The exhibition 2023

How does the synthetic population catalyst (SPC) simulate the populace’s daily activities, and how can we use this to solve issues of sustainability and life sciences at a population level? Using a spatial microsimulation model, The Alan Turing Institute’s Hadrien Salat and team create synthetic individuals mimicking the socio-economic, health and time-use characteristics of the British people. Drop in to meet the team and try out the software yourself, modelling the role of BMI in the pandemic, and climate change.

Interested in learning more? Some of the research featured was brought to you in collaboration with The Alan Turing Institute’s AI for science and government fund and Engineering and Physical Sciences Research Council (EPSRC).

While recent times have seen an unprecedented explosion in both the quality and availability of satellite data imaging the Earth, the information is highly unstructured. Daniel Arribas-Bel and colleagues (University of Liverpool) are using AI to bring it all into focus and put to use what the raw data gives us – the most accurate ever representation of Earth’s natural and built environments – to model, understand and plan the built environment. Drop in to meet the team behind the tech and gain a comprehensive, interactive introduction to this fascinating work.

Through the use of interactive 360 videos and VR headsets, UCL SpaceTimeLab’s Tao Cheng and Dr James Haworth (in collaboration with GLA and TfL) give three interactive case studies of urban footfall, travel and transport habits, and cycling safety. Meet the SpaceTimeAI team and get the latest across three of their most relevant and deep-dive projects, looking at the recovery of 600 London high streets post-pandemic, how low-traffic neighbourhoods are changing residents’ and traffic behaviours, and using the latest tech to enhance cyclists’ safety.

Polly Hudson (The Alan Turing Institute) showcases the Colouring Cities Research Programme (CCRP): a global collaboration programme utilising AI and machine learning to support international progress towards the UN’s Sustainable Development Goals – specifically, around our planet’s built environment. Currently working in eight far-reaching countries, CCRP facilitates knowledge and data sharing across and within governments, industry, academia and the third sector. Stop by to meet the team, learn about the project’s details and discover what’s next for collaborative learning and how we tackle issues of fragmentation, quality and accessibility – among others – in data.

Interested in learning more? Some of the research featured was brought to you in collaboration with The Alan Turing Institute’s AI for science and government fund and Engineering and Physical Sciences Research Council (EPSRC).

With high-stakes factors such as environmental health, people’s health and quality of life, and economic vitality influenced by land use, decision-making in this area is extremely complex. These far-reaching impacts are explored at Anna Zanchetta’s (The Alan Turing Institute) demo, in partnership with the Geospatial Commission and Turing colleagues. Their innovative programme constitutes a modelling system that leverages data science and AI to cut through inherent complications in land use determination: visit the interactive demo to find out more and meet the minds behind the tech.

While there is an almost endless range of situations in which artificial intelligence can have huge and meaningful impact, does it always function perfectly without a human touch? Headed by Mohammad Soorati, this University of Southampton team demonstrates their unique simulator, which utilises varying ratios of autonomy to human input for maximum efficiency. Visit their AI UK 2023 demo for a hands-on experience, meeting the team – and the swarm – and aiding decision-making in a search and rescue context.

Headed by Zena Wood (University of Exeter), this collaborative exhibition from the universities of Exeter, Liverpool and Surrey showcases the work of the Dstl-funded Data Defence Research Centre (DDRC). The transdisciplinary centre brings together experts from data science, computer science, and social science to help the Defence Science and Technology Laboratory (Dstl) use data more effectively. Visit the demonstration to deep dive into some of the DDRC’s fascinating ongoing data science and AI projects, advancing technical and cultural capabilities.

The answer to those sub-par holiday snaps may be just a click away at AI UK: discover David Berman’s (Queen Mary University of London, Cambridge Consultants) image distortion clean-up. Stop by the SharpWave demonstration to upload a photo and watch in real-time as the machine learning tech cleans up your pic. SharpWave utilises a generative adversarial network (GAN) machine learning framework, and the demo features camera, physical distorting semi-transparent screen and display screen, so you can witness machine learning in action.

