UK-National AI Strategy

National AI Strategy
Presented to Parliament
by the Secretary of State for Digital, Culture, Media and Sport
by Command of Her Majesty
September 2021
Command Paper 525

Version 1.2
2 1
Contents
Our ten-year plan to make Britain a global AI superpower 4 Pillar 2: Ensuring AI benefits all sectors and regions 40
Executive summary 7 AI deployment – understanding new dynamics 41
Summary of key actions 8 Creating and protecting Intellectual Property 42
Introduction 10 Using AI for the public benefit 42
10 Year Vision 11 Missions 44
The UK’s National AI Strategy 14 Net Zero 45
AI presents unique opportunities and challenges 16 Health 46
Reflecting and protecting society 16 The public sector as a buyer 46
The longer term 17 Pillar 3: Governing AI effectively 50
From Sector Deal to AI Strategy 18 Supporting innovation and adoption while protecting the public and building trust 51
Pillar 1: Investing in the long-term needs of the AI ecosystem 22 Alternative options 54
Skills and Talent 22 Regulators’ coordination and capacity 54
A new approach to research, development and innovation in AI 28 International governance and collaboration 55
International collaboration on Research & Innovation 30 AI and global digital technical standards 56
Access to data 30 AI Assurance 58
Data Foundations and Use in AI Systems 31 Public sector as an exemplar 59
Public sector data 32 AI risk, safety, and long-term development 60
Compute 33 Next steps 62
Finance and VC 35
Trade 36
Commercialisation 40
2 3
Our ten-year plan to National AI Strategy
make Britain a global This is the age of artificial intelligence.

Whether we know it or not, we all
AI superpower interact with AI every day - whether
it’s in our social media feeds and
smart speakers, or on our online
banking. AI, and the data that fuels
Over the next ten years, the impact of our algorithms, help protect us from
AI on businesses across the UK and fraud and diagnose serious illness.
the wider world will be profound - And this technology is evolving every
and UK universities and startups are day.
already leading the world in building
the tools for the new economy. We’ve got to make sure we keep up with
the pace of change. The UK is already a
New discoveries and methods for world leader in AI, as the home of
harnessing the capacity of machines to trailblazing pioneers like Alan Turing and
learn, aid and assist us in new ways Ada Lovelace and with our strong history
emerge every day from our universities of research excellence. This Strategy
and businesses. outlines our vision for how the UK can
maintain and build on its position as
AI gives us new opportunities to grow other countries also race to deliver their
and transform businesses of all sizes, own economic and technological
and capture the benefits of innovation transformations.
right across the UK.
The challenge now for the UK is to fully
As we build back better from the unlock the power of AI and data-driven
challenges of the global pandemic, and
KWASI KWARTENG technologies, to build on our early
NADINE DORRIES
prepare for new challenges ahead, we leadership and legacy, and to look
are presented with the opportunity to
SECRETARY OF STATE FOR forward to the opportunities of this
SECRETARY OF STATE FOR
supercharge our already admirable
BUSINESS, ENERGY AND coming decade.
DIGITAL, CULTURE, MEDIA
starting position on AI and to make these
INDUSTRIAL STRATEGY AND SPORT
technologies central to our development This National AI Strategy will signal to the
as a global science and innovation world our intention to build the most
superpower. pro-innovation regulatory environment
in the world; to drive prosperity across
With the help of our thriving AI the UK and ensure everyone can benefit
ecosystem and world leading R&D from AI; and to apply AI to help solve
system, this National AI Strategy will global challenges like climate change.
translate the tremendous potential of AI
into better growth, prosperity and social AI will be central to how we drive growth
benefits for the UK, and to lead the and enrich lives, and the vision set out in
charge in applying AI to the greatest our strategy will help us achieve both of
challenges of the 21st Century. those vital goals.
4 5
Executive summary
Artificial Intelligence (AI) is the fastest growing deep technology1 in the world, with huge
potential to rewrite the rules of entire industries, drive substantial economic growth and
transform all areas of life. The UK is a global superpower in AI and is well placed to lead the
world over the next decade as a genuine research and innovation powerhouse, a hive of
global talent and a progressive regulatory and business environment.
Many of the UK’s successes in AI were supported by the 2017 Industrial Strategy, which set out the
government’s vision to make the UK a global centre for AI innovation. In April 2018, the government
and the UK’s AI ecosystem agreed a near £1 billion AI Sector Deal to boost the UK’s global position as
a leader in developing AI technologies.
This new National AI Strategy builds on the UK’s strengths but also represents the start of a step-
change for AI in the UK, recognising the power of AI to increase resilience, productivity, growth and
innovation across the private and public sectors.
This is how we will prepare the UK for the next ten years, and is built on three assumptions about the
coming decade:
• The key drivers of progress, discovery and strategic advantage in AI are access to people, data,
compute and finance – all of which face huge global competition;
• AI will become mainstream in much of the economy and action will be required to ensure every
sector and region of the UK benefit from this transition;
• Our governance and regulatory regimes will need to keep pace with the fast-changing demands
of AI, maximising growth and competition, driving UK excellence in innovation, and protecting the
safety, security, choices and rights of our citizens.
The UK’s National AI Strategy therefore aims to:
• Invest and plan for the long-term needs of the AI ecosystem to continue our leadership as a
science and AI superpower;
• Support the transition to an AI-enabled economy, capturing the benefits of innovation in the
UK, and ensuring AI benefits all sectors and regions;
• Ensure the UK gets the national and international governance of AI technologies right to
encourage innovation, investment, and protect the public and our fundamental values.
This will be best achieved through broad public trust and support, and by the involvement of the
diverse talents and views of society.
7
Summary of key actions
Investing in the Long Term Needs of the AI Ecosystem Ensuring AI Benefits All Sectors and Regions Governing AI Effectively
■ Publish a framework for Government's role in enabling better data ■ Begin engagement on the Draft National Strategy for AI-driven ■ Publish the CDEI AI assurance roadmap
availability in the wider economy technologies in Health and Social Care, through the NHS AI Lab
■ Determine the role of data protection in wider AI governance
Short term ■ Consult on the role and options for a National Cyber-Physical ■ Publish the Defence AI Strategy, through the Ministry of Defence following the Data: A new direction consultation
(next 3 Infrastructure Framework
months): ■ Launch a consultation on copyright and patents for AI through ■ Publish details of the approaches the Ministry of Defence will
■ Support the development of AI, data science and digital skills the IPO use when adopting and using AI
through the Department for Education’s Skills Bootcamps
■ Develop an all-of-government approach to international AI
activity
■ Publish research into what skills are needed to enable employees ■ Publish research into opportunities to encourage diffusion of AI ■ Publish White Paper on a pro-innovation national position on
to use AI in a business setting and identify how national skills across the economy governing and regulating AI
provision can meet those needs
■ Consider how Innovation Missions include AI capabilities and
■ Evaluate the private funding needs and challenges of AI scaleups promote ambitious mission-based cooperation through bilateral ■ Complete an in-depth analysis on algorithmic transparency, with
and multilateral efforts a view to develop a cross-government standard
■ Support the National Centre for Computing Education to ensure AI
Medium programmes for schools are accessible ■ Extend UK aid to support local innovation in developing countries
term ■ Pilot an AI Standards Hub to coordinate UK engagement in AI
(next 6 ■ Support a broader range of people to enter AI-related jobs by ■ Build an open repository of AI challenges with real-world standardisation globally
months): ensuring career pathways highlight opportunities to work with or applications
develop AI
■ Establish medium and long term horizon scanning functions to
■ Implement the US UK Declaration on Cooperation in AI R&D increase government’s awareness of AI safety
■ Publish a review into the UK’s compute capacity needs to support

AI innovation, commercialisation and deployment
■ Roll out new visa regimes to attract the world's best AI talent to
the UK
■ Undertake a review of our international and domestic approach to ■ Launch joint Office for AI / UKRI programme to stimulate the ■ Explore with stakeholders the development of an AI technical
semiconductor supply chains development and adoption of AI technologies in high potential, standards engagement toolkit to support the AI ecosystem to
lower-AI-maturity sectors engage in the global AI standardisation landscape
■ Consider what open and machine-readable government
datasets can be published for AI models ■ Continue supporting the development of capabilities around ■ Work with partners in multilateral and multi-stakeholder fora, and
trustworthiness, adoptability, and transparency of AI invest in GPAI to shape and support AI governance in line with UK
■ Launch a new National AI Research and Innovation Programme technologies through the National AI Research and Innovation values and priorities
that will align funding programmes across UKRI and support the Programme
wider ecosystem ■ Work with The Alan Turing Institute to update guidance on AI
Long term ■ Join up across government to identify where using AI can provide ethics and safety in the public sector
(next 12 ■ Work with global partners on shared R&D challenges, a catalytic contribution to strategic challenges
months and leveraging Overseas Development Assistance to put AI at the heart ■ Work with national security, defence, and leading researchers to
beyond): of partnerships worldwide understand what public sector actions can safely advance AI
and mitigate catastrophic risks
■ Back diversity in AI by continuing existing interventions across top
talent, PhDs, AI and Data Science Conversion Courses and
Industrial Funded Masters
■ Monitor and use National Security and Investment Act to protect

national security while keeping the UK open for business
■ Include trade deal provisions in emerging technologies,

