We are looking to support Centres for Doctoral Training (CDTs) to deliver high quality, cohort-based doctoral training in the applications and implications of novel and existing AI technologies.
It is expected that all those trained through the CDTs should have a sufficient knowledge of AI and the chosen priority area such that at the end of their studies they will be able to develop and apply novel AI techniques within that area, discipline or sector.
CDTs should have an emphasis on a high value student experience promoting their long-term career development, wellbeing, and preparation to follow a diversity of career paths.
Applications are required to demonstrate the specific need for doctoral training through the CDT mechanism which includes the following key features:
- a clear need for doctoral level skills education in a specific area of focus
- the need for a cohort-based training approach
- the provision of both depth and breadth in the research training proposed to address the identified skills need
Co-creation between different disciplines and engagement with industry and users are strongly encouraged. Whilst high-quality proposals in any relevant area of UKRI’s remit are welcome, interdisciplinary proposals are particularly welcome.
CDTs should focus on AI applied and applicable in at least 1 of the priority areas. Centres may focus on more than 1 area. Centres should have a clear vision for the area or areas of AI training they will focus on, be able to articulate the national need for students in that area and explain the benefit of bringing the centre together in a multidisciplinary way.
Proposals must clearly identify which area or areas they will deliver against and the rationale for selection of the areas, including:
- potential for innovative research training agendas
- value added synergies
- evidence of capability to deliver relevant to these areas
The priority areas are outlined below.
Science and research
This priority area covers AI to transform research and discovery across all disciplines, enabling:
- novel hypotheses to be identified
- new questions to be explored
- advanced data-driven approaches to research
You should identify which fields of research the training within the CDT would be relevant to and where synergies may exist with other areas highlighted below, if relevant.
This priority area covers the development and deployment of AI in the understanding and management of health and disease. This includes the prevention, detection, and treatment of diseases through integration of multimodal data and across scales from molecular to populations. Using AI to support health, resilience and wellbeing through life is also within scope. Training may support the use of AI to generate:
- new cellular, molecular, mechanistic or causal insight
- understanding of emerging health threats
- accelerating development of improved health interventions
CDT training is expected to be collaborative and multidisciplinary, open to clinical trainees and other technical specialists. Training should be framed within the responsible delivery of AI in a human health context, and will need to consider complexity, equity, deployment, reproducibility, and fiduciary responsibilities.
Environment and energy
This priority area covers AI to advance our understanding of the natural world and to address critical global environmental and energy challenges. Using AI to harness our rich, complex and diverse environmental data will enable us to understand, predict, evaluate and mitigate the issues of our changing climate. Optimising energy systems with AI and using AI to develop new energy technologies will affect a long-term reduction in carbon emissions.
CDT training will deliver skilled researchers who can tackle interdisciplinary problems such as:
- weather and climate prediction
- modelling greenhouse gas emissions
- optimising use of energy and resources
- anticipating extreme events due to natural hazards
- understanding how behaviour change can address environmental challenges
- understanding uncertainty and risk in environmental models
Owing to the complexity of planetary and energy systems student training should also encompass systems approaches and end-user collaboration.
Sustainable agriculture and food
The agrifood sector must:
- adapt and build resilience to climate change
- decarbonise to reach net zero targets
- reverse biodiversity decline
- improve animal welfare
- combat disease threats
The sector must do this whilst also ensuring a sustainable supply of safe food and delivering positive nutritional, economic and environmental outcomes globally. CDT training will develop students with interdisciplinary AI expertise, including the ability to:
- harness diverse data to deepen our understanding of soils, crops, aquaculture, farmed animals, supply chains, and consumer demand
- generate novel insights across scales from lab to landscape
- generate solutions relevant to real-world needs from industry, policy and other stakeholders across food systems from farm to fork
Defence and security
AI is specifically cited in the UK government’s Integrated Review as a priority technology for development to ensure the ongoing security and defence of the UK.
Applications of AI which can augment better human decision-making at pace, better population responses to threats, automatically detect system vulnerabilities and self-patch software or enhance operational efficiency at scale have the potential to completely change defence and security.