Taking visitor attractions to the next level, Tag With Me offers location-based automated guides, AI recognition and interactive treasure hunt apps for parks, cultural destinations and visitor attractions. Eiman Kanjo’s (Nottingham Trent University) IoT and proximity-based digital game can also be used in a variety of beneficial settings. Stop by to find out how the AI has been put to use at Highbury Mental Health Hospital, Nottingham County Council and Rufford Country Park to improve patients’ health and well-being in a fun and interactive way.

Escape the dreary every day in this completely immersive, generative virtual reality (VR) environment. Plug in and enjoy the most advanced tech ever produced by this team in an organic generative AI virtual space, where your body movements cause the VR environment to respond and change shape. Not only is this impressive tech beautiful to see and experience, but its range of applications is remarkable, too. Stop by to meet the team – led by William Latham (Goldsmiths) – and find out how it’s been applied to protein folding and virus visualisation, including COVID-19.

Need to find a new pair of shoes online, but finding it hard to come up with a sufficiently precise text description? Yi-Zhe Song (University of Surrey) has you covered. This demo shows you a ground-breaking new paradigm where you can take out your phone, quickly scribble a rough sketch and watch as the system returns all shoes matching your sketch. The result of over a decade of sketch research at the University of Surrey’s SketchX lab, the system works across different drawing styles – and doesn’t baulk if your artistic skills are somewhat underdeveloped.

Drop by this futuristic demo to find out how you can build your very own metaverse – simply by finger scribbling in the air. This bleeding-edge research from the SketchX research lab (University of Surrey) uses the latest AI technologies in the form of deep neural networks, which convert your rough scribbles into realistic 3D objects. Visit Yi-Zhe Song’s (Centre for Vision, Speech and Signal Processing) demo for the chance to wear a VR headset and experience this immersive tech reality first-hand.

One of the most advanced polar research vessels in the world, RRS Sir David Attenborough represents the UK Government‘s largest investment in polar science since the 1980s. But, with shipping the largest carbon emissions contributor of the British Antarctic Survey (BAS), the BAS’ goal to be net zero by 2040 could be in jeopardy. See how George Coombs’ (BAS) AI virtual modelling system creates navigational routes that optimise fuel efficiency. Discover the RRS Sir David Attenborough Digital Twin, get to know the cutting-edge route-planning tech and plan your own Antarctic explorer routes at AI UK.

Following its appearance at AI UK 2022, IceNet is now here to demonstrate its impressive real-world progress. The machine learning’s operational (MLOps) capabilities and specific applications have grown in both predictive and deliverables capability on the key environmental goal of sea ice forecasting since 2022. Gain insight into MLOps architecture and utilisation of digital twin tech, experience tech demonstrations and explore the future possibilities of IceNet, from James Byrne and Ellen Bowler’s (British Antarctic Survey).

Interested in learning more? Some of the research featured was brought to you in collaboration with The Alan Turing Institute’s AI for science and government fund and Engineering and Physical Sciences Research Council (EPSRC).

Did you know that the world’s first underground farm is 33 metres below Clapham High Street? And, through sustainable and hydroponic farming techniques and a Cambridge-based digital twin, it’s shaping the future of urban farming. Its plants yield a harvest roughly every 10 days all year round, grow without soil on carpet cut-offs that were destined for landfill and are watered by tanks housed even further underground. Everything is measured and represented in a virtual 3D representation of what’s happening on the ground: the subterranean farm’s digital twin. Discover more about this incredible working project at AI UK 2023.

Interested in learning more? Some of the research featured was brought to you in collaboration with The Alan Turing Institute’s AI for science and government fund and Engineering and Physical Sciences Research Council (EPSRC).

The burgeoning landscape of digital twin technology has the potential to make almost unimaginable impact across in a vast array of industries. These virtual representations of physical systems can encompass connectivity, accessibility, data and models. Drop by this University of Sheffield and Turing Research and Innovation Cluster in Digital Twins (TRIC-DT) demo to discover the BAE T1A Hawk aircraft digital twin, which uses the DTOP framework to provide a simple browser-based user interface, connectivity to recorded datasets, access to remote computational resources, and connectivity to expert-made models using commercial software.