including AI
8 9
Introduction The UK’s National Artificial Intelligence Strategy aims to:
• Invest and plan for the long term needs of the AI ecosystem to continue our leadership as a
science and AI Superpower;
• Support the transition to an AI-enabled economy, capturing the benefits of innovation in the UK,
and ensuring AI benefits all sectors and regions;
• Ensure the UK gets the governance of AI technologies right to encourage innovation, investment,
and protect the public and our fundamental values.
This will be best achieved through broad public trust and support, and by the involvement of
Artificial Intelligence technologies (AI) prosperity and for shared global challenges the diverse talents and views of society.
offer the potential to transform the UK’s such as net zero, health resilience and
economic landscape and improve people’s environmental sustainability. AI capability is
lives across the country, transforming therefore vital for the UK's international
industries and delivering first-class public influence as a global science superpower.
services. 10 Year Vision research to building and shipping products. If
The National AI Strategy for the United you are a talented AI researcher from abroad,
AI may be one of the most important Kingdom will prepare the UK for the next ten Over the next decade, as transformative coming to the UK will be easier than ever
innovations in human history, and the years, and is built on three assumptions technologies continue to reshape our through the array of visa routes which are
government believes it is critical to both our about the coming decade: economy and society, the world is likely to see available.
economic and national security that the UK a shift in the nature and distribution of global
prepares for the opportunities AI brings, and • The key drivers of progress, discovery and power. We are seeing how, in the case of AI, If you run a business – whether it is a startup,
that the country is at the forefront of solving strategic advantage in AI are access to rapid technological change seeks to rebalance SME or a large corporate – the government
the complex challenges posed by an people, data, compute and finance – all of the science and technology dominance of wants you to have access to the people,
increased use of AI. which face huge global competition; existing superpowers like the US and China, knowledge and infrastructure you need to get
and wider transnational challenges demand your business ahead of the transformational
This country has a long and exceptional • AI will become mainstream in much of the greater collective action in the face of change AI will bring, making the UK a globally-
history in AI – from Alan Turing’s early work economy and action will be required to continued global security and prosperity. competitive, AI-first economy which benefits
through to DeepMind’s recent pioneering ensure every sector and region of the UK every region and sector.
discoveries. In terms of AI startups and benefit from this transition; With this in mind, the UK has an opportunity
scaleups, private capital invested and over the next ten years to position itself as By leading with our democratic values, the UK will
conference papers submitted, the UK sits in • Our governance and regulatory regimes the best place to live and work with AI; with work with partners around the world to make
the top tier of AI nations globally. The UK will need to keep pace with the fast- clear rules, applied ethical principles and a sure international agreements embed our ethical
ranked third in the world for private changing demands of AI, maximising pro-innovation regulatory environment. With values, making clear that progress in AI must be
investment into AI companies in 2020, behind growth and competition, driving UK the right ingredients in place, we will be both achieved responsibly, according to democratic
only the USA and China. excellence in innovation, and protecting a genuine innovation powerhouse and the norms and the rule of law.
the safety, security, choices and rights of most supportive business environment in the
The National AI Strategy builds on the UK’s our citizens. world, where we cooperate on using AI for And by increasing the number and diversity of
current strengths and represents the start of good, advocate for international standards people working with and developing AI, by
a step-change for AI in the UK, recognising This document sets out the UK’s strategic that reflect our values, and defend against the putting clear rules of the road in place and by
that maximising the potential of AI will intent at a level intended to guide action over malign use of AI. investing across the entire country, we will
increase resilience, productivity, growth and the next ten years, recognising that AI is a fast ensure the real-world benefits of AI are felt by
innovation across the private and public moving and dynamic area. Detailed and Whether it is making the decision to study AI, every member of society. Whether that is more
sectors. Building on our strengths in AI will measurable plans for the execution of the work at the cutting edge of research or spin accurate AI-enabled diagnostics in the NHS, the
take a whole-of-society effort that will span first stage of this strategy will be published up an AI business, our investments in skills, promise of driverless cars to make our roads
the next decade. This is a top-level economic, later this year. data and infrastructure will make it easier safer and smarter, or the hundreds of
security, health and wellbeing priority. The UK than ever to succeed. Our world-leading R&D unforeseen benefits that AI could bring to
government sees being competitive in AI as system will step up its support of innovators improve everyday life.
vital to our national ambitions on regional at every step of their journey, from deep
10 11
Introduction National AI Strategy
The National AI Strategy does not stand alone. • The forthcoming Digital Strategy, which The goals of this Strategy are that the The AI Council
It purposefully supports and amplifies the will build on DCMS's Ten Tech Priorities to UK:
other, interconnected work of government further set out the government’s The AI Council was established in 2019 to
including: ambitions in the digital sector; 1. Experiences a significant growth in both provide expert advice to the government and
the number and type of discoveries that high-level leadership of the AI ecosystem. The
• The Plan for Growth and recent • A new Defence AI centre as a keystone happen in the UK, and are AI Council demonstrates a key commitment
Innovation Strategy, which recognise piece of the modernisation of Defence; commercialised and exploited here; made in the AI Sector Deal, bringing together
the need to develop a diverse and respected leaders in their fields from across
inclusive pipeline of AI professionals with • The National Security Technology 2. Benefits from the highest amount of industry, academia and the public sector.
the capacity to supercharge innovation; Innovation exchange (NSTIx), a data economic and productivity growth due to Members meet quarterly to advise the Office
science & AI co-creation space that brings AI; and for AI and broader government on its current
• The Integrated Review, to find new together National Security stakeholders, priorities, opportunities and challenges for AI
paths for UK excellence in AI to deliver industry and academic partners to build 3. Establishes the most trusted and pro- policy.
prosperity and security at home and better national security capabilities; and innovation system for AI governance in
abroad, and shape the open international the world. In January 2021, the AI Council published its ‘AI
order of the future; • The upcoming National Resilience Roadmap’ providing 16 recommendations to
Strategy, which will in part focus on how This vision can be achieved if we build on the government on the strategic direction for
• The National Data Strategy, published the UK will stay on top of technological three pillars fundamental to the development AI. Its central call was for the government to
in September 2020, sets out our vision to threats. of AI: develop a National AI Strategy, building on the
harness the power of responsible data success of investments made through the AI
use to boost productivity, create new The government’s AI Council has played a 1. Investing in the needs of the ecosystem Sector Deal whilst remaining adaptable to
businesses and jobs, improve public central role in gathering evidence to inform to see more people working with AI, more future technological disruption. Since then, the
services, support a fairer society, and the development of this strategy, including access to data and compute resources to Council has led a programme of engagement
drive scientific discovery, positioning the through its roadmap published at the train and deliver AI systems, and access to with the wider AI community to inform the
UK as the forerunner of the next wave of beginning of the year, which represents a finance and customers to grow sectors; development of the National AI Strategy.
innovation; valuable set of recommendations reflecting
much of the wider AI community in the UK. 2. Supporting the diffusion of AI across the To guide the delivery and implementation of
• The Plan for Digital Regulation, which The wider ecosystem also fed in through a whole economy to ensure all regions, this strategy the government will renew and
sets out our pro-innovation approach to survey run by the AI Council in collaboration nations, businesses and sectors can strengthen the role of the AI Council, ensuring
regulating digital technologies in a way with The Alan Turing Institute. The benefit from AI; and it continues to provide expert advice to
that drives prosperity and builds trust in government remains grateful to the AI Council government and high-level leadership of the AI
their use; for its continued leadership of the AI 3. Developing a pro-innovation regulatory ecosystem.
ecosystem, and would like to thank those and governance framework that protects
• The upcoming National Cyber Strategy from across the United Kingdom who shared the public.
to continue the drive for securing their views during the course of developing
emerging technologies, including building this strategy.
security into the development of AI;
12 13
The UK’s National AI Strategy

Vision
To remain an AI and science superpower fit for the next decade
Strong
Benefits of AI UK maintains
Growth in the Protect and domestic AI
Impacts
adoption its position as a

UK’s AI sector, further capabilities to
shared across global leader in
contributing to fundamental address
every region AI research &
UK GDP growth UK values National
and sector development
Security issues
Pillar 1: Investing in the Pillar 2: Ensuring AI Pillar 3: Governing AI

long term needs of the AI benefits all sectors and effectively
ecosystem regions
A growing UK Increased diversity in Certainty for the UK AI

supplier base applied AI ecosystem
Reduced competition
Wider AI adoption in
for AI skills
industries & regions Improved public trust
Outcomes
in AI
New AI scientific
breakthroughs
Greater UK AI exports
Greater workforce
Increased responsible
diversity
innovation
Public Sector as
exemplar for AI
Applied AI technologies
procurement & ethics
to new use cases
UK maintains its
Increased investment Greater public value position as a global
in UK AI companies for money leader in AI
Activities
Government activity in this strategy and over the next 10 years
14 15
AI presents unique
opportunities and
challenges
‘Artificial Intelligence’ as a term can mean • In regulatory matters, a system’s lives, there are at least three reasons we The longer term
a lot of things, and the government autonomy raises unique questions should care about diversity in our AI
recognises that no single definition is around liability and fairness as well as ecosystem: Making specific predictions about the future
going to be suitable for every scenario. In risk and safety – and even ownership of impact of a technology – as opposed to the
general, the following definition is creative content3 – in a way which is • MORAL: As AI becomes an organising needs of those developing and using it today –
sufficient for our purposes: “Machines that distinct to AI, and these questions principle which creates new opportunities has a long history in AI. Since the 1950s various
perform tasks normally requiring human increase with the relative complexity of and changes the shape of industries and hype cycles have given way to so-called ‘AI
intelligence, especially when the machines the algorithm. There are also questions of the competitive dynamics across the winters’ as the promises made have perpetually
learn from data how to do those tasks.” transparency and bias which arise from economy, there is a moral imperative to remained ‘about 20 years away’.
The UK government has also set out a decisions made by AI systems. ensure people from all backgrounds and
legal definition of AI in the National parts of the UK are able to participate and While the emergence of Artificial General
Security and Investment Act.2 • There are often greater infrastructural thrive in this new AI economy. Intelligence (AGI) may seem like a science fiction
requirements for AI services than in concept, concern about AI safety and non-
Much like James Watt’s 1776 steam engine, AI cloud/Software as a Service systems. In • SOCIAL: AI systems make decisions human-aligned systems4 is by no means
is a ‘general purpose technology’ (or more building and deploying some models, based on the data they have been trained restricted to the fringes of the field.5 The
accurately, technologies) that has many access to expensive high performance on. If that data - or the system it is government’s first focus is on the economic and
possible applications, and we expect them to computing and/or large data sets is embedded in - is not representative, it social outcomes of autonomous and adaptive
have a transformational impact on the whole needed. risks perpetuating or even cementing new systems that exist today. However, we take the
economy. Already, AI is used in everyday forms of bias in society. It is therefore firm stance that it is critical to watch the
contexts like email spam filtering, media • Multiple skills are required to develop, important that people from diverse evolution of the technology, to take seriously the
recommendation systems, navigation apps, validate and deploy AI systems, and the backgrounds are included in the possibility of AGI and ‘more general AI’, and to
payment transaction validation and commercialisation and product development and deployment of AI actively direct the technology in a peaceful,
verification, and many more. AI technologies journey can be longer and more systems. human-aligned direction.6
will impact the whole economy, all of society expensive because so much starts with
and us as individuals. fundamental R&D. • ECONOMIC: There are big economic The emergence of full AGI would have a
benefits to a diverse AI ecosystem. These transformational impact on almost every aspect
Many of the themes in AI policy are similar to include increasing the UK’s human capital of life, but there are many challenges which
tech and digital policy more widely: the from a diverse labour supply, creating a could be presented by AI which could emerge
commercialisation journeys; the reliance on Reflecting and protecting society wider range of AI services that stimulate much sooner than this. As a general purpose
internationally mobile talent; the importance demand, ensuring the best talent is technology AI will have economic and social
of data; and consolidation of economic AI makes predictions and decisions, and fulfils discovered from the most diverse talent impacts comparable to the combustion engine,
functions onto platforms. However there are tasks normally undertaken by humans. While pool. the car, the computer and the internet. As each
some key examples of differences derived diverse opinions, skills, backgrounds and of these has disrupted and changed the shape of
from the above definition which differentiate experience are hugely important in designing the world we live in - so too could AI, long before
AI and require a unique policy response from any service – digital or otherwise – it is any system ‘wakes up.’
the government. particularly important in AI because of the
executive function of the systems. As AI
increasingly becomes an enabler for
transforming the economy and our personal
16 17
AI presents unique opportunities and challenges National AI Strategy
AlphaFold & AlphaFold 2
In November 2020, London-based DeepMind announced that they had solved one of the longest
running modern challenges in biology: predicting how proteins - the building blocks of life which
underpin every biological process in every living thing - take shape, or ‘fold’.
The choices that are made in the here and global destination for mobile AI researchers
now to develop AI will shape the future of after the USA.
humanity and the course of international
affairs. For example, whether AI is used to The government has invested more than £2.3
enhance peace, or a cause for war; whether billion into Artificial Intelligence across a range
AI is used to strengthen our democracies, or of initiatives since 2014.10 This portfolio of
embolden authoritarian regimes. As such we investment includes, but is not limited to:
have a responsibility to not only look at the
extreme risks that could be made real with • £250 million to develop the NHS AI Lab at
AGI, but also to consider the dual-use threats NHSX to accelerate the safe adoption of
we are already faced with today. Artificial Intelligence in health and care;
• £250 million into Connected and

Autonomous Mobility (CAM) technology
From Sector Deal to AI Strategy through the Centre for Connected and
Autonomous Vehicles (CCAV) to develop
The UK is an AI superpower, with particular the future of mobility in the UK;
strengths in research, investment and
innovation. The UK’s academic and • 16 new AI Centres for Doctoral Training at
commercial institutions are well known for universities across the country, backed by
conducting world-leading AI research, and the up to £100 million and delivering 1,000
UK ranks 3rd in the world for AI publication new PhDs over five years;
citations per capita.7 This research strength
was most recently demonstrated in • A new industry-funded AI Masters AlphaFold, DeepMind’s deep learning AI system, broke all previous accuracy levels dating back over
November 2020 when DeepMind, a UK-based programme and up to 2,500 places for AI 50 years, and in July 2021 the organisation open sourced the code for AlphaFold together with over
AI company, used AlphaFold to find a solution and data science conversion courses. This 350,000 protein structure predictions, including the entire human proteome, via the AlphaFold
to a 50-year-old grand challenge in biology.8 includes up to 1,000 government-funded database in partnership with EMBL-EBI.
scholarships;
The UK has the 3rd highest number of AI DeepMind’s decision to share this knowledge openly with the world, demonstrates both the
companies in the world after the US and • Investment into The Alan Turing Institute
opportunity that AI presents, as well as what this strategy seeks to support: bleeding-edge research
China. Alongside DeepMind, the UK is home and over £46 million to support the
to Graphcore, a Bristol-based machine Turing AI Fellowships to develop the next happening in the UK and with partners around the world, solving big global challenges.
learning semiconductor company; Darktrace, generation of top AI talent;
a world-leading AI company for cybersecurity; AlphaFold opens up a multitude of new avenues in research – helping to further our understanding
and BenevolentAI, a company changing the • Over £372 million of investment into UK of biology and the nature of the world around us. It also has a multitude of potential real-world
way we treat disease. The UK also attracts AI companies through the British applications, such as deepening our understanding of how bacteria and viruses attack the body in
some of the best AI talent from around the Business Bank for the growing AI sector; order to develop more effective prevention and treatment, or support the identification of proteins
world9 – the UK was the second most likely
and enzymes that can break down industrial or plastic waste.
18 19
AI presents unique opportunities and challenges National AI Strategy
• £172 million of investment through the the UK has a globally competitive R&D and
UKRI into the Hartree National Centre for industrial strength11 and has been widely
Digital Innovation, leveraging an additional cited as a set of technologies in which the UK
£38 million of private investment into must maintain a leading edge to guarantee
High Performance Computing. our continued security and prosperity in an
intensifying geopolitical landscape.
Further investments have been made into the
Tech Nation Applied AI programme – now in
its third iteration; establishing the Office for
National Statistics Data Science Campus; the
Crown Commercial Service’s public sector AI
procurement portal; and support for the
Department for International Trade attracting
AI related Foreign Direct Investment into the
UK.
As part of the AI Sector Deal, the government

established the AI Council to bring together
respected leaders to strengthen the
conversation between academia, industry,
and the public sector. The Office for Artificial
Intelligence was created as a new team within
government to take responsibility for
overarching AI policy across government and
to be a focal point for the AI ecosystem
through its secretariat of the AI Council. The
Centre for Data Ethics and Innovation (CDEI)
was established as a government expert body
focused on the trustworthy use of data and AI
in the public and private sector.
This strategy builds on the recent history of

government support for AI and considers the
next key steps to harness its potential in the
UK for the coming decade. In doing so, the
National AI Strategy leads on from the
ambitions outlined in the government’s
Innovation Strategy to enable UK businesses
and innovators to respond to economic
opportunities and real-world problems
through our national innovation prowess. AI
was identified in the Innovation Strategy as
one of the seven technology families where
20 21
Pillar 1:
Investing in the long-term
needs of the AI ecosystem
Investing in and planning for the long term needs of the AI ecosystem to remain a Increasing diversity and closing the skills gap through postgraduate conversion courses in
science and AI superpower data science and artificial intelligence
To maintain the UK’s position amongst the

