Consultation outcome

Centre for Data Ethics and Innovation: Government response to consultation

Updated 20 November 2018

1. Ministerial Foreword

Earlier this year the government launched a consultation setting out our vision for the new Centre for Data Ethics and Innovation. We are creating the Centre to ensure our society keeps pace with the rapid developments in data-driven technology, supporting ethical and innovative uses of data and Artificial Intelligence (AI).

We want technology to work for everyone – people, businesses and society as a whole. This ambition is at the heart of the Digital Charter, our programme of work to agree norms and rules for digital technologies and put them into practice. Establishing the Centre is a core part of this programme.

There was significant interest in this consultation, reflecting the increasingly important role which data and AI are playing in our lives and the complex challenges they are bringing to the fore. The ability of data and AI to produce powerful insights about our behaviour can deliver immense benefits to our economy and society, for example by making public services more personal and more effective. However, this technology also has the potential to develop in more harmful ways. We need to be able to respond quickly and effectively and make sure technology is used in a way that serves humanity.

The establishment of the Centre is a landmark moment for data ethics both in the UK and internationally. As the first body of its kind to be established anywhere in the world, the UK now has an unprecedented opportunity to lead the global debate on how we want data to be used, making sure it delivers the greatest benefits to society as a whole.

Alongside the publication of the government’s response, we are also announcing the appointment of the Centre’s board. Under the leadership of the chair, Roger Taylor, the Centre is now in a position to begin work on some of the most significant issues facing society. Together with my colleagues across government, I am eagerly awaiting their findings. Finally, I would like to express my gratitude to all those who contributed their views during the consultation process. Your submissions have been carefully considered and your continued input and support will be vital to the future success of the Centre.

I will place a copy of this response in the Libraries of both Houses.

Rt Hon Jeremy Wright QC MP

Secretary of State for Digital, Culture, Media and Sport

---

2. Introduction

Last year, the government announced its commitment to establish the Centre for Data Ethics and Innovation – a major new advisory body that will investigate and advise on how we govern the use of data and data-enabled technologies, including Artificial Intelligence (AI).

In June, we launched a consultation on how the Centre will operate and its priority areas of work. The consultation ran for a 12 week period and involved a series of roundtable discussions across the UK. We received over 100 responses from a wide range of organisations and individuals (see Annex C).

This document sets out the main points to emerge from the responses and the position the government has adopted after reviewing these comments, including its rationale for taking forward or amending the original proposals.

2.1 Why the Centre has been established

Data and data-enabled technologies are playing an increasingly important role in our world – from the way we find information online to the way we diagnose illness – transforming many different areas of our lives in profound and positive ways. At the same time, these new technologies and applications pose complex ethical and economic questions that need to be addressed to ensure they work for the benefit of people and society.

Our existing legislation and regulatory regimes already provide the essential foundations for addressing the challenges posed by data and data-enabled technologies. However, if we are to harness the full potential of these technologies, it is vital that government creates the ongoing capability to identify and address any areas where clearer guidelines or regulation are needed, now and in the future. That is why the Centre for Data Ethics and Innovation has been established. The Centre is a core component of the government’s Digital Charter, which seeks to agree norms and rules for the online world. The Centre will enable the UK to lead the global debate about how data and AI can and should be used.

2.2 The consultation proposals

The government’s consultation document set out detailed proposals on the Centre’s advisory remit and activities. It is important to note that these relate to the Centre’s initial phase of activity. It remains the government’s intention to establish the Centre on a statutory footing at the earliest appropriate opportunity, subject to securing the necessary approvals and legislative space, and with this objective in mind we will continue to review the effectiveness of the functions proposed, which may need to be changed or supplemented as future needs are identified.

The government’s consultation document proposed three core functions for the Centre:

• analysing and anticipating gaps in governance and regulation that could impede the ethical and innovative deployment of data and AI;
• agreeing and articulating best practice, codes of conduct and standards that can guide ethical and innovative uses of AI; and
• advising government on the specific policy or regulatory actions required to address or prevent barriers to innovative and ethical uses of data.

In addition to the above, the consultation document proposed a set of principles and guidelines for how the Centre should operate within the wider stakeholder and institutional landscape. The proposals also included specific provisions for how the Centre should report its findings and make recommendations to government.