Researchers at King's College London are looking at ways we can learn more about and better treat heart conditions like arrhythmias and heart failure. By creating a digital replica of a person's heart, the “digital twin” of the heart can be used to predict responses that cannot be obtained experimentally; for example, to see what a patient's response to a particular treatment method might be. These models can also be 3D printed to understand better their shape. Drop by this demo – part of the TRIC-DT group – to learn more, find out what these digital twins look like and even see how pacemakers have evolved over time.

Harnessing the power of digital twin tech to predict and mitigate the negative impacts of climate change, reduce natural hazard impact and support sustainable development, are key focuses of the Turing Research & Innovation Cluster in Digital Twins (TRIC: DT). Working with the Natural Environment Research Council (NERC/UKRI), this demo showcases a pilot digital twin of the Haig Fras Marine Protected Areas (UK), used to monitor biodiversity.

The mission: accelerate the world’s transition to renewable energy. The challenge: the planning process. The development of crucial infrastructure assets such as transmission lines, off- and onshore power cables and pipelines can be fraught with expensive and time-consuming setbacks. The infrastructure needed to meet net zero targets simply cannot be designed and approved in time with current methodologies. Continuum Industries are using AI to help major energy networks bridge the gap between existing procedures and the pace at which project development needs to happen.

Tackling the climate crisis one scoot at a time, University of Bristol’s James Thomas and Sam Gunner have taken pollution monitoring on the road by developing e-scooter mounted pollution monitoring hardware capable of measuring pollution in multiple locations and with flexibility hitherto unachievable. Focusing on public health impacts such as particulate matter, volatile organic compounds, carbon monoxide and nitrogen dioxide, the devices are also able to measure ambient noise. The project is ready to be trialled in Bristol this year: have a sneak peek at the hardware in action at AI UK 2023.

Do you live near a local street not designed for high vehicle traffic, but see many drivers shortcut through? Low-traffic neighbourhoods (LTNs) help keep drivers on main roads, but they can be hard to design properly. Come and meet Dustin Carlino (The Alan Turing Institute) to try out an interactive tool to decide how to place modal filters and adjust one-way streets. You can design an LTN anywhere in the UK with just a web browser, share the result with neighbours and even propose to your local council.

Interested in learning more? Some of the research featured was brought to you in collaboration with The Alan Turing Institute’s AI for science and government fund and Engineering and Physical Sciences Research Council (EPSRC).

Using data science and AI to quantify and visualise the risks climate change poses to populations, infrastructure and the economy, CLIMAR creates accessible decision-ready information for industry, policy makers and the public alike. The University of Exeter’s Matthew Thomas and the Met Office’s Theo McCaie present this game-changer on the path to Net Zero, crucial across multiple sectors and key in enhancing our resilience to climate change for everyone and everything, from industry to society’s most vulnerable individuals.

Displaying the enormous breadth and power of digital twins in transforming our world, this forward-thinking demo shows how new technologies can be leveraged to solve complex sustainability issues. Heavy goods vehicles (HGVs) generated over 4% of greenhouse gas emissions in 2019, and up to 30% of their journeys were freight-free: at AI UK 2023, Liming Xu and team (University of Cambridge) showcase an autonomous supply chain digital twin, including self-driving lorries. Can it influence the UK’s carbon emissions? Visit the interactive demo to find out.

With multiple recent events throwing the speedy spread of misinformation into the spotlight, Yulan He (King's College London) and team (University of Warwick, Queen-Mary University of London, Cranfield University) have worked together to shed light on the darker corners of the internet. Stop by their exhibition for a live demo of their cutting-edge web-based misinformation detection system, PANACEA. Focusing on COVID-19-related claims online, the AI-enabled evidence-driven framework for claim veracity assessment fights misinformation in two key ways: fact-checking and rumour detection.