Government’s aim is to greatly increase global AI superpowers and ensure the UK
the type, frequency and scale of AI continues to lead in the research,
discoveries which are developed and development, commercialisation and
exploited in the UK. deployment of AI, we need to invest in, plan
for, secure and unlock the critical inputs that
This will be achieved by: underpin AI innovation.
• Making sure the UK’s research, Skills and Talent

development and innovation system
continues to be world leading, providing Continuing to develop, attract and train
the support to allow researchers and the best people to build and use AI is at
entrepreneurs to forge new frontiers in the core of maintaining the UK’s world-
AI; leading position. By inspiring all with the
possibilities AI presents, the UK will
• Guaranteeing that the UK has access to a continue to develop the brightest, most
diverse range of people with the skills diverse workforce.
needed to develop the AI of the future
and to deploy it to meet the demands of Building a tech-savvy nation by supporting As a result of the growing skills gap in AI and data science, 2,500 new Masters conversion courses in
the new economy; skills for the future is one of the government’s
AI and data science are now being delivered across universities in England. The conversion course
ten tech priorities. The gap between demand
• Ensuring innovators have access to the and supply of AI skills remains significant and programme included up to 1,000 scholarships to increase the number of people from
data and computing resources necessary growing,12,13 despite a number of new AI skills underrepresented groups and to encourage graduates from diverse backgrounds to consider a
to develop and deliver the systems that initiatives since the 2018 AI Sector Deal. In future in AI and Data Science.
will drive the UK economy for the next order to meet demand, the UK needs a larger
decade; workforce with AI expertise. Last year there In the first year over 1,200 students enrolled, with 22% awarded scholarships. Over 40% of the total
was a 16% increase for online AI and Data students are women, one quarter are black students and 15% of students are disabled. 70% of the
• Supporting growth for AI through a pro- Science job vacancies and research found
total students are studying on courses based outside of London and the South East.
innovation business environment and that 69% of vacancies were hard to fill.14 Data
capital market, and attracting the best from an ecosystem survey conducted by the
people and firms to set up shop in the AI Council and The Alan Turing Institute These conversion courses are providing the opportunity to develop new digital skills or retrain to
UK; showed that 81% of respondents agreed help find new employment in the UK’s cutting-edge AI and data science sectors, ensuring that
there were significant barriers in recruiting industry and the public sector can access the greatest supply of talent across the whole country.
• Ensuring UK AI developers can access and retaining top AI talent in their domain
markets around the world. within the UK.
Research into the AI Labour Market showed