The Centre will be sponsored by the Department for Digital, Culture, Media and Sport and the chair will be accountable to the Secretary of State for the performance of the Centre.

2.3 Consultation responses

Respondents were invited to submit their views on all aspects of the proposals, including the Centre’s remit and objectives, its activities and outputs, and the way it will operate (see Annex A).

In general, respondents were supportive of the proposals for the Centre and the principle of establishing closer oversight of the way that data and data-driven technologies, such as AI, are used and regulated. There was also widespread recognition of the benefits and opportunities innovation in these technologies can bring if developed and deployed in a responsible manner.

There were a number of areas where respondents felt the proposals could be clarified and strengthened, or where they had specific suggestions for how the Centre should undertake its activities. A small number of respondents expressed concerns about the Centre’s proposed remit and potential impact, and questioned the rationale for establishing the creation of the Centre.

All suggestions and comments have been carefully reviewed and assessed by the government as part of its response.

2.4 Our response

This document sets out the government response to the consultation and the terms of reference for how the Centre will operate in its initial, pre-statutory form (see Annex A).

Taking into account the overall levels of support expressed by respondents and the detailed research which informed the design of the Centre, we have made limited changes to the original proposals. We have focused primarily on clarifying existing functions and strengthening the Centre’s reporting and recommendation functions. We have also set out where we expect the Centre to provide further clarity going forward and the mechanisms through which it should do this. We expect much of this to be addressed in the Centre’s first strategy document, due to be published by spring 2019.

3. 1. The Centre’s role and objectives

3.1 Summary of the consultation responses

There was strong support for greater oversight of the way that data is used. Most respondents welcomed the establishment of a new body to advise on the measures needed to ensure safe and ethical innovation and deployment of data and AI. They generally agreed that the Centre should not become a regulator, but should instead work with existing regulators to identify where ethical and governance challenges posed by the use of data and AI go beyond current law and practice.

Some respondents argued that there is an already complex and crowded governance landscape, and a proliferation of organisations and initiatives dealing with the challenges and opportunities that data and AI present. Many respondents stressed the importance of the Centre working closely with these existing organisations (including the Office for Artificial Intelligence and the AI Council) to bring consistency and coherence to the landscape, taking care not to duplicate work.

Respondents generally welcomed the Centre’s objectives to support both ethics and innovation, acknowledging these aims as closely linked and mutually reinforcing. It was observed that some respondents from business and industry expressed concern that the proposals focused excessively on the ‘ethics’ component of the Centre’s remit, with too little attention paid to ‘innovation’ and the varying needs of businesses of different sizes in this regard. Others, particularly those representing charities, non-governmental organisations, and academia, noted a possible tension between the ‘ethics’ and ‘innovation’ elements of the Centre’s remit, and expressed concern that the Centre may not be able to effectively challenge unethical practice while maintaining close links to business and industry.

A number of respondents emphasised the importance of the Centre building on the values, principles and norms already enshrined in UK and international law as it seeks to fulfil its remit. Furthermore, a small number of respondents felt existing legislation was sufficient to address the challenges posed by the use of data and AI, and questioned the need for the Centre to provide further guidance in this area.

3.2 Our response

The government welcomes the strong overall levels of support that respondents expressed for the creation of the Centre, while taking note of the voices of concern outlined above.

In line with the majority of respondents, we believe that the Centre has a key role to play in strengthening our governance frameworks around data and data-driven technologies, and so helping to build people’s confidence around their use.

The government acknowledges the need for the Centre to add clear value to the existing landscape, and believes the Centre has a unique and important role to play – convening, connecting and building on the work of existing institutions, and acting as the authoritative source of advice to government on the governance of data and AI. To help clarify its distinct role, the Centre should publish a strategic vision for how it proposes to operate with other organisations, including any formal agreement or memoranda of understanding it has reached. This strategic vision should also set out how the Centre intends to operate with the other institutions recently announced by the government; the AI Council and the Office for AI. ^{[footnote 1]}

We also recognise the complex relationship between innovation and ethics raised by many respondents. We believe the Centre should seek to support both ethics and innovation as mutually supporting objectives. As respondents recognised, the two are closely linked in the fields of data and AI innovation, where new advances routinely raise important ethical questions about issues as diverse as privacy, profiling, and liability. In this context, we believe stronger ethical guidelines will provide the clarity and confidence that is needed to drive the growth of responsible innovation. Equally, effective governance which supports well functioning markets and promotes innovation in the use of data will support valuable services for citizens.