CDTs will deliver students with expertise in the design and use of AI technologies in the defence and security sectors. This includes how AI systems will be used and affect decision making, and the ability to implement solutions ethically and responsibly. You should show how graduates will get the experience necessary to develop AI within the defence and security sectors.
The use and influence of AI in the creative industries is growing, impacting on the development of an increasing range of subsectors. This spans content creation, content consumption and analysis of creative outputs. The spread of AI is further accelerating through digital convergence between sectors, for example, the application of game development techniques in filmmaking.
CDT training will develop researchers with interdisciplinary skills for careers in the creative sector. Students will also develop deep understanding of significant ethical implications and questions of consumer trust, as well as regulatory and legal issues related to the use of AI in these contexts.
Responsible and trustworthy AI
The expanding capabilities and range of applications of AI necessitate new research into responsible approaches to AI that are secure, safe, reliable and that operate in a way we understand, trust, and can investigate if they fail. Ensuring the safe and ethical adoption of AI technologies is vital to ensure that they deliver societal and economic benefits.
CDT training in this area will equip students with interdisciplinary expertise in the development and deployment of responsible and trustworthy AI technologies, including technical knowledge and sociotechnical aspects such as fairness, bias and ethics.
This is a research-focused priority area.
All CDTs will be expected to provide relevant training in responsible research and innovation relating to AI.
In addition to the above priority areas applicants are encouraged to consider the following cross-cutting themes where appropriate. Applications involving cross-cutting themes should identify clear links with one or more of the priority areas, which sector or sectors the CDT will target, key industry or government partners, and the reciprocal benefits the partnerships will deliver.
AI for increasing business productivity
This theme addresses industry-oriented AI challenges, unlocking the potential of AI to boost innovation, competitiveness and economic activity. The capabilities of AI technology are mature enough for the economy to realise productivity gains, with adoption of AI being the last hurdle to make sure everyone benefits.
CDT training in this theme will ensure the UK has a new generation of researchers equipped with the skills needed by industry to facilitate responsible adoption and exploitation of AI.
Application of AI to government policy and public services
AI technologies have great potential to improve the effectiveness and efficiency of services to the public. By harnessing a range of data assets, it can also better enable data usage to inform policy and to prioritise what works, and for whom.
CDT training should result in students with an understanding of policy development and public services, who are also able to use AI techniques in this context, and who are familiar with datasets and challenges in their usage relevant to policy making.
Investing in CDTs in AI will train people across a spectrum, including:
- those with a background in AI wishing to apply their skills to a wide range of disciplines and challenges
- those who are from different disciplinary backgrounds, where AI could make a transformational contribution to that discipline or where that discipline could be brought to bear on the development of AI technology and approaches
CDTs should also consider the implications of AI into the intended domains, examining the legal, ethical and socio-economic consequences of potentially disruptive intelligent technologies before they are deployed.
Proposals focused on the mathematical and computational foundations of AI without a clear application to one of the priority areas should be submitted to the Engineering and Physical Sciences Research Council (EPSRC) CDT funding opportunity. Applicants who are unclear about which funding opportunity is best suited to the vision for their CDT should seek advice from UKRI.
UKRI reserves the right to move applications from this funding opportunity to the EPSRC CDT funding opportunity after the outline assessment, and vice versa, should an application better fit the scope of the other funding opportunity. Applicants who are unclear about which funding opportunity is best suited to the vision for their CDT should seek advice from UKRI.
Key features of CDTs
It is expected that the minimum cohort size will be 10, though exceptions may be made where they can be fully justified due to the nature of the training.