that technical AI skill gaps are a concern for
many firms, with 35% of firms revealing that a
22 23
Pillar 1: Investing in the long-term needs of the AI ecosystem National AI Strategy
lack of technical AI skills from existing Government will seek to build upon the £46 developing AI, will be key to ensuring the UK’s
employees had prevented them from meeting ‘Understanding the UK AI Labour Market’ million Turing AI Fellowships investment to success in developing AI and investing in the
their business goals, and 49% saying that a research attract, recruit, and retain a substantial cohort long term health of our AI ecosystem.
lack of required AI skills from job applicants of leading researchers and innovators at all
also affected their business outcomes.15 To In 2021, the Office for AI published research career stages. Our approach will enable Use: Empower employers and employees to
support the adoption of AI we need to ensure to investigate Artificial Intelligence and Data Fellows to work flexibly between academia upskill and understand the opportunities for
that non-technical employees understand the science skills in the UK labour market in 2020. and other sectors, creating an environment using AI in a business setting
opportunities, limitations and ethics of using Some key findings from the research: for them to discover and develop cutting edge
AI in a business setting, rather than these AI technologies and drive the use of AI to The AI Council ecosystem survey found that
being the exclusive domain of technical • Half of surveyed firms’ business plans had address societal, economic and only 18% agreed there was sufficient
practitioners. been impacted by a lack of suitable environmental challenges in the UK. We note provision of training and development in AI
candidates with the appropriate AI that recently, research breakthroughs in the skills available to the current UK workforce. As
We need to inspire a diverse set of people knowledge and skills. field of AI have been disproportionately driven the possibilities to develop and use AI grow,
across the UK to ensure the AI that is built by a small number of luminary talents and so will people's need to understand and apply
and used in the UK reflects the needs and • Two thirds of firms (67%) expected that their trainees. In line with the Innovation AI in their jobs. This will range from people
make-up of society. To close the skills gap, the the demand for AI skills in their Strategy, the government affirms our working adjacent to the technical aspects
government will focus on three areas to organisation was likely to increase in the commitment to empowering distinguished such as product managers and compliance,
attract and train the best people: those who next 12 months. academics. through to those who are applying AI within
build AI, those who use AI, and those we their business, such as in advertising and HR.
want to be inspired by AI. • Diversity in the AI sector was generally Research16 and industry engagement has Below degree level, there is a need to clearly
low. Over half of firms (53%) said none of demonstrated the need for graduates with articulate the skills employers and employees
Build: Train and attract the brightest and best their AI employees were female, and 40% business experience, indicating a need to need to use AI effectively in the workplace.
people at developing AI said none were from ethnic minority continue supporting industry/academic For example, industries have expressed their
backgrounds. partnerships to ensure graduates leave willingness to fund employees to undertake
To meet the demand seen in industry and education with business-ready experience. training but have not found training that suits
academia, the government will continue • There were over 110,000 UK job Our particular focus will be on software their needs: including training that is
supporting existing interventions across top vacancies in 2020 for AI and Data Science engineers, data scientists, data engineers, business-focused, modular and flexible.
talent, PhDs and Masters levels. This includes roles. machine learning engineers and scientists,
Turing Fellowships, Centres for Doctoral product managers, and related roles.
Training and Postgraduate Industrial-Funded The findings from this research will help the
Masters and AI Conversion Courses. Office for AI address the AI skills challenge We recognise that global AI talent is scarce,
and ensure UK businesses can take and the topic of fierce competition
advantage of the potential of AI and Data internationally. As announced in the
Science. Innovation Strategy, the government is
revitalising and introducing new visa routes
that encourage innovators and entrepreneurs
to the UK. Support for diverse and inclusive
researchers and innovators across sectors,
and new environments for collaboratively
24 25
Attracting the best AI talent from around the world
The UK is already the top global destination for AI graduates in the United States and we punch above our
weight globally in attracting talent. The UK nearly leads the world in its proportion of top-skilled AI researchers.
Government wants to take this to the next level and make the UK the global home for AI researchers,
entrepreneurs, businesses and investors.
As well as ensuring the UK produces the next generation of AI talent we need, the government is broadening the
routes that talented AI researchers and individuals can work in the UK, through the recently announced
Innovation Strategy.
• The Global Talent visa route is open to those who are leaders or potential leaders in AI - and those who
have won prestigious global prizes automatically qualify. Government is currently looking at how to broaden
this list of prizes. explore how these needs can be met and
Skills for Jobs White Paper mainstreamed through national skills
• A new High Potential Individual route will make it as simple as possible for internationally mobile provision.
individuals who demonstrate high potential to come to the UK. Eligibility will be open to applicants who have The Skills for Jobs: Lifelong Learning for
graduated from a top global university, with no job offer requirement. This gives individuals the flexibility to Opportunity and Growth White Paper was The government will also support people to
work, switch jobs or employers – keeping pace with the UK’s fast-moving AI sector. published in January 2021 and is focused on develop skills in AI, machine learning, data
giving people the skills they need, in a way science and digital through the Department
• A new scale-up route will support UK scale-ups by allowing talented individuals with a high-skilled job offer that suits them, so they can get great jobs in for Education’s Skills Bootcamps. The
from a qualifying scale-up at the required salary level to come to the UK. Scaleups will be able to apply sectors the economy needs and boost the Bootcamps are free, flexible courses of up to
through a fast-track verification process to use the route, so long as they can demonstrate an annual country’s productivity. 16 weeks, giving adults aged 19 and over the
average revenue or employment growth rate over a three-year period greater than 20%, and a minimum of opportunity to build up in-demand, sector-
10 employees at the start of the three-year period. These reforms aim to ensure that people can specific skills and fast-track to an interview
access training and learning flexibly with a local employer; improving their job
• A revitalised Innovator route will allow talented innovators and entrepreneurs from overseas to start and throughout their lives and that they are well- prospects and supporting the economy.
operate a business in the UK that is venture-backed or harnesses innovative technologies, creating jobs for informed about what is on offer, including
UK workers and boosting growth. We have reviewed the Innovator route to make it even more open to: opportunities in valuable growth sectors. This Inspire: Support all to be excited by the
will also involve reconfiguring the skills system possibilities of AI
• Simplifying and streamlining the business eligibility criteria. Applicants will need to demonstrate that their to give employers a leading role in delivering
business venture has a high potential to grow and add value to the UK and is innovative. the reforms and influencing the system to The AI Council’s Roadmap makes clear that
generate the skills they need to grow. inspiring those who are not currently using AI,
• Fast-tracking applications. The UK government is exploring a fast-track, lighter touch endorsement and allowing children to explore and be
process for applicants whose business ideas are particularly advanced to match the best-in-class amazed by the potential of AI, will be integral
international offers. Applicants that have been accepted on to the government’s Global Entrepreneur to ensuring we continue to have a growing
Programme will be automatically eligible. To more effectively use AI in a business and diverse AI-literate workforce.
setting, employees, including those who
• Building flexibility. Applicants will no longer be required to have at least £50,000 in investment funds to would not have traditionally engaged with AI, Through supporting the National Centre for
apply for an Innovator visa, provided that the endorsing body is satisfied the applicant has sufficient will require a clear articulation of the different Computing Education (NCCE) the government
funds to grow their business. We will also remove the restriction on doing work outside of the skills required, so they can identify what will continue to ensure programmes that
applicant’s primary business. training already exists and understand if there engage children with AI concepts are
is still a gap. accessible and reach the widest demographic.
• The new Global Business Mobility visa will also allow overseas AI businesses greater flexibility in
transferring workers to the UK, in order to establish and expand their business here. Using the Skills Value Chain approach piloted The Office for AI will also work with the
by the Department for Education,17 the Department for Education to ensure career
These reforms will sit alongside the UK government’s Global Entrepreneur Programme (GEP) which has a government will help industry and providers pathways for those working with or
track record of success in attracting high skilled migrant tech founders with IP-rich businesses to the UK. The to identify what skills are needed. Lessons developing AI are clearly articulated on career
programme will focus on attracting more international talent to support the growth of technology clusters learned from this pilot will support this work guidance platforms, including the National
including through working with academic institutions from overseas to access innovative spinouts and overseas to help businesses adopt the skills needed to Careers Service, demonstrating role models
talent. get the best from AI. The Office for AI will then and opportunities to those exploring AI. This
work with the Department for Education to will support a broader range of people to
Through the Graduate Route we are also granting international students with UK degrees 2 years, 3 years for
those with PhDs, to work in the UK post-graduation. This will help ensure that we can attract the best and
brightest from across the world while also giving students time to work on the most challenging AI problems.
These are all in addition to our existing skills visa schemes for those with UK job offers.
26 27
The National AI Research and Innovation (R&I) Programme has five main aims:
• Discovering and developing transformative new AI technologies, leading the world in the development
of frontier AI and the key technical capabilities to develop responsible and trustworthy AI. The programme
will support:
• foundational research to develop novel next generation AI technologies and approaches which could
address current limitations of AI, focusing on low power and sustainable AI, and AI which can work
differently with a diverse range of challenging data sets, human-AI interaction, reasoning, and the maths
underpinning the theoretical foundations of AI.
• technical and socio-technical capability development to overcome current limitations around the
responsible trustworthy nature of AI.
consider careers in AI. The government will At the same time, the field of AI is advancing
ensure that leaders within the National AI rapidly, with breakthrough innovations being • Maximising the creativity and adventure of researchers and innovators, building on UK strengths and
Research and Innovation Programme will play generated by a diverse set of institutions and developing strategic advantage through a diverse range of AI technologies. The programme will support:
a key role in engaging with the public and countries. The past decade has seen the rise
inspiring the leaders of the future. of deep learning, compute-intensive models, • specific routes to enable the exploration of high-risk ideas in the development and application of AI;
routine deployment of vision, speech, and
A new approach to research, language modelling in the real world, the • follow-on funding to maximise the impact of the ideas with the most potential.
development and innovation in AI emergence of responsible AI and AI safety,
among other advances. These are being • Building new research and innovation capacity to deliver the ideas, technologies, and workforce of
Our vision is that the UK builds on our developed by new types of research labs in the future, recruiting and retaining AI leaders, supporting the development of new collaborative AI
excellence in research and innovation in the private companies and public institutions ecosystems, and developing collaborative, multidisciplinary, multi-partner teams. The programme will
next generation of AI technologies. around the world. We expect that the next support:
decade will bring equally transformative
The UK has been a leader in AI research since breakthroughs. Our goal is to make the UK • the recruitment, retention, training and development of current and future leaders in AI, and flexible
it developed as a field, thanks to our the starting point for a large proportion of working across sectoral and organisational interfaces using tools such as fellowships, and building on
strengths in computational and mathematical them, and to be the fastest at turning them the success of the Turing AI Fellowships scheme;
sciences.18 The UK’s AI base has been built into benefits for all.
upon this foundation,19 and the recently • enhanced UK capacity in key AI professional skills for research and innovation, such as data scientists
announced Advanced Research and Invention To do this, UKRI will support the and software engineers.
Agency (ARIA) will complement our efforts to transformation of the UK’s capability in AI
cement our status as a global science by launching a National AI Research and • Connecting across the UK AI Research and Innovation ecosystem, building on the success of The Alan
superpower. The UK also has globally Innovation (R&I) Programme. The Turing Institute as the National Centre for AI and Data Science, and building collaborative partnerships
recognised institutes such as The Alan Turing programme will shift us from a rich but siloed nationally and regionally between and across sectors, diverse AI research and innovation stakeholders. The
Institute and the high-performing universities and discipline-focused national AI landscape programme will support:
which are core to research in AI.20 to an inclusive, interconnected, collaborative,
and interdisciplinary research and innovation • the development of a number of nationally distributed AI ecosystems which enable researchers and
Currently, AI research undertaken in the UK is ecosystem. It will work across all the Councils innovators to collaborate in new environments and integrate basic research through application and
world class, and investments in AI R&D of UKRI and will be fully-joined up with innovation. These ecosystems will be networked into a national AI effort with the Alan Turing Institute as
contribute to the Government’s target of business of all sizes and government its hub, convening and coordinating the national research and innovation programme and enabling
increasing overall public and private sector departments. It will translate fundamental business and government departments to access the UK’s AI expertise and skills capability e.g. the
R&D expenditure to 2.4% of GDP by 2027. But scientific discoveries into real-world AI catapult network and compute capability.
generating economic and societal impact applications, address some limitations in the
through adoption and diffusion of AI ability of current AI to be effectively used in • Supporting the UK's AI Sector and the adoption of AI, connecting research and innovation and
technologies is behind where it could be.21 numerous real world contexts, such as supporting AI adoption and innovation in the private sector. The programme will support:
There is a real opportunity to build on our tackling complex and undefined problems,
existing strengths in fundamental AI research and explore using legacy data such as non- • challenge-driven AI research and innovation programmes in key UK priorities, such as health and the
to ensure they translate into productive digital public records. transition to net zero;
processes throughout the economy.
• collaborative work with the public sector and government organisations to facilitate leading researchers
and innovators engaging with the AI transformation of the public sector;
• innovation activities in the private sector, both in terms of supporting the development of the UK’s
burgeoning AI sector and the adoption of AI across sectors.
28 29
International collaboration on Access to data The AI Council and the Ada Lovelace Institute professionals such as data engineers have a
Research & Innovation recently explored three legal mechanisms key role to play in opening up access to the
The National Data Strategy sets out the that could help facilitate responsible data most critical data and compute
As well as better coordination at home, the government's approach to unlocking the stewardship – data trusts, data cooperatives infrastructures on FAIR data principles, and in
UK will work with friends and partners around power of data. Access to good quality, and corporate and contractual mechanisms. accelerating the pathway to using AI
the world on shared challenges in research representative data from which AI can learn is The ongoing Data: A new direction technologies to make best use of the UK’s
and development and lead the global critical to the development and application of consultation asks what role the government healthy data ecosystem.
conversation on AI. robust and effective AI systems. should have in enabling and engendering
confidence in responsible data intermediary Data foundations are crucial to the effective
The UK will participate in Horizon Europe, The AI Sector Deal recognised this and since activity. The government is also exploring use of AI and it is estimated that, on average,
enabling collaboration with other European then the government has established how privacy enhancing technologies can 80% of the time spent on an AI project is
researchers, and will build a strong and varied evidence on which to make policies to remove barriers to data sharing by more cleaning, standardising and making the data
network of international science and harness the positive economic and social effectively managing the risks associated fit for purpose. Furthermore, when the source
technology partnerships to support R&I benefit of increased availability of data. This with sharing commercially sensitive and data needed to power AI or machine learning
collaboration. By shaping the responsible use includes the Open Data Institute’s original personal data. is not fit for purpose, it leads to poor or
of technology, we will put science and research into data trusts as a model of data inaccurate results, and to delays in realising
technology, including AI, at the heart of our stewardship to realise the value of data for AI. Data Foundations and Use in AI Systems the benefits of innovation.22 Poor quality
alliances and partnerships worldwide. We will The research established a repeatable model datasets can also be un-representative,
continue to use Official Development for data trusts which others have begun to Data foundations refer to various especially when it comes to minority groups,
Assistance to support R&D partnerships apply. characteristics of data that contribute to its and this can propagate existing biases and
with developing countries. overall condition, whether it is fit for purpose, exclusions when they are used for AI.
Mission 1 of the National Data Strategy seeks recorded in standardised formats on modern,
We are also deepening our collaboration with to unlock the value of data across the future-proof systems and held in a condition The government is looking to support action
the United States, implementing the US UK economy, and is a vital enabler for AI. This that means it is findable, accessible, to mitigate the effects of quality issues and
Declaration on Cooperation in AI Research mission explores how the government can interoperable and reusable (FAIR). A recent EY underrepresentation in AI systems. Subject
and Development. This declaration outlines apply six evidenced levers to tackle barriers to study delivered on behalf of DCMS has found to the outcomes of the Data: A new
a shared vision for driving technological data availability. The government will that organisations that report higher AI direction consultation, the government
breakthroughs in AI between the US and the publish a policy framework in Autumn adoption levels also have a higher level of will more explicitly permit the collection
UK. As we build materially on this partnership, 2021 informed by the outcomes of Mission data foundations. and processing of sensitive and protected
we will seek to enable UK partnership with 1, setting out its role in enabling better characteristics data to monitor and
other key global actors in AI, to grow data availability in the wider economy. The The government is considering how to mitigate bias in AI systems.
influential R&I collaborations. policy framework includes supporting the improve data foundations in the private and
activities of intermediaries, including data third sectors. Through the National AI R&I An important outcome for increasing access
trusts, and providing stewardship services Programme and ambitions to lead best to data and improving data foundations is in
between those sharing and accessing data. practices in FAIR data, we will grow our how technology will be better able to use that
capacity in professional AI, software and data data. Technological convergence – the
skills, and support the development of key tendency for technologies that were originally
new data infrastructure capabilities. Technical unrelated to become more closely integrated
(or even unified) as they advance – means
30 31
that AI will increasingly be deployed together to drive efficiency and improve public We need to strengthen the way that public Digital Catapult’s Machine Intelligence
with many other technologies of the future, services. The Central Digital and Data Office authorities can engage with private sector Garage
unlocking new technological, economic and (CDDO) has been created within the Cabinet data providers to make better use of data
social opportunities. For example, AI is a Office to consolidate the core policy and through FAIR data and open standards, For more than three years, Digital Catapult’s
necessary driver of the development of strategy responsibilities for data foundations, including making government data more Machine Intelligence Garage (MI Garage) has
robotics and smart machines, and will be a and will work with expert cross-sector easily available through application helped startups accelerate the development
crucial enabling technology for digital twins. partners to improve government’s use and programming interfaces (APIs), and of their industry-leading AI solutions by
These digital replicas of real-world assets, reuse of data to support data-driven encouraging businesses to offer their data addressing their need for computational
processes or systems, with a two-way link to innovation across the public sector. through APIs. Government will continue to power.
sensors in the physical world, will help make publish authoritative open and machine-
sense of and create insights and value from The CDDO also leads on the Open readable data on which AI models for both Some AI solutions being developed require
vast quantities of data in increasingly Government policy area, a wide-ranging and public and commercial benefit can greater computing capacity in the form of
sophisticated ways. And in the future, some open engagement programme that entails depend. The Office for AI will also work High Performance Computers (HPC) for
types of AI will rely on the step-change in ongoing work with Civil Society groups and with teams across government to consider unusually large workloads (such as weather
processing power that quantum computing is government departments to target new kinds what valuable datasets government simulation, protein folding and simulation of
expected to unlock. of data highlighted as having 'high potential should purposefully incentivise or curate molecular interactions) or access to AI
impact' for release as open data. The UK’s that will accelerate the development of focussed hardware like Graphcore’s
Government will consult later this year on ongoing investment in open data will serve to valuable AI applications. Intelligence Processing Unit (IPU), a new
the potential value of and options for a UK further bolster the use of AI and machine processor specifically designed for developing
capability in digital twinning and wider learning within government, the private Compute AI. MI Garage provides a channel through
‘cyber-physical infrastructure.’23 This sector, and the third sector. The application of which startups can connect with HPC centres
consultation will help identify how common, standards and improvements to the quality of Access to computing power is essential to the and access specialised hardware. HPC
interoperable digital tools and platforms, as data collected, processed, and ultimately development and use of AI, and has been a partners include the Hartree National Centre
well as physical testing and innovation spaces released publicly under the Open dominant trend in AI breakthroughs of the for Digital Innovation, the Edinburgh Parallel
can be brought together to form a digital and Government License will create further value past decade. The computing power Computing Centre, and the Earlham Institute.
physical shared infrastructure for innovators when used by organisations looking to train underpinning AI in the UK comes from a MI Garage has also worked with NVIDIA,
(e.g. digital twins, test beds and living labs). and optimise AI systems utilising large range of sources. The government’s recent Graphcore and LightOn to facilitate access to
Supporting and enabling this shared amounts of information. report on large-scale computing24 recognises special trials to lower the barrier to entry to AI
infrastructure will help remove time, cost and its importance in AI innovation, but suggests specialised hardware.
risk from the process of bringing innovation The Office for National Statistics (ONS) is that the UK’s infrastructure is lagging behind
to market, enabling accelerated AI leading the Integrated Data Programme in other major economies around the world
development and applications. collaboration with partners across such as the US, China, Japan and Germany.
government, providing real-time evidence, We also recognise the growing compute gap
Public sector data underpinning policy decisions and delivering between large-scale enterprises and
better outcomes for citizens while maintaining researchers. Access to compute is both a
Work is underway within the government to privacy. The 2021 Declaration on Government competitiveness and a security issue. It is also
fix its own data foundations as part of Mission Reform sets out a focus on strengthening not a one-size-fits-all approach – different AI
3 of the National Data Strategy, which focuses data skills across government including senior technologies need different capabilities.
on transforming the government's use of data leaders.
32 33
environment. The report will feed into UKRI’s Tech Nation