Our ambition is therefore that the Centre supports both innovation and ethics – separately and together – through enhanced governance. However, we agree that the Centre should be mindful of potential conflicts of interest, particularly in areas such as data sharing, and the fact that the benefits and risks associated with data use differ according to the individuals and interest groups involved. It is precisely here that the Centre has a vital role to play: convening and engaging different views to identify solutions that will deliver the greatest possible benefits to society as a whole.

4. 2. The Centre’s activities and outputs

4.1 Summary of the consultation responses

There was broad acceptance of the three proposed functions for the Centre: to analyse and anticipate risks and opportunities, to agree and articulate best practice, and to advise on the need for action. However, respondents noted that together they represented an ambitious and broad set of activities, particularly given the Centre’s early stage of development and modest budget for the set up phase.

A number of respondents highlighted that the Centre’s ‘analysing and anticipating’ role should also include identifying regulation that was not fit for purpose or otherwise limited innovation, as well as identifying where further regulation may be needed. Other respondents stressed that the Centre should focus on collating and building on existing research from renowned research institutes, learned societies and think tanks. Some expressed concern that in seeking to anticipate the harms of future technologies, the Centre would risk recommending actions that stifled innovation through inappropriate regulation.

Respondents from business and industry felt that activities to ‘agree and articulate best practice’ should focus on guidance for companies using data, as well as the development of clear, sector-specific frameworks for ethical data use. This group of respondents also favoured soft governance measures, such as voluntary codes of conduct, in preference to additional regulation. Sector regulators also welcomed this activity, noting that the fast pace of change in the technology sector makes it harder to respond quickly and meaningfully to emerging ethical issues. Respondents from research and academia indicated that the Centre could develop infrastructure for sharing of knowledge and best practice amongst experts, such as mapping AI research initiatives and facilitating discussions amongst experts, including technicians.

A significant number of respondents also called for the Centre to establish clearer guidelines relating to the sharing of public sector data, to ensure that the large volumes of data gathered by the NHS and other public bodies could be used to greater effect. However, this proposal was not universally endorsed by respondents, with a small number expressing significant reservations about the Centre’s potential role in this regard and the public’s willingness to share their data. Respondents representing academia and small businesses also highlighted that lack of access to data is a barrier to research and innovation in AI, which they felt the Centre could address through the development of frameworks such as data trusts.

On ‘advising on the need for action’ many respondents indicated that the Centre’s priority should be to undertake projects investigating issues where there is a recognised need for immediate action. Suggested issues included data monopolies, the use of predictive algorithms in policing, the use of data analytics in political campaigning, and the possibility of bias in automated recruitment decisions.

A prominent theme in a number of the responses was the importance of active and extensive public engagement. Some felt the Centre should take a more active role in facilitating debate and building consensus around the ethics of data and AI use. A minority went further to suggest it has a role in delivering education and data literacy campaigns, to ensure the public is able to engage with the issues in an informed manner. However, it was recognised that any enhanced public engagement activity would require a commensurate level of funding.

There was also widespread support for the high-level themes proposed as a working framework for the Centre’s activities: targeting, fairness, transparency, liability, data access, and IP and ownership. However, respondents advised that the Centre will need to do further work to define the focus of activities within these broad and complex areas. Respondents also suggested other categories that could usefully frame the Centre’s work, for example, cross-cutting themes such as privacy and consent or specific areas of focus such as the future of work.

Some respondents stressed that the ultimate goal of a strong, trusted and dynamic governance regime will only be achieved through a work programme that balances both the ethical and innovation concerns, and that this would be most effectively developed by close engagement with key stakeholders.

4.2 Our response

The government welcomes the strong degree of support that respondents expressed for the Centre’s three core functions. We note the reservations that a small number of respondents expressed in relation to the Centre’s anticipatory function, but believe this will be central to the Centre’s role in identifying how to support and enable the development of responsible innovation.

