Adarga comments on UK Government’s National AI Strategy

By  Adarga CEO Rob Bassett Cross & Head of Applied Science Dr. Daniel Clarke.

The National AI Strategy is powerful confirmation of the UK’s commitment to enhance and speed the development of world-leading technology so we can all realise the significant transformative benefits and opportunities that AI affords. It is recognition of the vital importance of AI for our future economic growth, prosperity, the protection of our society, our fundamental values, and our national security.

We share the government’s clearsighted belief in AI’s profound impact and benefits, and this strategy represents an excellent foundation on which the UK can reinforce and extend its competitive advantage in this vital field.

As a leading developer of AI software in Defence and National Security, Adarga is proud to be part of a thriving UK AI sector, revolutionising how organisations exploit data and information with cutting-edge technology. We look forward to helping the government to realise their strategy and have outlined key highlights, along with recommendations to help fulfil this ambitious vision.

Executive Summary

- The UK has a thriving AI ecosystem; to successfully execute the ambitious National AI Strategy we strongly encourage the government to call on this diverse community of expert, world-leading practitioners who are developing these innovative technologies. Building closer, more effective partnerships with industry and academia will help to increase the supply of skills and reinforce commercial adoption.

- We urge the government to prioritise efforts to ensure that the appropriate digital infrastructure and data foundations are in place, critical if the UK is to lead the world in AI. The government has a key role to play as an early adopter, leading by example in the application and use of commercialised technology.

- Clear articulation of the UK’s approach to governance, that ensures responsible AI development and use, should be accelerated in order to sustain the rapid pace of innovation and avoid divergence in international standards.

- We hope to see the ‘Next Steps’ of the National AI Strategy include measurable goals and appropriate funding commitments so the UK can realise this vision of global competitiveness and maintain a truly world-leading role.

People and skills

It is encouraging that the Strategy set out clear and strong support for the development of skills for the future, foundational for a thriving AI ecosystem. Importantly, the Strategy also recognises the widening gap between the increasing range of opportunities offered in the field of AI and supply of relevant technical expertise.

The measures proposed by the Strategy, aimed at developing and training the best people to develop AI-driven technology, are warmly welcomed. They build on a range of excellent initiatives already being undertaken to address supply-side constraints, such as the Institute of Coding’s Industrial Masters programme and the EPSRC’s funded Centres for Doctoral Training. Sustained investment in skills and education is vital if the UK is to continue to expand and sustain the brightest, most diverse AI workforce.

This is an exciting time to be a UK technology company. AI-driven technologies are set to transform every industry, in much the same way electricity did over a century ago. The advent of the Information Age has already catalysed AI’s ascension into a realm previously reserved exclusively for humans. During the next decade, AI will again make further leaps in capability and be a powerful accelerant for all other technologies. But the pace of change we are already witnessing will only get faster. The continued rapid cycle of innovation will also be reflected in an evolving demand for relevant, up-to-date skills.

It is imperative, therefore, that as the Strategy and supporting policies and plans mature, the government works in lockstep with industry and academia to understand, map and invest in the diverse range of developing skills, at all levels of education. This will not only ensure that effective technical career pathways are in place, all the way from early education through to becoming a data scientist or software engineer, but also provide opportunities to upskill those in jobs today. It is they, after all, who will make up the majority of the next decade’s workforce.

The dawning of a new era can be defined by the break it represents with the past. The increasing adoption of AI will catalyse a shift to a period when much more of our productivity and prosperity will be derived from the systems and machines we create. Data and information processing will become fully the domain of AI. Not only will many routine tasks be delegated to machines, but much of the typical lower-level, time intensive decision-making that currently consumes people in their day-to-day work will also be assisted by AI.

Economic advantage for the UK will not only be achieved by supporting people to enter new, AI-driven fields alone, but also by accelerating the deployment of a myriad of different applications across a range of industries and activities. Our people will remain our greatest asset in this endeavour; providing access to fundamental AI skills across all sectors and roles will underpin our future national capability. It is our role in industry to help the government to better understand the range of skills, technical and non-technical, that we will need in 3, 5, 10+ years to truly enable our economy and society to prosper.

Commercial adoption

Adarga is encouraged by the government’s vision for a pro-innovation environment and a business-friendly regulatory framework. Successful commercialisation and deployment of AI technologies will be key to maintaining the UK’s position as a global superpower in AI. We welcome the measures that seek to reinforce the UK’s already strong foundations in computational and mathematical sciences. Supporting the UK’s world-leading research and development (R&D) will be critical to the UK remaining at the forefront of developing the next generation of AI technologies.

However, whilst it is vital to consider new approaches to the development of cutting-edge, emerging technology, there is also an opportunity to accelerate the use of established and available capabilities. AI is not in its infancy; it already forms the technical cornerstone of most of the world’s most valuable and successful companies. The Strategy must, therefore, also help to cultivate a commercial environment that will support the creation, growth and sustainment of UK technology companies to replicate the success of contemporary global tech leaders.

Appropriate and sufficient levels of investment will be fundamental if those UK companies developing, and indeed adopting, such world-leading technologies are to achieve extraordinary commercial success. The scale of support required, across such a wide range of nationally critical technologies, will require both the public and private sectors to share the responsibility of investment.