wider work on Digital Research
Infrastructure.26 Tech Nation is a predominantly government-
funded programme, built to deliver its own
Alongside access to necessary compute initiatives that grow and support the UK’s
capacity, the competitiveness of the AI burgeoning digital tech sector. This includes
hardware will be critical to the UK's overall growth initiatives aiming to help businesses
research and commercial competitiveness in successfully navigate the transition from start-
the sector. The UK is a world leader in chip up to scale-up and beyond, network initiatives
and systems design, underpinned by to connect the UK digital ecosystem, and the
processor innovation hubs in Cambridge and Tech Nation Visa scheme, which offers a route
Bristol. We have world-leading companies into the UK for exceptionally talented
supporting both general purpose AI – individuals from overseas.
Graphcore has built the world's most complex
AI chip,27 and for specific applications – XMOS Recent growth programmes include Applied
is a leader in AI for IOT. The government is AI, their first to help the UK’s most promising
currently undertaking a wider review of its founders who are applying AI in practical
international and domestic approach to areas and creating real-world impact; Net
the semiconductor sector. Given Zero, a six-month free growth programme for
commercial and innovation priorities in AI, tech companies that are creating a more
further support for the chip design sustainable future; and Libra, which is focused
community will be considered. on supporting Black founders and addressing
racial inequality in UK tech.
Finance and VC
The government will continue to evaluate
AI innovation is thriving in the UK, backed by the state of funding specifically for
our world-leading financial services industry. innovative firms developing AI
In 2020, UK firms that were adopting or technologies across every region of the
Sustained public and private investment in a Understanding our domestic AI computing creating AI-based technologies received UK. This work will explore if there are any
range of facilities from cloud, laboratory and capacity needs and their relationship to £1.78bn in funding, compared to £525m significant investment gaps or barriers to
academic department scale, through to energy use is increasingly important25 if we raised by French companies and £386m accessing funding that AI innovative
supercomputing, will be necessary to ensure are to achieve our ambitions. To better raised in Germany.28 More broadly, companies are facing that are not being
that accessing computing power is not a understand the UK’s future AI computing investment in UK deep tech companies has addressed. Government commits to
barrier to future AI research and innovation, requirements, the Office for AI and UKRI increased by 291% over the past five years, reporting on this work in Autumn 2022.
commercialisation and deployment of AI. In will evaluate the UK’s computing capacity though deal sizes remain considerably
June 2021, the government announced joint needs to support AI innovation, smaller compared to the US.29
funding with IBM for the Hartree National commercialisation and deployment. This
Centre for Digital Innovation to stimulate high study will look at the hardware and broader
performance computing enabled innovation needs of researchers and organisations, large
in industry and make cutting-edge and small, developing AI technologies,
technologies like AI more accessible to alongside organisations adopting AI products
businesses and public sector organisations. and services. The study will also consider the
possible wider impact of future computing
requirements for AI as it relates to areas of
proportional concern, such as the
34 35
Accessing the right finance at the right time is introduced the National Security and
critical for AI innovators to be able to develop Investment Act that will provide new powers
their idea into a commercially viable product to screen investments effectively and
and grow their business, but this is efficiently now and into the future. It will give
complicated by the long timelines often businesses and investors the reassurance
needed for AI research and development that the UK continues to welcome the right
work.30,31 The AI Council’s Roadmap suggests a talent, investment and collaboration that
funding gap at series B+, meaning that AI underpins our wider economic security.
companies are struggling to scale and stay
under UK ownership. Trade
While the UK’s funding ecosystem is robust, AI is a key part of the UK’s digital goods and
the government is committed to ensuring the services exports, which totalled £69.3bn in
system is easy for businesses and innovators 2019.32 Trade can support the UK’s objectives
to navigate, and that any existing gaps are to sustain the mature, competitive and
addressed. The recent Innovation Strategy innovative AI developer base the UK needs to
signalled the Government’s efforts to support access customers around the world.
innovators by bringing together effective
private markets with well-targeted public As part of its free trade agenda, the
investment. In it, the government set out government is committed to pursuing
plans to upskill lenders to assess risk when ambitious digital trade chapters to help place
lending to innovative businesses and outlined the UK as a global leader. As the UK secures
work across Innovate UK and the British new trade deals, the government will
Business Bank to investigate how businesses include provisions on emerging digital
interact with the public support landscape, to technologies, including AI, and champion
maximise accessibility for qualifying international data flows, preventing
businesses. A good example of this is the unjustified barriers to data crossing borders
Future Fund: Breakthrough, a new £375 while maintaining the UK’s high standards for
million UK-wide programme launched in July personal data protection.
2021, will encourage private investors to co-
invest with the government in high-growth In doing so, the UK aims to deliver digital
innovative businesses to accelerate the trade chapters in agreements that: 1) provide
deployment of breakthrough technologies. legal certainty; 2) support data flows; 3)
protect consumers; 4) minimise non-tariff
Our economy’s success and our citizens’ barriers to digital trade; 5) prevent
safety rely on the government’s ability to discrimination against trade by electronic
protect national security while keeping the UK means; and 6) promote international
open for business with the rest of the world. cooperation and global AI governance. All of
Within this context, we will ensure we protect these aims support a pro-innovation agenda.
the growth of welcome investment into the
UK’s AI ecosystem. The government has
36 37
Pillar 1 - Investing in the Long Term Needs of the AI Ecosystem 6. Publish a report on the UK’s compute capacity needs to support AI innovation, commercialisation
and deployment. The report will feed into UKRI’s wider work on infrastructure.
Actions:
7. Continue to publish open and machine-readable data on which AI models for both public and
1. Launch a new National AI Research and Innovation Programme, that will align funding commercial benefit can depend.
programmes across UKRI Research Councils and Innovate UK, stimulating new investment in
fundamental AI research while making critical mass investments in particular applications of AI. 8. Consider what valuable datasets the government should purposefully incentivise or curate that
will accelerate the development of valuable AI applications.
2. Lead the global conversation on AI R&D and put AI at the heart of our science and technology
alliances and partnerships worldwide through: 9. Undertake a wider review of our international and domestic approach to the semiconductor
sector. Given commercial and innovation priorities in AI, further support for the chip design
1.
i Work with partners around the world on shared AI challenges, including participation in Horizon community will be considered.
Europe to enable collaboration with other European researchers.
10. Evaluate the state of funding specifically for innovative firms developing AI technologies in the UK,
ii2. Use of Overseas Development Assistance to support partnerships with developing AI nations. and report on this work in Autumn 2022.
iii3. Deliver new initiatives through the US UK Declaration on Cooperation in AI R&D. 11. Protect national security through the National Security & Investment Act while keeping the UK
open for business with the rest of the world, as our economy’s success and our citizens’ safety
3. Develop a diverse and talented workforce which is at the core of maintaining the UK’s world rely on the government’s ability to take swift and decisive action against potentially hostile foreign
leading position through: investment.
1.
i Scoping what is required to upskill employees to use AI in a business setting. Then, working 12. Include provisions on emerging digital technologies, including AI, in future trade deals alongside
with the Department for Education, explore how skills provision can meet these needs championing international data flows, preventing unjustified barriers to data crossing borders
through the Skills Value Chain and build out AI and data science skills through Skills and maintaining the UK’s high standards for personal data protection.
Bootcamps.
ii2. Supporting existing interventions across top talent, PhDs and Masters levels and developing
world leading teams and collaborations, the government will continue to attract and develop
the brightest and best people to build AI.
iii3. Inspiring all to be excited by the possibilities of AI, by supporting the National Centre for
Computing Education (NCCE) to ensure AI programmes for children are accessible and reach
the widest demographic and that career pathways for those working with or developing AI
are clearly articulated on career guidance platforms.
iv4. Promoting the revitalised and new visa routes that encourage innovators and entrepreneurs
to the UK, making attractive propositions for prospective and leading AI talent.
4. Publish a policy framework setting the government's role in enabling better data availability in the
wider economy. The government is already consulting on the opportunity for data intermediaries
to support responsible data sharing and data stewardship in the economy and the interplay of AI
technologies with the UK’s data rights regime.
5. Consult on the potential role and options for a future national ‘cyber-physical infrastructure’
framework, to help identify how common interoperable digital tools and platforms and cyber-
physical or living labs could come together to form a digital and physical ‘commons’ for
innovators, enabling accelerated AI development and applications.
38 39
Pillar 2:
Ensuring AI benefits all
sectors and regions
Supporting the transition to an AI-enabled economy, capturing the benefits of AI
innovation in the UK, and ensuring AI technologies benefit all sectors and regions
To ensure that all sectors and regions of the access to relevant skills discussed above, more evidence, the Office for AI will publish
Government’s aim is to diffuse AI across UK economy can benefit from the positive other challenges reported by businesses are research later this year into the drivers of
the whole economy to drive the highest transformation that AI will bring, the the lack of engagement with end users, AI adoption and diffusion.
amount of economic and productivity government will back the domestic design limiting adoption and commercialisation.
growth due to AI. and development of the next generation of AI Commercialisation outcomes are also often To stimulate the development and
systems, and support British business to constrained by business models rather than adoption of AI technologies in high-
This will be achieved by: adopt them, grow and become more technical issues and a lack of understanding potential, low-AI maturity sectors the
productive. The UK has historically been of AI-related projects’ return on investment. Office for AI and UKRI will launch a
• Supporting AI businesses on their excellent at developing new technologies but programme that will:
commercial journey, understanding the less so at commercialising them into products
unique challenges they face and helping and services. • Support the identification and creation of
them get to market and supporting AI deployment – understanding new opportunities for businesses, whether
innovation in high potential sectors and As well as smart action to support both dynamics SMEs or larger firms, to use AI and for AI
locations where the market currently suppliers, developers and adopters, developers to build new products and
doesn’t reach; government also has a role to play when it To grow the market and spread AI to more services that address these needs;
comes to the use of AI, both as a significant areas of our economy, the government aims
• Understanding better the factors that market pull in terms of public procurement, to support the demand side as well as the • Create a pathway for AI developers to
influence the decisions to adopt AI into such as the NHS and the defence sector, with means for commercialising AI - understanding start companies around new products
organisations – which includes an a dedicated Defence AI Strategy and AI what, why, when and how companies choose and services or to extend and diversify
understanding of when not to; Centre, but also in terms of using the to incorporate AI into their business planning their product offering if they are looking
technology to solve big public policy is a prerequisite to any attempt to encourage to grow and scale;
• Ensuring AI is harnessed to support challenges, such as in health and achieving wider adoption and diffusion across the UK.
outcomes across the Government’s net zero. Finally, it requires being bold and • Facilitate close engagement between
Innovation Strategy, including by experimental, and supporting the use of AI in EY research delivered on behalf of DCMS businesses and AI developers to ensure
purposefully leveraging our leading AI the service of mission-led policymaking. shows that AI remains an emerging products and services developed address
capabilities to tackle real-world problems technology for private sector and third sector business needs, are responsibly
facing the UK and world through our organisations in the UK. 27% of UK developed and implemented, and
Innovation Missions,33 while driving organisations have implemented AI designed and deployed so that
forward discovery; Commercialisation technologies in business processes; 38% of businesses and developers alike are
organisations are planning and piloting AI prepped and primed for AI
• Leveraging the whole public sector’s Developing a commercial AI product or technology; and 33% of organisations have implementation; and
capacity to create demand for AI and service is more than just bringing an idea to not adopted AI and are not planning to.
markets for new services. market or accessing the right funding. Recent Consistent with studies of AI adoption,34 the • Incentivise investors to learn about these
analysis from Innovate UK suggests that size of an organisation was found to be a new market opportunities, products, and
obtaining private funding is only one among large contributing factor to the decision to services, so that, where equity finance is
many other obstacles to successful adopt AI, with large organisations far more needed, the right financing is made
commercial outcomes in AI-related projects. likely to have already done so. Recognising available to AI developers.
As well as the well known barriers such as that for many sectors this is the cutting edge
access to data, labour market supply and of industrial transformation, and the need for
40 41
Pillar 2: Ensuring AI benefits all sectors and regions National AI Strategy
AI and COVID-19
When the pandemic began it created a unique environment where AI technologies were developed
to identify the virus more quickly, to help with starting treatments earlier and to reduce the likelihood
that people will need intensive care.
Working with Faculty, NHS England and NHS Improvement developed the COVID-19 Early Warning
System (EWS). A first-of-its-kind toolkit that forecasts vital metrics such as COVID-19 hospital
AI and Intellectual Property (IP): Call for admissions and required bed capacity up to three weeks in advance, based on a wide range of data
Views and Government Response from the NHS COVID-19 Data Store. This gave national, regional and local NHS teams the confidence
to plan services for patients amid any potential upticks in COVID-related hospital activity.
An effective Intellectual Property (IP) system is
fundamental to the Government’s ambition Creating and protecting Intellectual At the same time over the past year, the NHS AI Lab has collected more than 40,000 X-ray, CT and
for the UK to be a ‘science superpower’ and Property MRI chest images of over 13,000 patients from 21 NHS trusts through the National COVID-19 Chest
the best place in the world for scientists, Imaging Database (NCCID), one of the largest centralised collections of medical images in the UK. The
researchers and entrepreneurs to innovate. Intellectual Property (IP) plays a significant NCCID is being used to study and understand the COVID-19 illness and to improve the care for
To ensure that IP incentivises innovation, our part in building a successful business by patients hospitalised with severe infection. The database has enabled 13 projects to research new AI
aspiration is that the UK’s domestic IP rewarding people for inventiveness and technologies to help speed up the identification, severity assessment and monitoring of COVID-19.
framework gives the UK a competitive edge. creativity and enabling innovation. IP supports
business growth by incentivising investment, UK AI companies have also shown how AI can help accelerate the search for potential drug
In support of this ambition, the IPO published safe-guarding assets and enabling the sharing candidates, streamline triage and contribute to global research efforts. BenevolentAI, a world-leading
its AI and IP call for views to put the UK at the of know-how. The Intellectual Property Office AI company focused on drug discovery and medicine development, used their biomedical knowledge
forefront of emerging technological (IPO) recognises that AI researchers and graph to identify a potential coronavirus treatment from already approved drugs that could be
opportunities, by considering how AI impacts developers need the right support to repurposed to defeat the virus. This was later validated through experimental testing from
on the existing UK intellectual property commercialise their IP, and helps them to AstraZeneca. UK AI company DeepMind have adapted their AI-enabled protein folding breakthrough
framework and what impacts it might have for understand and identify their intellectual to better understand the virus’ structure, contributing to a wider understanding of how the virus can
AI in the near to medium term. assets, providing them with the skills to function.
protect, exploit and enforce their rights to
In March this year, the government published improve their chances of survival and growth.
its response to the call for views, which
committed to the following next steps:
These challenges will require public sector government's strategic goals and where AI
• To consult on the extent to which Using AI for the public benefit leadership, identifying strategic priorities that can provide a catalytic contribution,35
copyright and patents should protect AI can maximise the potential of AI for the including through Innovation Missions and
generated inventions and creative works; AI can contribute to solving the greatest betterment of the UK. the Integrated Review’s ‘Own-Collaborate-
challenges we face. AI has contributed to Access’ framework.36
• To consult on measures to make it easier tackling COVID-19, demonstrating how these The government has a clear role to play. In
to use copyright protected material in AI technologies can be brought to bear stimulating and applying AI innovation to The COVID-19 pandemic has shown that
development; alongside other approaches to create priority applications and wider strategic goals, global challenges need global solutions. The
effective, efficient and context-specific the government can help incentivise a group UK’s international science and technology
• An economic study to enhance solutions. of different actors to harness innovation for partnerships, global network of science and
understanding of the role the IP improving lives, simultaneously reinforcing innovation officers, and research and
framework plays in incentivising There are many areas of AI development that the innovation cycle that can drive wider innovation hubs, are working alongside UK
investment in AI. have matured to the point that industry and economic benefits – from creating and universities, research institutes and investors
third sector organisations are investing invigorating markets, to the role of open to foster new collaborations to tackle the
The consultation, on copyright areas of significantly in AI tools, techniques and source in the public, private and third sectors, global challenges we all share, including in
computer generated works and text and data processes. These investments are helping to to raising productivity. Over the next six to innovations on global health and to achieve
mining, and on patents for AI devised move AI from the lab and into commercial twelve months, the Office for AI will work net zero emissions around the globe.
inventions, will be launched before the end of products and services. But there remain more closely with the Office for Science and
the year so that the UK can harness the complex, cross-sector challenges that Technology Strategy and government
opportunities of AI to further support industry is unlikely to solve on its own. departments to understand the
innovation and creativity.
42 43
AI and net zero
AI works best when presented with specific