The Centre has a critical role to play in facilitating, shaping and informing public debates about how to use and regulate data and AI. This is something the Centre will be uniquely placed to do through its convening role and it is fundamental to all three functions outlined in the proposals. However, we do not consider it would be appropriate and within scope for the Centre to extend its engagement activities to encompass large scale education and data literacy campaigns. We believe there are other organisations within civil society and government that are better placed – in terms of their resources and established expertise – to carry out these activities. Depending on the priority areas of its initial work programme, the Centre may have a role in identifying what form and objectives such educational or literacy campaigns should take.

More broadly, it is important to reiterate that the Centre is being established on an interim footing. This will allow the government time to test the value and utility of the Centre’s functions ahead of the creation of a future statutory advisory body, and to identify how these might need to be expanded or adapted going forward. The findings and views of the interim Centre will have an important role to play in this assessment

We also welcome the support which respondents expressed for the thematic working areas outlined in the consultation. We recognise the broad scope of these areas and the corresponding need for the Centre to focus its activities carefully within them, and the possibility of extending those themes to include other cross-cutting or specific issues, with the caveat that these issues must relate directly to the Centre’s core focus on governance and regulation.

As respondents have pointed out, the list of projects the Centre could undertake is potentially vast. We recognise the need for the Centre to carefully prioritise its activities to ensure it is delivering the greatest possible value. Prioritisation should reflect:

1. the value generated by projects, in terms of their impact on innovation and public trust in the ethical use of data and AI;
2. the rationale for the Centre doing the work (relative to other organisations, in or outside government); and
3. the importance of undertaking projects now, either to address issues which have been identified as urgent or to identify and anticipate longer term challenges.

As announced in the 2018 Autumn Budget, the Centre has been commissioned to study the use of data in shaping people’s online experiences, and the potential for bias in decisions made using algorithms. An interim update on progress is expected in summer 2019.

The chair of the Centre must agree its annual work programme with the Secretary of State for Digital, Culture, Media and Sport. We will agree the exact governance mechanisms with the chair and board following publication of this response.

5. 3. How the Centre will operate

5.1 Summary of the consultation responses

Overall, there was strong support for placing the Centre on a statutory footing, with many respondents agreeing that this would help the Centre build its long-term capacity, independence and authority. A smaller number expressed concerns that this could potentially confuse the regulatory landscape.

There was considerable consensus amongst respondents that the Centre should only be established as a statutory body after an initial phase of activity, when its functions and work programme have been reviewed and refined. There was less agreement on what statutory powers, if any, the Centre might need at this stage, although a number of suggestions were offered (for example, the power to request information).

Many respondents highlighted the need for the Centre to effectively engage with the numerous institutions and organisations already working in this field (nationally and internationally), drawing upon existing research and expertise. It was particularly emphasised that the Centre should work closely with regulators – including the Information Commissioner’s Office, the National Data Guardian and sector regulators – drawing upon their specific expertise and taking care not to duplicate their functions. Similarly, respondents argued that it was important for the Centre to clearly articulate its role in relation to the Office for AI and the AI Council (and vice versa) as its activity and work programme develops. It was felt this would secure productive, efficient and non-duplicative relationships with stakeholders who are likely to be interacting with all three bodies.

A number of respondents also noted the importance of the Centre fully engaging across the UK with the different centres of expertise and innovation. This included suggestions that the Centre should itself be located outside of London in proximity to these regional hubs. The need for the Centre to work closely with the devolved administrations was also emphasised.

More broadly, many respondents emphasised the importance of the Centre seeking to shape debate and best practice in international forums, arguing that it was only through international agreement that the cross-jurisdictional issues of data and AI governance can be adequately addressed. Respondents noted the potential for the UK to play a leading role in shaping these international discussions at a time when many countries have begun to grapple with similar questions, and the important contribution which the Centre itself could make.

In relation to the Centre’s reports and recommendations, a number of respondents stressed the importance of swift response from government given the rapidly evolving nature and impacts of data innovation and data-driven technologies. In view of the strong public interest in these issues, respondents made repeated calls for the Centre to report to Parliament on its findings and recommendations.

Respondents also strongly supported a transparent approach to the Centre’s activities and recommendations, focusing in particular on the need for board meetings to be open to the public as far as possible, and for recommendations to be made public at the point at which they are delivered to government. Exceptions were noted around sensitive data and anything pertaining to national security.

