Public Sector Innovation Conference
13 March 2024
This year's Public Sector Innovation Conference was a deep dive into the early uses of AI inn Government. We had a new conference format with lots of quick-fire examples of early “lighthouse” projects including inputs from the AI Safety Institute, Cabinet Office, CDDO, NCSC and No.10. Conference Chair Mark Thompson has written up the day in the link below.
Background
Gen AI burst into the public consciousness on 30th November 2022 and already has 100 million regular monthly users.
Gen AI is already being used in small pockets internally within public sector organisations, and in a much more limited way to support direct citizen interactions. However, usage remains piecemeal, as its implementation seems to hang mainly on the knowledge and existing skills of CTOs across Government.
In May 2023 The independent Committee on Standards in Public Life argued that public sector organisations are not being sufficiently transparent about their use of AI and that it is too difficult to find out where machine learning is currently being used in government. More recently the EU Bloc has called for less restrictive thinking by Governments when regulating generative AI, and arguing that there was a far greater risk of such behaviour stifling innovation.
Most recently, in November 2023 the UK government held the first Global Safety Summit to bring together international governments, leading AI companies and civil society groups to consider the risks of AI, especially at the frontier of development.
Themes for the day
Session One - Is AI Innovation?
AI had already led to the creation of intelligent systems, algorithms, and applications that could automate processes, improve efficiency, and generate insights from vast amounts of data and solve complex problems. So much for digitization: but could AI have helped us to ‘digitalize’ – to genuinely transform - our public service delivery models, by rethinking them altogether? Were we being ambitious enough? We looked at the government’s policy, thought about the relationship between AI and Innovation, and looked at examples of AI already being used in the public sector.
Session Two - Ethics and AI in the Public Sector
After coffee, we considered Ethics and AI in the Public Sector. When the public sector used AI, serious ethical questions arose that prompted careful consideration. For example: fairness and bias; transparency and explainability; employment and social impact; and security and safety. The session considered these ethical questions and discussed how trust, fairness, and accountability could be built into the deployment of these technologies.
Finally, we looked at how the Public Sector Procurement rules could allow the Public Sector to engage with this new rapidly developing technology. Had the 2023 Procurement Act sufficiently laid the way forward to allow experimentation, risk taking, and procurement of services where the ultimate outcome, methodology, and delivery mechanism was uncertain?
Session Three - AI for good and bad
AAfter lunch, we broadened the discussion to the topic 'AI for good and bad'.
How did we strike the balance between AI being used for bad (e.g., lowering the barrier to entry for criminality, or for cyber attacks) and AI being used for good (e.g., allowing otherwise marginalized societies in the global South access to technologies, or advancements in health research)?
So far, the global narrative around AI had been largely a negative one, reinforcing how AI was being used for bad, and not exploring enough the positive opportunities that come from AI.
Speakers
-
Imeh Akpan
Advisor Nesta’s Civic AI Observatory
-
Seb Barker
COO, Beam (Kingston work)
-
Geoff Connell
Director of Digital Services, Norfolk
-
Brigadier Stefan Crossfield
CDO, Army
-
Dr. David Gerouville-Farrell
AI Lead, FCDO
”
-
Dr. Keith Grimes
GP, Founder, and Principal Consultant
-
Yvonne Gallacher
Digital Director, NAO
-
Sabby Gill
CEO, DEXT
-
Jack Greenhalgh
CTO, Skin Analytics
-
Andrew Grill
The Actionable Futurist
-
Malcolm Harbour CBE
Connected Places Catapult
-
Shelina Hargrove
Head of Digital, MoJ
-
Harriett Harman
Member of Parliament, UK
-
Oliver Ilott
Director, AI Safety Institute
-
Clive Kelman
Principle AI Tech, CDDO
-
Alex Kilcoyne
Data Science Principal Adviser , Natural England
-
Dan Klein
Head of AI and Data, Zuhlke
-
David Lawton
Technical Director Informed Solutions
-
Maria Luciana Axente
Head of AI Public Policy, PwC
-
Niamh McKenna
CIO NHS Resolution
-
Professor Mark Thompson
Professor in Digital Economy, University of Exeter Business School
-
Sarah Peña
Head of Emerging Technology, Swindon Borough Council
-
Rebecca Rees
Head of Public Procurement at Trowers & Hamlins
-
Tremaine Richard-Noel
Head of Emerging Technology at Northampton General Hospital.
-
Himanshu Sahni
AI/ML Specialist, AWS
-
Rikesh Shah
CEO, IPEC
-
Ollie Whitehouse
CTO, National Cyber Security Centre
-
Sarah Wyer
Bias in Generative AI, Durham University
Agenda
The conference day was in three parts with interlocking themes. Each session comprised a keynote, a panel to discuss the topic in detail, and then 5 lighthouse examples of projects already underway to inspire the audience.
9:40: Session One: AI and Innovation in the public sector
Keynote: AI: the opportunity for the UK
Panel: AI and Innovation in the public Sector
Showcase: There’s an AI for that - Government
11:00: Session Two: Ethics and AI in the Public Sector
Panel: Ethics and AI in the Public Sector
Fireside chat: Procuring innovation
14:00: Session Three: AI for good and bad
Keynote: Ensuring the safety of AI
Panel: AI for Good and Bad
Showcase: There’s an AI for that - Health and Education
In collaboration with:
Venue
RSA House, 8 John Adam St, London WC2N 6EZ
Central London's award-winning RSA House is a focal point for social change and enlightened thinking whilst being at the forefront of numerous discoveries and inventions. Providing an inspiring venue, its charming Georgian features are balanced with progressive technology, high-ceilinged and bright rooms and atmospheric exposed brick Vaults.