problem areas with clear system boundaries
Missions At the same time, well-specified challenges Net Zero and where there are large datasets being
have also led to some of the most impactful produced. In these scenarios, AI has the
The Innovation Strategy set out the moments of progress in AI. Whether through The Prime Minister’s Ten Point Plan for a capability to identify complex patterns, unlock
government's plans to stimulate innovation to Imagenet, CIFAR-10, MNIST, GLUE, SquAD, Green Industrial Revolution highlights the new insights, and advise on how best to
tackle major challenges facing the UK and the Kaggle, or more, challenge-related datasets development of disruptive technologies such optimise system inputs in order to best
world, and drive capability in key technologies. and benchmarks have generated as AI for energy as a key priority, and in achieve defined objectives.
This will be achieved through Innovation breakthroughs in vision, language, concert with the government’s Ten Tech
Missions,37 which will draw on multiple recommender systems, and other subfields.39 Priorities to use digital innovations to reach There are a range of climate change
technologies and research disciplines towards The government believes that challenges net zero, the UK has the opportunity to lead mitigation and adaptation challenges that fit
clear and measurable outcomes. They will be could be created that simultaneously the world in climate technologies, supporting this description. These include:
supported by Innovation Technologies,38 incentivise significant progress in Innovation us to deliver our ambitious net zero targets.
including AI, supporting their capability to Missions while rapidly progressing the This will be key to meet our stated ambition in • using machine vision to monitor the
tackle pressing global and national challenges development in the technology along the Sixth Carbon Budget, and with it a need to environment;
while supporting their adoption in novel desirable lines. consider how to achieve the maximum
areas, boosting growth and helping to possible level of emissions reductions. • using machine learning to forecast
consolidate our position as a science and AI To this end, the government will develop a electricity generation and demand and
superpower. repository of short, medium and long term Over the last ten years there have been a control its distribution around the
AI challenges to motivate industry and series of advances in AI. These advances offer network;
Some of these challenges have been society to identify and implement real- opportunities to rapidly increase the
articulated and revolve around the future world solutions to the strategic priorities. efficiency of energy systems and help reduce • using data analysis to find inefficiencies in
health, wellbeing, prosperity and security of These priorities will be identified through the emissions across a wide array of climate emission-heavy industries; and
people, the economy, and our environment – Missions Programme, and guided by the change challenges. The AI Council’s AI
in the UK and globally. These challenges are National AI R&I Programme. Roadmap advocates for AI technologies to • using AI to model complex systems, like
worthwhile and therefore difficult, and will play a role in innovating towards solutions to Earth’s own climate, so we can better
require harnessing the combined intellect Climate change and global health threats are climate change, and literature is emerging prepare for future changes.
and diversity of the AI ecosystem and the examples of shared international challenges, that shows how ‘exponential technologies’
whole nation, and will consider a full range of and science progresses through open such as AI can increase the pace of AI applications for energy and climate
possible impacts of a given solution. The pace international collaboration. This is particularly decarbonisation across the most impactful challenges are already being developed, but
of AI development is often fast, parallel and the case when AI development is able to take sectors. AI is increasingly seen as a critical they are predominantly outliers and there are
non-linear, and finding the right answer to advantage of publicly available coding technology to scale and enable these many applications across sectors that are not
these challenges will require a collection of platforms to produce new algorithms. The UK significant emissions cuts by 2030.40,41,42 yet attempted. A study by Microsoft and PwC
actors beyond just government departments, will extend its science partnerships and its estimated that AI can help deliver a global
agencies and bodies to consider the technical work investing UK aid to support local reduction in emissions of up to 4% by 2030
and social implications of certain solutions innovation ecosystems in developing compared to business as usual, with a
and increase the creativity of problem solving. countries. Through our leadership in concurrent uplift of 4.4% to global GDP. Such
In doing so, the UK will be able to find new international development and diplomacy, estimates are likely to become more accurate
paths for AI to deliver on our security and we will work to ensure international over time as the potential of AI becomes
prosperity objectives at home and abroad. collaboration can unlock the enormous more apparent.
potential of AI to accelerate progress on
global challenges, from climate change to
poverty.
44 45
Missions will also be continued through the provide a facilitative environment to enable processes around public sector culture, Ministry of Defence using AI to reduce
Innovation Strategy’s Missions Programme, the health and social care system to expertise and incentive structures. This costs and meet climate goals
which will form the heart of the government’s confidently adopt safe, effective and ethical complements previous work across
approach to respond to these priorities, and AI-driven technologies at pace and scale. government to inform and empower buyers The MOD is trialling a US startups’ Software
we will develop these missions in a way in the public sector, helping them to evaluate Defined Electricity (SDE) system, which uses AI
that considers the promise of AI The NHS AI Lab is creating a National suppliers, then confidently and responsibly to optimise electricity in real time, to help
technologies, particularly in areas of Strategy for AI in Health and Social Care in procure AI technologies for the benefit of meet its climate goals and reduce costs. Initial
specific advantage such as energy. line with the National AI Strategy. The citizens.45 tests suggest it could reduce energy draw by
strategy, which will begin engagement on at least 25% which, given the annual electricity
Government will ensure that, in key areas of a draft this year and is expected to launch The government has outlined how it plans to bill for MOD’s non-PFI sites in FY 2018/19 was
international collaboration such as the US UK in early 2022, will consolidate the system rapidly modernise our Armed Forces,46,47 and £203.6M, would equate to savings of £50.9M
Declaration on Cooperation in AI Research transformation achieved by the Lab to how investments will be guided.48,49 The every year and significant reductions in CO2
and Development and the Global Partnership date and will set the direction for AI in Ministry of Defence will soon be publishing its emissions.
on AI, we will pursue technological health and social care up to 2030. AI strategy which will contribute to how we will
developments in world-leading areas of achieve and sustain technological advantage,
expertise in the energy sector to maximise and be a great science power in defence. This
our strategic advantage. will include the establishment of the new AI Dynamic Purchasing System
The public sector as a buyer Defence AI Centre which will champion AI
Health development and use, and enable rapid The Crown Commercial Service worked
To build a world-leading strategic advantage development of AI projects. Defence should closely with colleagues in the Office for AI and
In August 2019, the Health Secretary in AI and build an ecosystem that harnesses be a natural partner for the UK AI sector and across government during drafting of
announced a £250 million investment43 to innovation for the public good, the UK will the defence strategy will outline how to guidelines for AI procurement. This was used
create the NHS AI Lab in NHSX to accelerate need to take a number of approaches. As the galvanise a stronger relationship between to design their AI Dynamic Purchasing System
the safe, ethical and effective development government, we can also work with industry industry and defence. (DPS) agreement to align with these
and use of AI-driven technologies to help leaders to develop a shared understanding guidelines, and included a baselines ethics
tackle some of the toughest challenges in and vision for the emerging AI ecosystem, assessment so that suppliers commit only to
health and social care, including earlier cancer creating longer-term certainty that enables bidding where they are capable and willing to
detection, addressing priorities in the NHS new supply chains and markets to form. deliver both the ethical and technical
Long Term Plan, and relieving pressure on the dimensions of a tender.
workforce. This requires leveraging public procurement
and pre-commercial procurement to be more The Crown Commercial Service is piloting a
AI-driven technologies have the potential to in line with the development of deep and training workshop to help improve the public
improve health outcomes for patients and transformative technologies such as AI. The sector’s capability to buy AI products and
service users, and to free up staff time for recent AI Council ecosystem survey revealed services, and will continue to work closely with
care.44 The NHS AI Lab along with partners, that 72% agreed the government should take the Office for AI and others across
such as the Accelerated Access Collaborative, steps to increase buyer confidence and AI government to ensure we are addressing the
the National Institute of Health and Care capability. The Innovation Strategy and key drivers set out in the National AI Strategy.
Excellence and the Medicines and Healthcare forthcoming National Procurement Policy
products Regulatory Agency, are working to Statement have recently articulated how we
can further refine public procurement
46 47
Pillar 2 - Ensuring AI Benefits all Sectors and Regions
Actions:
1. Launch a programme as part of UKRI’s National AI R&I Programme, designed to stimulate the
development and adoption of AI technologies in high-potential, lower-AI maturity sectors. The
programme will be primed to exploit commercialisation interventions, enabling early innovators
to access potential market opportunities where their products and services are relevant.
2. Launch a draft National Strategy for AI in Health and Social Care in line with the National AI
Strategy. This will set the direction for AI in health and social care up to 2030, and is expected to
launch in early 2022.
3. Ensure that AI policy supports the government’s ambition to secure strategic advantage through
science and technology.
4. Consider how the development of Innovation Missions also incorporates the potential of AI
solutions to tackling big, real-world problems such as net zero. This will also be complemented by
pursuing ambitious bilateral and multilateral agreements that advance our strategic advantages
in net zero sectors such as energy, and through the extension of UK aid to to support local
innovation ecosystems in developing AI nations.
5. Build an open repository of AI challenges with real-world applications, to empower wider civil
society to identify and implement real-world solutions to the strategic priorities identified
through the Missions Programme and guided by the National AI Research and Innovation
Programme.
6. Publish research into the determinants impacting the diffusion of AI across the economy.
7. Publish the Ministry of Defence AI Strategy, which will explain how we can achieve and sustain
technological advantage and be a science superpower in defence, including detail on the
establishment of a new Defence AI Centre.
48 49
Pillar 3:
Governing AI effectively
Ensuring that national governance of AI technologies encourages innovation,
investment, protects the public and safeguards our fundamental values, while
working with global partners to promote the responsible development of AI
internationally
Government’s aim is to build the most An effective governance regime that supports Supporting innovation and adoption undermine privacy and human agency; and
trusted and pro-innovation system for AI scientists, researchers and entrepreneurs to while protecting the public and physical, economic and financial harms being
governance in the world. innovate while ensuring consumer and citizen building trust enabled or exacerbated by AI technologies.
confidence in AI technologies is fundamental For example, cyber security should be
This will be achieved by: to the government’s vision over the next The UK has a strong international reputation considered early in the development and
decade. for the rule of law and technological deployment of AI systems to prevent such
• Establishing an AI governance framework breakthroughs. To build on this the harms from arising, by adopting a ‘secure by
that addresses the unique challenges and In a world where systematic international government set out its pro-innovation design’ approach to mitigate against cyber
opportunities of AI, while being flexible, competition will have significant impacts on approach through its Plan for Digital security becoming an afterthought.
proportionate and without creating security and prosperity around the world, the Regulation. The Plan recognises that well-
unnecessary burdens; government wants the UK to be the most designed regulation can have a powerful This is not to say that AI is currently
trustworthy jurisdiction for the development effect on driving growth and shaping a unregulated. The UK already regulates many
• Enabling AI products and services to be and use of AI, one that protects the public thriving digital economy and society, whereas aspects of the development and use of AI
trustworthy, by supporting the development and the consumer while increasing poorly-designed or restrictive regulation can through ‘cross-sector’ legislation and different
of an ecosystem of AI assurance tools and confidence and investment in AI technologies dampen innovation. The Plan also regulators. For example, there is coverage in
services to provide meaningful information in the UK. acknowledges that digital businesses, which areas like data protection (Information
about AI systems to users and regulators; include those developing and using AI Commissioner’s Office), competition
Effective, pro-innovation governance of AI technologies, are currently operating in some (Competition & Markets Authority), human
• Growing the UK’s contribution to the means that (i) the UK has a clear, instances without appropriate guardrails. The rights and equality (Equality & Human Rights
development of global AI technical proportionate and effective framework for existing rules and norms, which have so far Commission). As well as through ‘sector-
standards, to translate UK R&D for regulating AI that supports innovation while guided business activity, were in many cases specific’ legislation and regulators, for
trustworthy AI into robust, technical addressing actual risks and harms, (ii) UK not designed for these modern technologies example financial services (Financial Conduct
specifications and processes that can regulators have the flexibility and capabilities and business models. In addition, these Authority) and medical products (Medicines
support our AI governance model, ensure to respond effectively to the challenges of AI, technologies are themselves disrupting these and Healthcare products Regulatory Agency).
global interoperability and minimise the and (iii) organisations can confidently innovate established rules and norms.
costs of regulatory compliance; and adopt AI technologies with the right tools As the use of AI increases, the UK has
and infrastructure to address AI risks and This is especially the case for AI which, with its responded by reviewing and adapting the
• Building UK regulators’ capacities to use harms. The UK public sector will lead the way powerful data processing and analytical regulatory environment. For example, the
and assess AI, ensuring that they can deliver by setting an example for the safe and ethical capabilities, is disrupting traditional business Data: A new direction consultation, published
on their responsibilities as new AI-based deployment of AI through how it governs its models and processes.50 There is growing earlier this month, invites views on the role of
products and services come to market; own use of the technology. awareness in industry and by citizens of the the data protection framework within the
potential risks and harms associated with AI broader context of AI governance. Specifically,
• Setting an example in the safe and ethical We will collaborate with key actors and technologies. These include concerns around the consultation examines the role of
deployment of AI, with the government partners on the global stage to promote the fairness, bias and accountability of AI systems. sensitive personal data in bias detection and
leading from the front; responsible development and deployment of For example, the report from the Commission mitigation in AI systems, and the use of the
AI. The UK will act to protect against efforts to on Race and Ethnic Disparities raised term ‘fairness’ in a data protection context.
• Working with our partners around the adopt and apply these technologies in the concerns around the potential for novel ways
world to promote international agreements service of authoritarianism and repression. for bias to be introduced through AI. Other
and standards that deliver for our Through our science partnerships and wider concerns include the ability of AI to
prosperity and security, and promote development and diplomacy work, we will
innovation that harnesses the benefits of AI seek to engage early with countries on AI
as we embed our values such as fairness, governance, to promote open society values
openness, liberty, security, democracy, rule and defend human rights.
of law and respect for human rights.
50 51
Pillar 3: Governing AI effectively National AI Strategy
Data: A new direction consultation
The UK data protection framework (UK General Data Protection Regulations and Data Protection Act
2018) is technology neutral and was not intended to comprehensively govern AI systems, or any
other specific technologies. Many AI systems do not use personal data at all.
Navigating and applying relevant data protection provisions can be perceived as a complex or
confusing exercise for an organisation looking to develop or deploy AI systems, possibly impeding
uptake of AI technologies.
DCMS is currently running a consultation on potential reforms to the data protection framework,
closing on the 19th November 2021. The consultation calls for views on specific data protection
2018, the government agreed with the House connectivity, the move to mobile working, provisions that are currently triggered in the process of developing and deploying AI. In particular,
of Lords’ view that “blanket AI-specific the dominant role of major platforms etc. the consultation covers:
regulation, at this stage, would be It is often hard to unpick the specific
inappropriate... [and] that existing sector-specific impact of AI; focusing regulation on the • Clarifying the use and reuse of personal data for research (including AI development) (Ch 1);
regulators are best placed to consider the impact particular use cases where there is risk
on their sector of any subsequent regulation allows risks to be addressed holistically, • Clarifying the use and reuse of personal data under the legitimate interests test, including bias
which may be needed.” and simplifies things for innovators. detection and mitigation anonymisation (Ch 1);
There are some strong reasons why our Having embraced a strong sector-based • Explicitly authorising the use of sensitive personal data (special category data) for bias detection
sector-led approach makes sense: approach to date, now is the time to decide and mitigation in AI systems (Ch 1);
whether our existing approach remains the
1. The boundaries of AI risks and harms right one. • Clarifying the use of the term ‘fairness’ in a data protection context (Ch 1);
are grey, because the harms raised by
these technologies are often non-AI, or As the UK’s regulators have begun to respond • Assessing the challenges with the current data protection framework in developing and
extensions of non-AI, issues, and also to the emergence of AI, challenges have deploying AI responsibly (Ch 1);
because AI is rapidly developing and emerged. These include:
therefore what counts as the AI part of a • Assessing the general suitability and operation of UK GDPR Article 22 (rights relating to
system is constantly changing. • Inconsistent or contradictory automated decision-making and profiling) (Ch 1);
approaches across sectors. While a
2. Use cases for AI, and their wider sector-led approach allows • Mandatory transparency requirements for the use of algorithmic decision-making in the public
impacts, can be highly complex in their responsiveness to sector specific sector (Ch 5).
own right. There is a big limitation in challenges, it could create barriers to
what can be covered in cross-cutting adoption across sectors by creating
legislation on AI, and regardless of the confusing or contradictory compliance
overall regulatory approach, the detail will requirements;
always need to be dealt with at the level These challenges raise the question of Working with the AI ecosystem the Office for
of individual harms and use cases. • Overlap between regulatory whether the UK’s current approach is AI will develop our national position on
mandates, creating uncertainty about adequate, and whether there is a case for governing and regulating AI, which will be
3. Individual regulators and industries responsibility, the potential for issues to greater cross-cutting AI regulation or greater set out in a White Paper in early 2022. The
are already starting to respond to the fall between the gaps, and increased consistency across regulated sectors. White Paper will set out the government’s
risks of AI, and to work with innovators in need for coordination; position on the potential risks and harms
their sectors to guide on interpretation of At the same time, alternative methods and posed by AI technologies and our proposal to
existing regulations, and on what further • AI regulation could become framed approaches to governing AI have emerged address them.
regulatory responses are appropriate. narrowly around prominent, existing from multilateral and multi stakeholder fora,
Enabling and empowering individual cross-cutting frameworks, e.g. the data at international and regional levels, including
bodies to respond is a much quicker protection framework, while the range of global standards development organisations,
response to individual harms than AI risks and harms is much broader; academia, thought leaders, and businesses.
agreeing to an AI regulatory regime that This has raised awareness about the
makes sense across all sectors. • The growing activity in multilateral and importance of AI governance, but also
multi stakeholder fora internationally, potentially confusion for the consumer about
4. AI is not the only ongoing technology and global standards development what good AI governance looks like and
change, and its impacts are often organisations that addresses AI across where responsibility lies.
interlinked with other innovations and sectors could overtake a national effort to
behaviour changes, including increased build a consistent approach.
52 53
Alternative options potentially other, options for governing AI Greater cooperation is also being enabled helped found the Global Partnership on AI
technologies. Having exited the EU, we have through initiatives such as through the Digital (GPAI), providing significant support for
The UK’s 2018 policy position that “existing the opportunity to build on our world-leading Regulation Cooperation Forum, a recently evidence underpinning these initiatives, such
sector-specific regulators are best placed to regulatory regime by setting out a pro- formed voluntary forum comprising the as the recently announced £1m investment in
consider the impact on their sector of any innovation approach, one that drives Competition & Markets Authority (CMA), GPAI’s data trust research by BEIS.
subsequent regulation which may be needed” prosperity and builds trust in the use of AI. Financial Conduct Authority (FCA), Information
will be tested in our work towards the We will consider what outcomes we want to Commissioner's Office (ICO) and Office of The UK will act to protect against efforts to
development of a White Paper, along with achieve and how best to realise them, across Communications (Ofcom) to deliver a joined adopt and apply these technologies in the
potential alternatives. The main alternative existing regulators’ remits and consider the up approach to digital regulation. service of authoritarianism and repression
options are: role that standards, assurance, and and through our science partnerships and
international engagement plays. wider development and diplomacy work seek
1. Removing some existing regulatory to engage early with countries on AI
burdens where there is evidence they are International governance and governance, including when existing
creating unnecessary barriers to collaboration technology governance is less developed, to
innovation. Regulators’ coordination and promote open society values and defend
capacity The UK will work with partners to support the human rights.
2. Retaining the existing sector-based international development of AI governance
approach, ensuring that individual While some regulators are leading the way in in line with our values. We will do this by UK Defence has a strong record of
regulators are empowered to work understanding the implications of AI for their working with partners around the world to collaboration with international partners and
flexibly within their own remits to ensure sector or activity, we need all regulators to be shape approaches to AI governance under allies. Key collaborations include engagement
AI delivers the right outcomes. able to do this. The cross-sector and development, such as the proposed EU AI Act with NATO allies to lead AI integration and
disruptive nature of AI also raises new and potential Council of Europe legal interoperability across the Alliance, and
3. Introducing additional cross-sector challenges in terms of regulatory overlap. For framework. We will work to reflect the UK’s supporting the AI Partnership for Defence, a
principles or rules, specific to AI, to example, concerns around fairness relate to views on international AI governance and 14-nation coalition providing values based
supplement the role of individual algorithmic bias and discrimination issues prevent divergence and friction between global leadership for defence AI.
regulators to enable more consistency under the Equality Act, the use of personal partners, and guard against abuse of this
across existing regimes. data (including sensitive personal data) and critical technology. The government will continue to work
sector-specific notions of fairness such as the with our partners around the world to
For any of these options, it will be necessary Financial Conduct Authority’s Fair Treatment The UK is already working with like-minded shape international norms and standards
to ensure that regulators and other relevant of Customers guidance. partners to ensure that shared values on relating to AI, including those developed
bodies are equipped to tackle the challenges human rights, democratic principles and the by multilateral and multistakeholder
raised by AI. This may require additional The government is working with The Alan rule of law shape AI regulation and bodies at global and regional level. This will
capabilities, capacity, and better coordination Turing Institute and regulators to examine governance frameworks, whether binding or support our vision for a global ecosystem that
among existing regulators; new guidance; or regulators’ existing AI capacities. In particular, non-binding, and that an inclusive multi- promotes innovation and responsible
standards to better enable consistency across this work is exploring monitoring and stakeholder approach is taken throughout development and use of technology,
existing regulatory regimes. assessing products and services using AI and these processes. As the international debate underpinned by our shared values of
dealing with complexities arising from cross- on these frameworks has gained momentum, freedom, fairness, and democracy.
In developing our White Paper position, the sectoral AI systems.51 the UK has proactively engaged on AI at the
Office for AI will consider all of these, and OECD,52 Council of Europe and UNESCO, and
54 55
What are technical standards and how do they benefit the UK?
Global technical standards set out good practice that can be consistently applied to ensure that
products, processes and services perform as intended – safely and efficiently. They are generally
voluntary and developed through an industry-led process in global standards developing
organisations, based on the principles of consensus, openness, and transparency, and benefiting
from global technical expertise and best practice.53
We want global technical standards for AI to benefit UK citizens, businesses, and the economy by:
• Supporting R&D and Innovation. Technical standards should provide clear definitions and
processes for innovators and businesses, lowering costs and project complexity and improving
product consistency and interoperability, supporting market uptake. The government is also exploring with The UK is leading the way on AI technical
stakeholders to: standards internationally
• Supporting trade. Technical standards should facilitate digital trade by minimising regulatory
requirements and technical barriers to trade. • Pilot an AI Standards Hub to expand the The UK’s global approach to AI
UK’s international engagement and standardisation is exemplified by our
• Giving UK businesses more opportunities. Standardisation is a co-creation process that spans thought leadership; and leadership in the International Organisation
different roles and sectors, providing businesses with access to market knowledge, new for Standardisation and International
customers, and commercial and research partnerships. • Develop an AI standards engagement Electrotechnical Commission (ISO/IEC) on four
toolkit to guide multidisciplinary UK active AI projects, as well as the UK’s initiation
• Delivering on safety, security and trust. The Integrated Review set out the role of technical stakeholders to engage in the global AI of and strong engagement in the Industry
standards in embedding transparency and accountability in the design and deployment of standardisation landscape. Specification Group on Securing AI at the
technologies. AI technical standards (e.g. for accuracy, explainability and reliability) should ensure European Telecommunications Standards
that safety, trust and security are at the heart of AI products and services. Internationally, the government is: Institute (ETSI).
• Supporting conformity assessments and regulatory compliance. Technical standards should • Increasing bilateral engagement with At ISO/IEC, the UK, through BSI, is leading the
support testing and certification to ensure the quality, performance, reliability of products before partners, including strengthening development of AI international standards in
they enter the market. This includes providing a means of compliance with requirements set out coordination and information sharing. concepts and terminology; data; bias;
in legislation. governance implications; and data life cycles.
• Bringing together conversations at At ETSI we have published, among other
standards developing organisations and documents, ETSI GR SAI 002 on Data Supply
multilateral fora. BSI and the government Chain Security, which was led by the UK’s
AI and global digital technical The UK is taking a global approach to shaping are members of the Open Community for National Cyber Security Centre.
standards technical standards for AI trustworthiness, Ethics in Autonomous and Intelligent
seeking to embed accuracy, reliability, Systems (OCEANIS), which unites global The ISO/IEC work programme includes the
The UK’s Plan for Digital Regulation sets out security, and other facets of trust in AI SDOs, businesses, and research development of an AI Management System
our ambition to use digital technical technologies from the outset. The institutes. Standard (MSS), which intends to help solve
standards to provide an agile and pro- government’s work to date on AI technical some of the implementation challenges of AI.
innovation way to regulate AI technologies standards with international partners, • Engaging in the OECD’s Network of This standard will be known as ISO/IEC 42001
and build consistency in technical industry, and other stakeholders provides a Experts Group on Implementing and will help an organisation develop or use
approaches, as part of a wider suite of potential foundation to complement our Trustworthy AI, collaborating with artificial intelligence responsibly in pursuing
governance tools complementing ‘traditional’ governance and regulatory approach. governments, academics, and experts to its objectives, and deliver its expected
regulation. build guidance. obligations related to interested parties.
Domestically, the government has established
The integration of standards in our model for a strategic coordination initiative with the • Promoting the 2021 Carbis Bay G7
AI governance and regulation is crucial for British Standards Institution (BSI) and the Leaders’ Communiqué, on supporting
unlocking the benefits of AI for the economy National Physical Laboratory to explore ways inclusive, multi-stakeholder approaches
and society, and will play a key role in to step up the UK’s engagement in global to standards development, by ensuring
ensuring that the principles of trustworthy AI standards developing organisations.54 our UK approach to AI standards is
are translated into robust technical multidisciplinary, and encourages a wide
specifications and processes that are globally- set of stakeholders in standards
recognised and interoperable. developing organisations.
56 57
AI Assurance An AI assurance ecosystem is emerging within Public sector as an exemplar decisions affecting individuals, highlighting the
both the public and private sectors, with a importance of stewarding AI systems in a
Understanding whether AI systems are safe, range of companies including established The government must lead from the front and responsible manner to increase overall trust
fair or are otherwise trustworthy requires accountancy firms and specialised start-ups, set an example in the safe and ethical in their use.
measuring, evaluating and communicating a beginning to offer assurance services. A deployment of AI. The Office for AI and the
variety of information, including how these number of possible assurance techniques55 Government Digital Service worked with The To ensure that citizens have confidence and
systems perform, how they are governed and have been proposed and regulators are Alan Turing Institute to produce guidance on trust in how data is being processed and
managed, whether they are compliant with beginning to set out how AI might be assured AI ethics and safety in the public sector in analysed to derive insights, the Central
standards and regulations, and whether they (for example, the ICO’s Auditing Framework 2019. This guidance identifies the potential Digital and Data Office (CDDO) is
will reliably operate as intended. AI assurance for AI). harms caused by AI systems and proposes conducting research with a view to
will play an important enabling role, unlocking measures to counteract them. The developing a cross-government standard
economic and social benefits of AI systems. However, the assurance ecosystem is government is working with The Alan for algorithmic transparency in line with the
currently fragmented and there have been Turing Institute to update this guidance in commitment in the National Data Strategy.
several calls for better coordination, including order to provide public servants with the
from the Committee on Standards in Public most current information about the state The CDDO work is being conducted
What is Assurance? Life and the Office for Statistics Regulation. of the art in responsible AI innovation. This collaboratively with leading organisations in AI
The CDEI’s recently published review into bias update incorporates the delivery of interactive and data ethics and it has been informed by a
Assurance covers a number of governance in algorithmic decision-making also points to workbooks aimed to equip public sector range of public engagement processes. To
mechanisms for third parties to develop trust the need for an ecosystem of industry stakeholders with the practical tools and skills date, no other country has developed a
in the compliance and risk of a system or standards and professional services to help needed to bring the content of the original standard for algorithmic transparency at a
organisation. Assurance as a service draws organisations address algorithmic bias in the guidance to life.56 national level. Proactive transparency in this
originally from the accounting profession, but UK and beyond. field will be an extension of the UK’s long
has since been adapted to cover many areas The Ministry of Defence is moving quickly standing open data and data ethics
such as cyber security, product safety, quality Playing this crucial role in the development against a fast-evolving threat picture to secure leadership.
and risk management. and deployment of AI, assurance is likely to the benefits of these transformative
become a significant economic activity in its technologies. The Ministry of Defence has
In these areas, mature ecosystems of own right and is an area in which the UK, with rigorous codes of conduct and regulation
assurance products and services enable particular strengths in legal and professional which uphold responsible AI use, and is
people to understand whether systems are services, has the potential to excel. working closely with the wider
trustworthy and direct their trust or distrust government on approaches to ensure
appropriately. These products and services To support the development of a mature clear alignment with the values and norms
include: process and technical standards; AI assurance ecosystem, the CDEI is of the society we represent.
repeatable audits; impact assessments; publishing an AI assurance roadmap. This
certification schemes; advisory and training roadmap clarifies the set of activities needed As the CDEI conducts its ongoing work to
services. to build a mature assurance ecosystem and address bias in algorithmic decision-making,
identifies the roles and responsibilities of the Commission on Race and Ethnic
different stakeholders across these activities. Disparities recommended that a mandatory
transparency obligation be placed on all
public sector organisations applying
algorithms that have an impact on significant
58 59
Pillar 3 - Governing AI Effectively
Actions:
1. Develop a pro-innovation national position on governing and regulating AI, which will be set out
in a White Paper, to be published in early 2022.
2. Publish the CDEI assurance roadmap and use this to continue work to develop a mature AI
AI risk, safety, and long-term well as using powers through the National assurance ecosystem in the UK.
development Security & Investment Act to mitigate risks
arising from a small number of potentially 3. Pilot an AI Standards Hub to coordinate UK engagement in AI standardisation globally, and
The government takes the long term risk of concerning actors. At a strategic level, the explore with stakeholders the development of an AI standards engagement toolkit to support the
non-aligned Artificial General Intelligence, and National Resilience Strategy will review our AI ecosystem to engage in the global AI standardisation landscape.
the unforeseeable changes that it would approach to emerging technologies; the
mean for the UK and the world, seriously. Ministry of Defence will set out the details of 4. Continue our engagement to help shape international frameworks, and international norms and
the approaches by which Defence AI is standards for governing AI, to reflect human rights, democratic principles, and the rule of law on
There are also risks, safety and national developed and used; the National AI R&I the international stage.
security concerns that must be considered Programme’s emphasis on AI theory will
here and now - from deepfakes and targeted support safety; and central government will 5. Support the continuing development of new capabilities around trustworthiness, acceptability,
misinformation from authoritarian regimes, to work with the national security apparatus to adoptability, and transparency of AI technologies via the national AI Research and Innovation
sophisticated attacks on consumers or critical consider narrow and more general AI as a top- Programme.
infrastructure. As AI becomes increasingly level security issue.
ubiquitous, it has the potential to bring risks 6. Publish details of the approaches which the Ministry of Defence will use when adopting and using
into everyday life, into businesses and into AI.
national security and defence. So as AI
becomes more general and is simply used in 7. Develop a cross-government standard for algorithmic transparency.
more domains, we must maintain a broad
perspective on implications and threats, with 8. Work with The Alan Turing Institute to update the guidance on AI ethics and safety in the public
the tools to understand its most subtle sector.
impacts, and ensure the UK is protected from
bad actors using AI, as well as risks inherent in 9. Coordinate cross-government processes to accurately assess long term AI safety and risks, which
unsafe future versions of the technology itself. will include activities such as evaluating technical expertise in government and the value of
research infrastructure.
The Office for AI will coordinate cross-
government processes to accurately 10. Work with national security, defence, and leading researchers to understand how to anticipate
assess long term AI safety and risks, which and prevent catastrophic risks.
will include activities such as evaluating
technical expertise in government and the
value of research infrastructure. Given the
speed at which AI developments are
impacting our world, it is also critical that the
government takes a more precise and timely
approach to monitoring progress on AI, and
the government will work to do so.
The government will support the safe and

