UK Government launches ambitious National AI Strategy

• United Kingdom

• Consumer
• Energy and infrastructure
• Financial services
• Health and life sciences
• Industrial engineering
• Technology, Media and Telecoms

27-09-2021

The UK Government has launched its eagerly anticipated National AI Strategy for the next decade.

The aim of the Strategy is to position the UK as a leading global superpower in AI, recognising the transformative power and wide-ranging potential of AI in all sectors of society and the economy and in providing solutions to some of the big issues of the day including climate change, national security and bounce-back from the COVID-19 pandemic. Its specific goals are for the UK to experience significant growth in discoveries that are not only made in the UK but are also commercialised and exploited there; to benefit from economic and productivity growth due to AI; and to establish the most trusted and pro-innovation system for AI governance on the planet – albeit on this later objective that is unsurprisingly a stated objective of other countries too.

The Strategy is set in the context of existing work and initiatives including the UK’s Plan for Growth, Innovation Strategy, Integrated Review, National Data Strategy and Plan for Digital Regulation and the anticipated National Cyber Strategy, National Resilience Strategy and Digital Strategy. However, the Strategy is expressed to be the beginning of a step-change for AI in the UK “recognising that maximising the potential of AI will increase resilience, productivity, growth and innovation across the private and public sectors”.

In order to achieve its stated goals the Strategy is founded on three pillars:

• Pillar 1: investing in the long-term needs of the AI ecosystem. This includes measures to attract and cultivate diverse AI skills and talent; the launch of a National AI Research and Innovation Programme; international collaboration on research and innovation; a new policy framework (to be published in the autumn) aimed at improving access to good quality, representative data which is crucial for the machine learning aspects of AI technology development; a consultation later in 2021 on potential options for a UK capability in digital twinning and shared wider cyber physical infrastructure for innovation; and an evaluation of the computing capacity needed to support AI innovation, commercialisation and deployment.

• Pillar 2: ensuring that AI benefits all sectors and regions. This involves supporting businesses to develop and commercialise AI technologies as well as supporting the demand side; i.e. understanding the factors that inform decisions taken by a business on whether to adopt an AI solution and then encouraging wider scale adoption. As part of this ambition the Government will launch a programme intended to stimulate development and adoption of AI technologies in high potential, lower AI maturity sectors. Other measures include the issue of a consultation later this year on AI and IP protection, as part of the Government’s ambition to support innovation and creativity, and the development of Innovation Missions that can incorporate potential AI solution to serious real-world problems including achievement of net zero.

• Pillar 3: governing AI effectively. The Government recognises the inherent tension between the need for AI to be regulated in order to ensure that it is safe, secure, fair, accountable, trustworthy and ethically sound and the need for regulation to be sufficiently flexible and light-touch so that it remains fit for purpose in a fast-moving environment without stifling innovation. The Government therefore aims to have a clear, proportionate, flexible and effective governance framework and it recognises that the current sector-led approach to regulation, whilst having many benefits, has the potential for inconsistent, fragmented and divergent approaches. It therefore intends for the Office for AI to publish a White Paper in early 2022 to set out the UK position on governing and regulating AI. The Government also intends to work with international partners on setting international technical standards for AI and there are various other policies and roadmaps in the pipeline, including a CDEI AI assurance roadmap, a cross-government standard for algorithmic transparency and updated guidance on AI ethics and safety in the public sector.

Setting a long-term strategy for the UK in relation to AI is critical and I am pleased to see this announcement, having been involved in inputting thinking as part of my role on the CBI Working Group for Artificial Intelligence. Whilst the UK is deservedly a leading power-house of AI talent, innovation and in attracting investment, it is essential that it builds on this and ensures that the UK has the necessary platform/foundations from which to cement its place as a global AI super-power. This includes, as part of the strategy, the need to attract international and in-country investment; support the current and future generation(s) of AI talent; and ensure trust in the building, adoption of, investment in and licensing or sale of AI technology within and into the UK. As part of this, it will be essential that the regulatory landscape the UK sets for AI is robust, factors in ESG considerations, and enshrines ethical and trustworthy AI into it, particularly in relation to more high-risk areas that rely on AI. It will be imperative for the UK to ensure that the laws and rules around use of data utilised by the AI (which the UK has already been looking at in relation to personal data via the Information Commissioner for some time) ensure clear, transparent and fair use, and do not allow biased results (whether due to the AI, the data used and/or the training of the AI). Key to driving trust and adoption will include ensuring the requisite level of oversight within businesses adopting AI into its core. But, balancing against all of this, the regulatory landscape must not stifle innovation or go too far in terms of its reach. Whilst a small number of countries and blocs have come out early with their proposed and actual legislation, arguably certain of that legislation is excessive in parts including applying to machine learning technology (as opposed to AI) that has been in place for years without commensurate proportional merit given to the purpose of that AI, creating unnecessary expense and administrative burden for companies innovating in this space or seeking to sell or license their AI in the relevant country/countries. Intellectual property rights as well as usage rights are critical parts of the equation for the UK getting this right and it is pleasing to see the Intellectual Property Office will be looking at this aspect as part of a collection of regulatory reforms as part of the strategy. The strategy also mentions potential use of tools to help drive adoption of and trust in AI technology. A number of other countries (including Hong Kong and Dubai) have already started utilising tools to assist with ethical adoption and the UK considering tools as a potential resource for organisations is to be welcomed, as long as these are well thought-through.

The Strategy is wide-ranging and ambitious and the passion and belief of those involved in shaping it is palpable when reading the document. It is undoubtedly an exciting time for the UK in the AI space. Whilst some of the aims and visions are inevitably broad-brush, there are a large number of important deliverables promised, together with a commitment to publish a plan to execute the vision “in the near future”. If the vision of the strategy can be realised and is implemented in the right way, it should absolutely lay the right foundations to drive investment, growth, adoption, and success both for the UK and the rest of the world, helping ensure the huge benefits that ethical AI can bring to societal issues are driven forwards - including  speed to drug discovery, cyber protection, dealing with energy and food shortages and more.

We will be closely monitoring developments.