5.2 Our response

The government remains committed to establishing the Centre on a statutory footing after its initial phase of operation and views this as the most effective way to secure its long-term credibility, accountability and effectiveness. This will be pursued based on a careful assessment of its statutory needs and informed by the interim Centre’s early findings and recommendations. In the meantime, we will ensure that the interim Centre is able to work effectively and independently, and will seek to maximise the opportunities for it to develop its own identity and voice during its pre-statutory phase of work.

We expect the Centre to publish its operational strategy by spring 2019, which should include details of how it will work with the different institutions and stakeholder interest groups, including any formal memoranda of understanding it has agreed. This must also capture how it intends to work with devolved administrations and build in a nationwide, representative perspective.

While it is imperative that the Centre engages fully across the UK, it is important to note that, in its initial, pre-statutory form, the Centre will be located in London. This reflects the organisational status of the interim Centre, which is being established as an advisory committee of DCMS and staffed by civil servants from the department. The decision on where the statutory Centre will subsequently be located will be decided in line with Cabinet Office guidance and following a careful analysis of needs and impacts.

We recognise the importance of ensuring that the Centre’s reporting mechanisms reflect the pace of technological development in this area, the urgency of the issues data innovation is raising and the strong public interest in how they are addressed. As such, we have strengthened our original proposals to include provision for:

1. The government to respond to the Centre’s substantive recommendations no later than six months after the recommendations have been made; and
2. The Centre to submit an annual report to SoS outlining its recommendations, activities and progress to date, which the SoS will lay before parliament.

In addition – and in line with our original proposals – the Centre should also publish an overarching assessment of data use and the governance landscape, including any recommendations it has made and the steps the government has taken to implement them. The first of these assessments should be published at the end of the Centre’s initial, pre-statutory phase of work, with further assessments following at periodic intervals thereafter.

Finally, we remain committed to the Centre operating as transparently as possible. Reports will therefore be made public. We likewise expect that, where possible and practical, board meetings are open to the public.

---

6. Annex A

6.1 Centre for Data Ethics and Innovation: Terms of Reference

The Centre for Data Ethics and Innovation will help to maximise the benefits of data and Artificial Intelligence (AI) for our society and economy. The Centre will:

1. Analyse and anticipate risks and opportunities. The Centre will identify the opportunities for strengthening ethical and innovative uses of data and AI by:

a. Commissioning and bringing together research and analysis into the ethical and economic implications of uses of data and AI

b. Reviewing the existing regulatory framework to identify gaps in response to the uses of data and AI and barriers to ethical innovation

c. Consulting regulators, industry, public bodies and civil society on specific issues relating to the use and governance of data and AI

d. Engaging citizens and consumers to understand the range of societal attitudes towards the use and regulation of data and AI

e. Horizon-scanning new and emerging data-driven and AI-based technologies and associated governance implications

2. Agree and articulate best practice. The Centre will identify best practice for the responsible use of data and AI. This might include:

a. Engaging industry bodies, public service providers and consumers to coordinate world-leading standards and codes of conduct in data and AI uses

b. Responding to, and seeking to shape, the international debate on standards

c. Working with stakeholders to identify and assess effective and ethical frameworks for sharing data

d. Advising on measures to build capability amongst data users, through training or organisational governance

e. Working with the public and the private sector to investigate the utility of technological approaches to key challenges

3. Advise on the need for action. The Centre will support the government to enable safe and ethical innovation in the use of data and AI by:

a. Identifying steps to ensure that the law, regulation and guidance keep pace with developments in data-driven and AI-based technologies

b. Publishing recommendations to government on how it can support safe and ethical innovation in data and AI through policy and legislation

c. Identifying opportunities to inform and influence international agreements and frameworks for the use of data and AI

d. Providing expert advice and support to regulators on the implications of data and AI uses and areas of potential harm

In carrying out the above functions, the government expects the Centre to:

• appropriately balance objectives for ethical and innovative uses of data and AI to ensure they deliver the greatest benefit for society and the economy
• take into account the economic implications of its advice, including the UK’s attractiveness as a place to invest in the development of data-driven technologies
• provide advice that is independent, impartial, proportionate and evidence-based
• work closely with existing regulators and other institutions to ensure clarity and consistency of guidance

In addition to this, as part of its initial, pre-statutory phase of activity the Centre will:

