(written with PPI panel) Many aspects of a young person's life can affect their mental health(MH), and there is a crisis in our ability to support childhood mental illness. Problems often have to become serious before young people can access Child & Adolescent Mental Health Services(CAMHS). CAMHS are stretched, offering help to only a quarter of those in need, and often intervene late. Early identification and treatment are beneficial, but could swamp services and create even longer waits. Some young people are reluctant to access CAMHS because of stigma (e.g. self-harm). Inequity also limits access (e.g. those experiencing economic hardship or from minority groups). These variations leave many struggling to get help, affecting their health lifelong and their and their families' lives. We need to re-think how CAMHS are delivered. Using digital tools to make CAMHS fairer and more efficient could help young people get the right treatment sooner. For example, apps or websites could be used to: (1) identify problems early before someone needs intensive treatments, (2) signpost young people to the most useful services for them rather than sending everyone to CAMHS, or (3) help predict who would benefit most from which treatments, so young people get the right treatment first time. This could be achieved by harnessing the power of 'big data'. Information (data) about a young person's life could help. For example, the risk of serious problems is indicated by an accumulation of factors such as early childhood experiences (e.g. bullying, neglect, racism), the environment (e.g. housing, diet, the amount of green space near home) or physical factors (e.g. genetics, inflammation, brain chemistry). Data like these are already collected from a range of sources such as maternity, health visitors, GP records, schools and social care, but are never brought together. This information, if brought together, could be used to create digital tools to identify patterns using artificial intelligence (AI). However, there are problems to solve first. We do not know which data are most useful, how best to bring data together securely, or the most effective AI methods. Importantly, we have not got agreement on which information should be used for which purposes. For example, it might be acceptable to use genetic information in a hospital to decide which medication is safest, but maybe not to identify who is at risk of suffering from a problem in the community. We must get this right. In this study, we will access data from a broad range of sources, some of which we will collect and organise in the early stage of this project, and use it to establish the best way to develop digital tools to support CAMHS. We will then work with the public, and experts who work with or have experience of MH problems, to translate AI algorithms into digital tools. These digital tools must be part of a clinical service that can intervene early. We want to create a new early identification and prevention service and establish what digital tools are needed to make early detection work effectively, safely, and fairly. We will bring together experts who are doing ground-breaking work in academia, industry, and the clinic, with policy makers. We want to turn their attention to solving these problems, together with young people, their carers, and people with lived experience. The people whose data is used should direct the building of these tools and new clinical pathways. We need their help thinking about which data should be used for what purposes, for which people, what should happen when a young person is thought to be developing MH problems, and how to use digital tools to support treatment decisions. In later years we will explore the effectiveness of the early identification and prevention approach, create recommendations for overhauling inefficient systems and develop a template for data-guided, individualised, and timely MH interventions for the future.

The ambition of this partnership between NATS and The Alan Turing Institute is to develop the fundamental science to deliver the world's first AI system to control a section of airspace in live trials. Our research will take a hierarchical approach to air traffic control (ATC) by developing a digital twin alongside a multi-agent machine-learning control system for UK airspace. Furthermore, the partnership will develop technical approaches to deploy trustworthy AI systems, considering how safety, explainability and ethics are embedded within our methods, so that we can deliver new tools which work in harmony with human air traffic controllers in a safety-critical environment. Little has changed in the fundamental infrastructure of UK airspace in the past 50 years, but demand for aviation has increased a hundredfold. Aviation 2050, a recent government green paper, underlines the importance of the aviation network to the prosperity of the UK to the value of £22 billion annually. Yet our nation is at risk without rapid action to modernise our airspace and control methods, to ensure they can handle a future increase in UK passenger traffic of over 50% by 2050 and new challenges arising from unmanned aircraft, both against a backdrop of increasing global pressures to transform the sector's environmental impact. The augmentation of live air traffic control through the use of AI agents which can handle the complexity and uncertainties in the system has transformative potential for NATS's business. This will positively impact live operations, as well as a research tool and training facility for new ATCOs. Correspondingly, NATS's research vision is to exploit new approaches to AI that enable increases in safety, capacity and environmental sustainability while streamlining air traffic controller training. The anticipated benefits of AI systems to air traffic control have come at a critical time, providing us with an opportunity to respond effectively to the unprecedented challenges which arise from a triad of crises: the coronavirus 2019 (Covid-19) pandemic, Brexit and global warming. The UK must develop independent technical advances in the sector, without compromising sustainability targets. The Alan Turing Institute is positioned at the rapidly evolving frontiers of probabilistic machine learning, safe and trustworthy AI and reproducible software engineering. Matching this with the world-leading expertise of NATS, supported by a world-first data set of more than 20 million flight records, means that this partnership is in a unique position to build the first multi AI agents system to deliver tactical control of UK airspace.