ethical development of these technologies as
60 61
Next steps
The National AI Strategy proposes three core The government’s AI Council, an independent
pillars which, taken together, are areas the UK expert group formed to represent high-level
can make the biggest impact to set the leadership of the UK’s AI ecosystem, has
country on its way to being an AI and science played a key role in reaching a National AI
superpower fit for the coming decade. Strategy and informing its direction. As we
move into an implementation phase, the AI
By their nature, strategies are a response to Council will continue to help galvanise action
the moment in which they exist - further from across the ecosystem in fulfilling our
actions will also be required to elaborate on objectives and holding the government to
the paths set out in this document in a way account on the actions contained in the
that responds to the fast-changing landscape strategy. The recently established Office for
in the years to come. A plan to execute Science and Technology Strategy, National
against the vision set out in this strategy will Science and Technology Council and National
be published in the near future. Alongside Technology Adviser will work with the rest of
this, we will put mechanisms in place to government to drive forward Whitehall’s
monitor and assess progress. science and technology priorities from the
centre. As a part of this, we will collectively
We will publish a set of quantitative indicators, identify the technological capabilities required
given the far-ranging and hard-to-define in the UK and in the government to deliver
impacts AI will have on the economy and the Prime Minister’s global science
society. We will publish these indicators superpower ambitions through AI.
separately to this document and at regular
intervals to provide transparency on our
progress and to hold ourselves to account.
Given the cross-cutting nature of AI,
collaboration across a wide range of sectors
and stakeholders will be paramount. The
Office for AI will be responsible for overall
delivery of the strategy, monitoring progress
and enabling its implementation across
government, industry, academia and civil
society.
We will also continue talking with the wider