With stroke a major cause of death and long-term disability, Michael Allen and team (University of Exeter) have implemented machine learning to explore inter-hospital variation in clinical practice in the emergency stroke pathway. Supported by Stroke Association and the National Clinical Director of Stroke, SAMuel elucidates those patient characteristics that most influence the varying treatment decisions across the country. Visit the demo to see this machine learning and data science in action, and find out about the team’s detailed stroke outcome models, predicting likely disability.

This pioneering live demonstration featuring the companion robot and digital twin from Alexandre Colle (Edinburgh Centre for Robotics) combines innovative design, technology for social good, and data ethics. The customisable and sustainable robot, Maah, is a complex sensory device with a pillow-like form that uses a wide range of skins developed with 3D knitting technologies. Able to capture many data from users, the robot and its digital twin can reveal to both users and care stakeholders a wide range of relevant information.

Has AI really been as reliable as some thought in identifying COVID-19 infection? Harry Coppock (Imperial College London) and team’s interactive demo of AI audio models shows data in action through voice and cough audio sampling and interactive heat maps. Aimed at educating policy makers, AI practitioners and members of the public as to the dangers of falsely inflated AI performance, this project has input from across the UK and abroad with the demo featuring findings from the Bioacoustics Markers team in the Turing-RSS Health Data Lab.

Can AI decision systems reduce treatment delays? Medical imaging is routinely used for cancer diagnosis and treatment monitoring, but newer analysis techniques can lead to bottlenecks in treatment journeys. Lorena Escudero Sanchez (University of Cambridge), alongside colleagues from UK and Swedish research institutes, gives a live demo of how they have integrated deep learning models for automated segmentation of ovarian cancer, and how they are using machine learning models to predict chemotherapy response.

Discover how Charles Harris (University of Cambridge, Cambridge Centre for AI in Medicine) and team are designing new drugs using AI models. This forward-thinking demo displays a highly visual generative process, showing how state-of-the-art technologies can generate new drugs within a protein pocket structure. Find out how diffusion models could supercharge drug discovery at the demo: choose your own molecules and properties, optimise them and watch the model do its work.

Can AI be used to detect future dementia in patients? Working collaboratively to build AI models – as well as the necessary supportive research infrastructure, public engagement and policy strategy – this collaboration between Alzheimer’s Research UK, the Alan Turing Institute, the National Physical Laboratory and Oxford, Cambridge, Exeter and Newcastle Universities has a resounding answer: absolutely. Visit the EDoN demo to learn about this initiative utilising an inexpensive, low-burden digital toolkit to develop machine learning models capable of detecting dementia 15-20 years in advance of current techniques.

Discover the future of x-ray assessment with the University of Manchester’s Claudia Linder and team: BoneFinder is the software tool that automatically locates bones and joints in x-rays. It’s already being used across the world, enabling researchers and clinicians to examine differences between patient groups, how bone shapes are changed by disease, and whether certain shapes make someone more likely to suffer a particular disease. Stop by the team’s demo to see the software in action on new images, including x-rays from children with Cerebral Palsy.

Young people’s education, work and social life is increasingly influenced by AI systems. This project from Coventry University provides innovative opportunities for young people to deepen their understanding of AI through an interactive audiovisual performance – God Mode – and an online game – God Mode Epochs. Created by artist duo dmstfctn, both the performance and game (which run in real-time 3D software) aim to demystify AI by inviting the audience to interact with the process of training such systems. With sell-out performances under their belt, the team invites you to visit their demo at AI UK 2023 and find out more.

AI trailblazers have made leaps and bounds when it comes to tech progress: but how good are we at working together internationally? Dedicated to advancing trustworthy and responsible machine learning, AI Standards Hub's Florian Ostmann (the Alan Turing Institute) leads on developing an interactive online platform offering a range of pathbreaking tools and resources to support you. Drop by to watch a demonstration of the platform’s features, explore the platform on your own device and engage with members of the Hub team. Demo partners: DCMS, Office for AI, British Standards Institution, and the National Physical Laboratory.