• explicitly review these functions to ensure the Centre’s resources are being deployed in the most effective way
• identify what additional functions the Centre may need to undertake to deliver its mandate effectively
• assess where these functions may need to be amended or augmented with specific powers when the Centre is established on a statutory footing

---

7. Annex B: Summary of Consultation Questions

1. Do you agree with the proposed role and objectives for the Centre?
2. How best can the Centre work with other institutions to ensure safe and ethical innovation in the use of data and AI? Which specific organisations or initiatives should it engage with?
3. What activities should the Centre undertake? Do you agree with the types of activities proposed?
4. Do you agree with the proposed areas and themes for the Centre to focus on? Within these or additional areas, where can the Centre add the most value?
5. What priority projects should the Centre aim to deliver in its first two years, according to the criteria set out above?
6. Do you agree the Centre should be placed on a statutory footing? What statutory powers does the Centre need?
7. In what ways can the Centre most effectively engage stakeholders, experts and the public? What specific mechanisms and tools should it use to maximise the breadth of input it secures in formulating its actions and advice?
8. How should the Centre deliver its recommendations to government? Should the Centre make its activities and recommendations public?

---

8. Annex C: Written consultation responses and roundtables

8.1 Written responses

We received over 100 responses, including from the following organisations. Individual respondents have not been named.

Academy of Medical Sciences

Advertising Standards Authority

Alan Turing Institute

ARM

Association of Accounting Technicians

Association of Medical Research Charities

Asthma UK

BAE Systems

Barclays

Biometrics and Forensics Ethics Group

British Academy & Royal Society (joint response)

British Heart Foundation

British Insurance Brokers’ Association

British Library

British Standards Institution

BT Group

Codeplay

Coelition

Community Foundation

Crowe

DeepMind Technologies

DefendDigitalMe

Department of Health and Social Care

Digital Catapult

Direct Marketing Association

Doteveryone

EDF Energy

Ethical Leadership

Ernst & Young

Financial Services Consumer Panel

Federation of Small Businesses

Genetic Alliance UK

Google

Hazy & Adapt (joint response)

Health Data Research UK

Health Research Authority

Helena Kennedy Centre for International Justice

House of Commons Science and Technology Select Committee

House of Lords Select Committee on Artificial Intelligence

Human Rights Centre at Essex

IBM

Information Commissioner’s Office

Institute and Faculty of Actuaries

Institute for the Future of Work

Institute of Advanced Legal Studies

Involve

ISACA

The Law Society

Leverhulme Centre for the Future of Intelligence

Mastercard

medConfidential

Microsoft Corporation

Minister for Public Finance & Digital Economy (Scotland)

Money and Mental Health Policy Institute

National Data Guardian

National Physical Laboratory

Nesta

The Nia Project

Nuffield Council on Bioethics

Nuffield Foundation

ORBIT

Ofgem

Open Data Institute

The Open University

Ordnance Survey

Pearson

PHG Foundation

Pinsent Masons

Polygeia

Positive Change UK

PricewaterhouseCoopers

R3

Regulatory Institute

Royal Academy of Engineering

Royal College of Physicians

Royal Geographical Society

Royal Statistical Society

Sage Group

Samsung Group

Science Advisory Council

Scottish Care

Sefton Council

Sense about Science

Society of Motor Manufacturers and Traders

Tableau Software

techUK

UKCloud

United Kingdom Accreditation Service

United Kingdom Research and Innovation

University of Manchester University of York

use MY data

Wellcome Trust & Understanding Patient Data (joint response)

West Midlands Open Data Forum

Which?

Wikimedia Foundation

8.2 Roundtables

We are grateful to the following organisations that generously organised roundtables for the government over the consultation period:

• Alan Turing Institute
• Confederation of British Industry
• Nesta
• techUK
• UK Regulators Network
• Which?

1. The Office for AI is a joint BEIS-DCMS unit responsible for overseeing implementation of the AI Sector Deal and putting the UK at the forefront of the AI and data revolution. Its mission is to drive responsible and innovative uptake of AI technologies for the benefit of everyone in the UK. It will do this by engaging organisations, fostering growth and delivering recommendations around data, skills and adoption. The Office for AI will act as secretariat for the AI Council, which will be an expert committee formed of independent members, convening experts from across industry, academia and government to create a strong dialogue, provide high level leadership and momentum for the implementation of the Industrial Strategy Sector Deal. ↩