There is an unprecedented integration of AI assistants into everyday life, from the personal AI assistants running in our smart phones and homes, to enterprise AI assistants for increased productivity at the workplace, to health AI assistants. Only in the UK, 7M users interact with AI assistants every day, and 13M on a weekly basis. A crucial issue is how secure AI assistants are, as they make extensive use of AI and learn continually. Also, AI assistants are complex systems with different AI models interacting with each other and with the various stakeholders and the wider ecosystem in which AI assistants are embedded. This ranges from adversarial settings, where malicious actors exploit vulnerabilities that arise from the use of AI models to make AI assistants behave in an insecure way, to accidental ones, where negligent actors introduce security issues or use AIS insecurely. Beyond the technical complexities, users of AI assistants are known to have mental models that are highly incomplete and they do not know how to protect themselves. SAIS (Secure AI assistantS) is a cross-disciplinary collaboration between the Departments of Informatics, Digital Humanities and The Policy Institute at King's College London, and the Department of Computing at Imperial College London, working with non-academic partners: Microsoft, Humley, Hospify, Mycroft, policy and regulation experts, and the general public, including non-technical users. SAIS will provide an understanding of attacks on AIS considering the whole AIS ecosystem, the AI models used in them, and all the stakeholders involved, particularly focusing on the feasibility and severity of potential attacks on AIS from a strategic threat and risk approach. Based on this understanding, SAIS will propose methods to specify, verify and monitor the security behaviour of AIS based on model- based AI techniques known to provide richer foundations than data-driven ones for explanations on the behaviour of AI-based systems. This will result in a multifaceted approach, including: a) novel specification and verification techniques for AIS, such as methods to verify the machine learning models used by AIS; b) novel methods to dynamically reason about the expected behaviour of AIS to be able to audit and detect any degradation or deviation from that expected behaviour based on normative systems and data provenance; iii) co-created security explanations following a techno-cultural method to increase users' literacy of AIS security in a way that users can comprehend.

The last decade has seen a proliferation of social data that betrays human behaviour--from a person's clicks to a group's language to a state's signal data. Concurrently there has been rapid developments in our ability to analyse these data, to identify patterns that enable inferences not conceivable before. These advances open exciting possibilities for understanding human behaviour at the individual, group and population level. But they are also a source of insecurity and vulnerability, challenging the meaning of privacy and exposing individuals to manipulation. NABS+ will deliver 'next generation behavioural science' that addresses these opportunities and threats in the context of security and defence. It embraces new forms of data and analytics for behavioural science, while acknowledging the need to simultaneously address ethical and societal consequences. It will promote a generation of 'analytical behavioural scientists' who are equipped to respond to existential and acute threats, can understand and navigate policy dilemmas, and are adept at working with diverse data. NABS+ will be home to a vibrant community of researchers and end-users who recognise the value of combining social scientific theory and novel data analytics. Our vision is to stimulate an interdisciplinary community that is: (1) focused on identifying, prioritising and developing knowledge in analytical behavioural science through a theory-led, data-driven lens; (2) structured to be agile and responsive to international security trends and stakeholder needs, recognising that diverse data and methods are needed for different challenges; and (3) driven to create a community that is home to an emerging generation of data literate, behavioural scientists. NABS+ will deliver four pillars of activity. The Community pillar will deliver activities that shape new forms of public-private partnerships, connecting academic, industry, and government stakeholders to deliver challenge-driven collaborations. NABS+ is premised on the value of bringing together behavioural and data scientists. It thus invests in outputs that break down traditional groupings and increasing awareness (e.g., white papers, website). The Research pillar will co-develop and deliver interdisciplinary research that responds to immediate needs and prepares for future threats. The work may be an event, a synthetic review, or novel research, and we strive for a diversity of data and methodological approaches. In all cases, however, projects will be theory-led, data-driven activities that consider ethical and society implications at all stages. The Disseminate pillar recognises that progress in analytical behavioural science will depend on researchers understanding the activities and concepts of others. The pillar will deliver world-class activities that 'translate' knowledge so that understanding spreads beyond specialist audiences. As part of this, NABS+ will support researchers inform policy and practice, utilising unique Research to Practice Fellow support. The Build pillar seeks to ensure growth in the NABS+ community so that it becomes home to a next generation of behavioural scientist. The pillar involves activities that foster the skills and leadership of members through training and by providing secondment opportunities (e.g., academic researchers into industry). As part of this, NABS+ will implement a programme of actions designed to progress the network toward being self-supporting. The cumulative impact of these activities will be a world-class analytical behavioural science capability; a vibrant community with shared understanding, common tools, and a programme of ongoing research that shapes understanding and contributes profoundly to policy and practice.

The proposed programme of research will establish the machine learning foundations and artificial intelligence methodologies for Digital Twins. Digital Twins are digital representations of real-world physical phenomena and assets, that are coupled with the corresponding physical twin through instrumentation and live data and information flows. This research programme will establish next-generation Digital Twins that will enable decision makers to perform accurate but simulated "what-if" scenarios in order to better understand the real world phenomena and improve overall decision making and outcomes.