This argues for further economic incentives to encourage and promote not only founders of exciting new companies, but also to attract greater levels of private investment capital to support early and growth-stage businesses. The government’s current review of the enterprise management incentive (EMI) scheme and tax credit system for qualifying expenditure on R&D provide two such areas to directly support increased investment in innovation and offer the opportunity for swift reform.

Where gaps emerge, as may arise in the case of some of the most promising but higher-risk technologies for national security or other niche strategic applications, strengthening the direct financial firepower of state-backed vehicles, such as the National Security Strategic Investment Fund, will also be crucial.

The government also has a key role to play as an early adopter, leading by example in the application and use of already commercialised AI products across the day-to-day business of government. This will require support for not only those technologies that have obvious national utility, such as environmental, defence or security applications, but also those traditionally thought of as purely commercial in nature. This will require swift evolution of the government’s traditional processes and practices; these are now out-of-step with modern approaches to the procurement of software and algorithm-driven capabilities.

As the government develops its further detailed plans for the execution of the first stage of the Strategy, we encourage policy makers to work even more closely with a broader number of members of the UK’s diverse technology sector to better understand the needs of the tech community - from the engineers, data scientists and founders who are already building the AI-enabled products and services, through to the investors who provide the capital for those businesses to grow.

Governance

Adarga welcomes the importance the Strategy places on ensuring the appropriate governance of AI technologies, and the importance of responsible AI development to protect the public and our fundamental values.

When it comes to regulation, swift and clear articulation of the UK’s approach must be prioritised if we are to all operate effectively in the complex international network of standards. We have already noted the pace, scale and impact of AI technology development. If the UK is too slow in setting out its approach to governance, we are likely to end up operating in a divergent, contradictory global patchwork of standards that will slow the pace of AI development and stifle innovation.

The pervasive spread and utility of AI means that centralised regulation is inappropriate. A self-regulatory or more agile, co-regulatory approach, with governance norms applied flexibly and adaptively, would enable policy safeguards while sustaining innovation. The UK must endeavour to lead the way, shaping and supporting a collaborative, international ecosystem that promotes effective standards and responsible development and use of technology. Such an approach would, of course, be underpinned by shared values of freedom, fairness, and democracy.

Infrastructure and data

AI advancements continue to outpace Moore’s Law with ever greater computation power being made available through new generations of processors and the hyperscaling of AI models leading to further increases in performance. From a technology standpoint therefore, it is excellent to see the Strategy stress the importance of computing power.

Whilst trustworthy and well-designed algorithmic models are a fundamental component of AI, it is the data and the enabling digital infrastructure that will provide the essential foundations for world-leading capabilities and empower the innovations that will push the boundaries of AI.

The UK must, therefore, place even greater emphasis on the provision and availability of the critical digital infrastructure required to accelerate, optimise and scale the development of AI-driven technologies, recognising the foundational role played by hyperscale cloud, open-source technology and accessible and appropriate data resources in driving forward AI innovation and adoption on a national scale.

This supporting digital infrastructure needs to form a fundamental mesh of capabilities across the UK, starting at schools and ensuring all children have access to a computer from an early age. We need to place as much emphasis on prioritising access to this infrastructure as we do high speed internet to ensure we can support and deliver the next generation of AI capabilities.

This is again where we urge the government to lead by example by being early adopters, users and champions of modern, data infrastructures as well as the custodians and distributors of shared, trustworthy data resources.

Partnerships

Closer cooperation and more effective partnership between government, industry and academia will be critical to realising the bold ambitions of the Strategy. The UK has one of the world’s leading AI academic communities; we must continue to strengthen this foundation by ensuring it is considering problems and challenges not only in the technology, but also in its industrial and societal application.

We think it will be more important than ever to have senior industry technologists sitting on research councils and in leadership positions in universities to contribute to steering research direction towards solving some of the most complex challenges the world faces using AI.

International partnerships and working within our existing global alliances will also be vital in successfully achieving the government’s ambitious vision and shaping the future of AI. In this respect we are particularly pleased to see the significant commitment made in the recent AUKUS tri-lateral security partnership announcement to deepen cooperation and sharing of AI expertise and resources. It is also excellent to see the Strategy set an immediate action to implement the US/UK Declaration on Cooperation in AI Research and Development.

We look forward to the publication of ‘Next Steps’; detailed, measurable plans for the execution of the first stage of the Strategy in the coming months. We and our team at Adarga stand ready to assist with drawing up these approaches if requested to do so. Whilst we urge the government to heighten its urgency to release the further papers and strategies on related areas, such as a broader Digital Strategy, the Defence AI Strategy, a National Resilience Strategy, and the outcomes of the ‘UK’s Data: A New Direction consultation’, we also ask the government to seek the wider council of the expert practitioners who will be behind the development and commercialisation of these innovative technologies. It is this expertise that will ensure that the course for this critical UK strategy may run straight and true. Achieving this vision of global competitiveness will require the UK to take a world-leading approach. We hope that the government will stand-by these ambitious commitments by focusing on the removal of barriers that hinder technology led growth and with appropriate funding in the forthcoming Spending Review.

The National AI Strategy can be found here.