Get hands-on with Policy Priority Inference (PPI), a brand-new web app created at the Turing, and explore the nuances of allocating government spending. Upload development indicators to use this AI tool to simulate policy experiments and investigate budgetary scenarios using an evidence-based decision model. At AI UK 2023, you can watch demonstrations on how to use this app and even try it out for yourself.

Interested in learning more? Some of the research featured was brought to you in collaboration with The Alan Turing Institute’s AI for science and government fund and Engineering and Physical Sciences Research Council (EPSRC).

10-20% of UK adults have experienced online abuse and, with large-scale public abuse part of daily life for many, it’s clear we need new and improved safety measures. Project-led by Jonathan Bright (The Alan Turing Institute), the Turing Online Safety team are using state-of-the-art language models to automatically detect abuse in social media data and produce responses. Explore the science behind these models and discover how data is created, with interactive language models and real-time results, at the team’s AI UK 2023 demo.

Interested in learning more? Some of the research featured was brought to you in collaboration with The Alan Turing Institute’s AI for science and government fund and Engineering and Physical Sciences Research Council (EPSRC).

Mobilising for Data Justice is the 20-minute documentary produced by the ethics team at The Alan Turing Institute and Fable Studios that fills the gap in data justice research and practice, and provides resources that help policymakers, practitioners and impacted communities gain a broader understanding of data governance. This demonstration provides an accessible introduction to the broader project using audio-visual elements as well as the interactive premiere of the documentary-style film. Be among the first to see the premiere of the second episode in a three-part documentary series, at AI UK 2023.

There’s no doubt that AI is revolutionary to many key industries. But, with greater data science comes greater data responsibility; especially when it comes to healthcare, education, finance and more. We’ve seen great progress on this with the advent of Trusted Research Environments (TREs) – computing environments with restrictions in place to keep sensitive data secure. Discover and learn more about TREs with Hari Sood (The Alan Turing Institute) through a comprehensive overview as well as live demos, games and discussion.

How can Turing.jl boost scientific research, foster collaboration, and achieve social good? The University of Cambridge’s Hong Ge, Zoubin Ghahramani and Yongchao Huang present the Turing.jl: the Julia-based, high-performance probabilistic programming language.

Drop by to meet the team; understand what Turing.jl is, how it works and how we can benefit from its use. Experience the fast, flexible, modular and easy-to-use language in three real-world probabilistic modelling uses: demonstrating its applications in fraud-detection (digital economy), patients’ drug-tracking (health and life sciences), and exhibiting its novel Bayesian approach for Li-Ion battery fade modelling (sustainable environment).

The situation: you are in charge of climate policy. The challenge: can you make the right decisions to avoid disaster? Test your theories with the University of Exeter’s Uncertainty Quantification modelling apps. Peter Challenor and team present two interactive apps replicating the real methods behind crucial and intimidatingly large-scale policy making decisions, one modelling future global temperature increases and one aiming to reduce prohibitive costs in state-of-the-art green engineering. Meet the team behind the tech, test your climate-saving mettle and deep-dive into uncertainty quantification with the University of Exeter team.

Looking for deeper insights into your data? Raphtory is an analytics platform combining graph structures and time-series analysis, allowing you to look for time-respecting paths through your data, model community evolution, extract temporal patterns and much more. With potential for meaningful application across a huge range of research, visit the demo to meet the team (Pometry / The Alan Turing Institute / QMUL) for an exciting overview of the tech and a comprehensive understanding of how the platform works.

Does the increasing sophistication of AI present more of a risk or more of an opportunity when it comes to safeguarding children? Stephen Anning (Trilateral Research) and Lincolnshire Police have co-designed CESIUM, a ground-breaking ethical AI system that identifies and prioritises vulnerable children. The system – which pays particular attention to explainability – has already demonstrated success, having identified 16 vulnerable children months before they were referred through normal channels. AI has the potential to unlock otherwise-hidden insights than more general-purpose safeguarding tools: at AI UK, meet the team behind this tech, ask the big questions and discover proof of concept with discussion and case study walk-throughs.

