintain the UK's pro-innovation regulatory approach while ensuring safety and assurance standards protect citizens from significant AI risks.
- ObligationApplies to: Public authorityDrafted by the Index
Government should finance the AI Safety Institute to expand its research on model evaluation, foundational safety and societal resilience, and maintain its world-leading position in pre-deployment frontier model assessment.
- ObligationApplies to: Public authorityDrafted by the Index
Government should urgently reform UK text and data mining intellectual property rules to achieve competitiveness matching the EU, addressing rights holder control and creative industry growth.
Scan 3 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government should appoint an AI lead within each mission delivery area to identify where AI could address mission-critical challenges, considering user needs from the start.
- ObligationApplies to: Public authorityDrafted by the Index
Government should establish a cross-government technical capability to track evolving AI capabilities and market developments, working with mission leads to coordinate expertise.
- ObligationApplies to: Public authorityDrafted by the Index
Government should engage in bilateral relationships with AI vendors and startups to understand upcoming product releases, anticipate new developments and signal public sector needs.
Pilot 7 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government should establish and refine a consistent framework for sourcing AI systems through in-house development, procurement, or innovation challenges, adapting it over time based on evidence and context.
- ObligationApplies to: Public authorityDrafted by the Index
Government should support open-source AI solutions that other organisations can adopt and design procurement processes to accommodate startups and innovators.
- ObligationApplies to: Public authorityDrafted by the Index
Government should build a rapid prototyping capability with technical and delivery staff to create and evaluate proofs of concept using internal AI expertise and external specialists.
- ObligationApplies to: Public authorityDrafted by the Index
Government should hire external AI talent, establish a senior technical civil servant career stream, set internal AI role pay at least 75% of private-sector levels, and use rigorous technical screening for recruitment.
- ObligationApplies to: Public authorityDrafted by the Index
Government should provide a data-rich experimental environment with streamlined access to datasets, language models and computational infrastructure.
- ObligationApplies to: Public authorityDrafted by the Index
Government should introduce a faster, multi-stage procurement process for AI with minimal bureaucracy at pilot scale and increasing oversight as investment size grows.
- ObligationApplies to: Public authorityDrafted by the Index
Government should compensate startups for completing rounds in competitive flexible procurement procedures.
Scale 4 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government should establish a scaling service to support successful pilots, providing senior backing and central funding to enable expansion beyond organisational boundaries toward national reach.
- ObligationApplies to: Public authorityDrafted by the Index
Government should conduct nationwide AI tenders focused on mission objectives to drive rapid adoption across decentralised systems, designed to encourage new market entrants.
- ObligationApplies to: Public authorityDrafted by the Index
Government should acquire or build a scalable AI technology platform capable of supporting both specialised narrow and large language models serving millions of citizen interactions across the UK.
- ObligationApplies to: Public authorityDrafted by the Index
Government should require standardised AI infrastructure, modular and reusable code design, and open-sourcing of code where feasible to promote infrastructure compatibility and code sharing.
2.3 4 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government should purchase AI systems using contract terms that establish performance and quality standards, incorporating regular technology review and adoption provisions to maintain flexibility within the changing AI market.
- ObligationApplies to: Public authorityDrafted by the Index
Government should make public sector data and systems available through application programming interfaces to enable third parties to integrate their own tools and create new services.
- ObligationApplies to: Public authorityDrafted by the Index
Government should establish a single shared hub where technical and non-technical users from public and private sectors can access documented frameworks, guidance, results, case studies and open-source resources on AI adoption.
- ObligationApplies to: Public authorityDrafted by the Index
The Digital Centre of Government should identify early adoption opportunities and quick wins to support the scan, pilot and scale approach within three months.
2.4 3 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government should use the Industrial Strategy to identify high-impact AI use cases and productivity opportunities in key industries and address regulatory barriers and skills gaps.
- ObligationApplies to: Public authorityDrafted by the Index
Government should appoint AI Sector Champions in key industries such as life sciences, financial services and creative industries to develop AI adoption plans with business and government.
- ObligationApplies to: Public authorityDrafted by the Index
Government should use local intermediaries and supply chain partnerships to promote widespread AI adoption across all regions, with particular support for small and medium-sized enterprises.
3 1 control
- ObligationApplies to: Public authorityDrafted by the Index
Government should use all available powers to ensure the UK develops national champion companies at the frontier of AI capabilities for national and economic security.
Conclusion 3 controls
- ObligationApplies to: Public authorityDrafted by the Index
Government must ensure long-term investment in physical infrastructure and human capital that will support all future AI development.
- ObligationApplies to: Public authorityDrafted by the Index
Government should take immediate action alongside its long-term strategy, set an example for the broader economy, work closely with industry, and accept some risk when facing uncertain conditions.
- ObligationApplies to: Public authorityDrafted by the Index
Government must maintain sustained high-level leadership attention and agency-wide commitment to driving rapid progress in delivering the Action Plan.
Change history
No changes recorded since we began monitoring this policy.