community to get their feedback on AI in the
UK. Taken together, this quantitative analysis
and qualitative intelligence will enable us to
track progress and course-correct if we are at
risk of falling short in any particular area.
62 63
Published in September 2021

by the Office for Artificial Intelligence
© Crown copyright 2021
This publication is licensed under the terms of the Open Government Licence
v3.0 except where otherwise stated. To view this licence,
visit: nationalarchives.gov.uk/doc/open-government-licence/version/3
Where we have identified any third party copyright information you will need to
obtain permission from the copyright holders concerned.
ISBN 978-1-5286-2894-5
E02674508 09/21
This publication is available at www.gov.uk/official-documents

Any enquiries regarding this publication should be sent to:
enquiries@dcms.gov.uk
64 65
Any enquiries regarding this publication
should be sent to: enquiries@dcms.gov.uk
ISBN 978-1-5286-2894-5
E02674508 09/21
Published in September 2021

by the Office for Artificial Intelligence
© Crown copyright 2021

UK-National AI Strategy

Uploaded by

Copyright:

Available Formats

UK-National AI Strategy

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

UK-National AI Strategy

Uploaded by

Copyright:

Available Formats

National AI Strategy

Command Paper 525

Executive summary 7 AI deployment – understanding new dynamics 41

Summary of key actions 8 Creating and protecting Intellectual Property 42

Introduction 10 Using AI for the public benefit 42

10 Year Vision 11 Missions 44

The UK’s National AI Strategy 14 Net Zero 45

AI presents unique opportunities and challenges 16 Health 46

Reflecting and protecting society 16 The public sector as a buyer 46

The longer term 17 Pillar 3: Governing AI effectively 50

Pillar 1: Investing in the long-term needs of the AI ecosystem 22 Alternative options 54

Skills and Talent 22 Regulators’ coordination and capacity 54

Access to data 30 AI Assurance 58

Data Foundations and Use in AI Systems 31 Public sector as an exemplar 59

Public sector data 32 AI risk, safety, and long-term development 60

Compute 33 Next steps 62

make Britain a global This is the age of artificial intelligence.

The UK’s National AI Strategy therefore aims to:

■ Publish a review into the UK’s compute capacity needs to support

■ Monitor and use National Security and Investment Act to protect

■ Include trade deal provisions in emerging technologies,

Introduction The UK’s National Artificial Intelligence Strategy aims to:

The UK’s National AI Strategy

To remain an AI and science superpower fit for the next decade

adoption its position as a

Pillar 1: Investing in the Pillar 2: Ensuring AI Pillar 3: Governing AI

A growing UK Increased diversity in Certainty for the UK AI

Government activity in this strategy and over the next 10 years

AlphaFold & AlphaFold 2

• £250 million into Connected and

As part of the AI Sector Deal, the government

This strategy builds on the recent history of

To maintain the UK’s position amongst the

• Making sure the UK’s research, Skills and Talent

Research into the AI Labour Market showed

Attracting the best AI talent from around the world

environment. The report will feed into UKRI’s Tech Nation

AI and net zero

AI works best when presented with specific

Pillar 2 - Ensuring AI Benefits all Sectors and Regions

Data: A new direction consultation

Pillar 3 - Governing AI Effectively

The government will support the safe and

We will also continue talking with the wider

Published in September 2021

© Crown copyright 2021

This publication is available at www.gov.uk/official-documents

Published in September 2021

© Crown copyright 2021

You might also like