What are the major pitfalls in data ethics? And how do we – as the UK’s leaders in data science, machine learning and AI – ensure we’re treading the path to ethical progress together? The oft-fraught area of data ethics is absolutely key to both progress and success. Discover Nina Di Cara and Natalie Zelenka’s (University of Bristol) Data Hazards system at AI UK 2023 and have your say in this crucial conversation.

The expert team behind the Turing Way are here at AI UK 2023 to meet you, at their pop-up collaboration café. Drop by for a chin wag over pertinent topics like open source, AI ethics, reproducibility or public and stakeholder engagement or pop in for one of several introductory sessions to different areas covered by The Turing Way and the newly funded Turing Practitioners Hub.

Led by experts from across the board and with an important breadth of know-how, the sessions will present concepts and practices from each guide, as well as practical sessions on how to use The Turing Way book, contribute to the project and reuse its open-source infrastructure.

Interested in learning more? Some of the research featured was brought to you in collaboration with The Alan Turing Institute’s AI for science and government fund and Engineering and Physical Sciences Research Council (EPSRC).

Launching a fascinating AI/human experiment, this cross-institutional team led by Matija Franklin (UCL) presents The Blame Game. Get seriously involved playing judge to fictional crimes with suspects and witnesses – both human and AI. Decide how you’ll apportion blame in this visual novel, interview witnesses, investigate sources, and witness your decisions impact the narrative. Inspired by the Moral Machine Experiment, this project aims to collect data on how people think about the actions of AI agents: be part of the discussion on the world of humans and machines at AI UK 2023.

The need for comprehensive understanding of vast quantities of data has never been more crucial. From disease spread and epidemiology education to cyber security, Daniel Archambault, Cagatay Turkay and Nick Holliman (of Swansea University, University of Warwick and King's College London, respectively) pose the question: is comprehensive data visualisation the answer to bridging the gap between human and computer to achieve actionable insights? Stop by to experience live demos of the team’s innovative AI-driven data visualisation technique.

Biomass is disappearing at a truly alarming rate. With Germany reporting a massive 75% decline in biomass, leading to substantial concerns of an “insect Armageddon”, The Centre for Ecology and Hydrology’s (UKCEH) Tom August invites you to set mind against machine for the ultimate solution. UKCEH present a new automated, non-lethal AI moth-monitoring tool; but can it outperform current data science methods? Try your hand at the game and catch up with the humans behind the machine at AI UK 2023. 

The Turing’s Data Study Groups (DSGs) bring together some of the country’s top talent from data science, AI and wider fields to analyse real-world data science challenges in intensive “collaborative hackathons”, which connect organisations from industry, government and the third sector with talented multi-disciplinary researchers from across a huge academic range. Join The Turing’s Jules Manser and team for an informal walk-through of how they scope DSG challenges – and how a DSG could transform your business problem – and take part in a DSG taster brainstorming session. Pop by the stand to register for a taster session, or email the team ahead of time at: [email protected]

The Alan Turing Institute is launching its very first online learning platform. Sarah Nietopski, Chloe Poon and Mishka Nemes (Turing Skills Team) invite you to be one of the first to experience this exciting new venture; visit the AI UK demo at any time across the two days to learn about the new suite of courses focused on responsible AI. With skills team members on hand to introduce the platform and guide you through one of two sample courses, discover the fresh beginnings of a stimulating new access point to artificial intelligence.

Discover how NATS (formerly National Air Traffic Services) is using the latest in AI technology to advance their decades of airspace simulation experience at this demo, headed by Kayleigh Ockelford (NATS). From the “Time to conflict” model – which proactively reduces reliance on humans and enhances safety – and the machine learning “Michelangelo” – which forecasts air traffic flows for detailed recommendations – to bespoke cutting-edge products such as the Demand Capacity Balancer and the Strategic-Airport Capacity Management, there is a wealth of know-how and fascinating tech use to explore.