The doctoral education delivered by the CDTs should provide:
- the support for student cohorts on a 4-year doctorate or equivalent, with a critical mass of supervisors (around 20 to 40) of internationally recognised research excellence and having a track record of doctoral supervision
- a cohort approach to training through peer-to-peer learning both within and across cohorts. This cohort approach to training should be provided throughout the lifetime of student’s doctorate training programme
- opportunities for significant, challenging and original research projects leading to the award of a doctoral level degree in accordance with a university’s standard regulations
- doctoral projects that are designed in such a way that (barring exceptional circumstances) students are able to submit their thesis within their funded period
- a formal, assessable programme of taught coursework, which should develop and enhance, for example, technical interdisciplinary knowledge such as software and data skills. Courses should also prepare students for future careers, providing trainings in areas such as management, entrepreneurship, commercialisation, responsible innovation and environmental sustainability
- a significant commitment to and support for the training environment by the hosts and key partners including appropriate co-creation of the centre
- opportunities for all students to gain experience beyond their doctoral projects
- appropriate user and employer engagement in the research and training
- a diverse and inclusive research environment to support people in achieving world-class research and career development
- mechanisms by which students funded through other routes can benefit from the training experience offered by the centre, and for the centre to reach out to the broader research and user community
You should also consider the aspects listed in the enhanced training section below.
The design and management of CDTs should aim to support the graduation of students with research doctoral level qualifications. Centres can offer all students a PhD or professional doctoral award, for example EngD as appropriate to the individual student, the research project, and the benefit to their future career. Universities are free to choose the type of research doctoral qualification that is offered to students.
Centres may choose to offer all students the same type of qualification or a mixture. Some qualifications have their own requirements so centres must ensure that students are able to meet these criteria if these are to be offered. Centre bids will be assessed against the appropriateness of the training provision offered, not the choice of qualifications to be awarded.
Students must receive training in responsible innovation taking into account the wider implications of research and innovation.
Find out more about responsible innovation.
Students should gain an appreciation of social responsibility, the consideration of ethics and inclusive user engagement as part of designing and conducting research. We would expect students to receive training in the general topic of responsible innovation as well as in issues more specific to the scientific areas relevant to the centre and their project.
Impact and translation
CDTs will support students to maximise the impact of the research they undertake, by providing them with an understanding of how research projects can be designed to include considerations of impact from the start.
Depending on the nature of the CDT, impact training may cover knowledge exchange and maximising academic, environmental, societal and economic impacts from research.
Students should understand the research and innovation lifecycle in which they are participating, including translation of research and consideration of end-use. Where appropriate, training should develop people who are able to work with and across industry sectors, and who can foster new innovative approaches.
In some areas, training could provide understanding of intellectual property, entrepreneurship and commercialisation. Others may require understanding of regulatory and policy considerations.
UKRI encourages user engagement across the entirety of its doctoral training. The extent of that engagement varies according to the nature of the research and training and may also vary with the size of the company or user. We encourage all forms of user engagement and contributions where this is beneficial to the training provision. The appropriateness of the support offered will vary depending on both the area, sectors, and type of partner. This should be demonstrated and will be assessed based on the added value of the engagement, not its monetary value.
Wider training experience
Enabling UKRI sponsored research students to benefit from experience outside their home organisation can contribute to the wider training experience possible through a CDT. This can be in the form of, for example:
- industrial experience
- entrepreneurial training
- public engagement activities
- a period of time spent in an overseas academic collaborator’s laboratory
Facilities and research tools
To carry out AI research, researchers need to be able to access and use a wide range of equipment, facilities and e-infrastructure (software, digital research infrastructure and data).
CDT students will therefore need to be trained in how to use the essential tools for their research. Students should benefit from the environment and accessibility of infrastructure. Access to the necessary infrastructure is good evidence of the suitability of the bidding organisation or organisations as a host for the CDT.
If appropriate to their research, students should also have access to large facilities and national research facilities.
Find an EPSRC facility or resource.
It is not expected that centres create bespoke training courses in the use of essential research tools if access to existing courses is available. Funding for students to attend these courses should be included in applications.
UKRI expects applicants to liaise with the appropriate contacts throughout the development of their application to secure commitment from the facility or trainer. Centres requiring significant interactions with facilities should describe how they will ensure the students receive an excellent grounding in the experimental techniques for their research.
Computational and data-driven research
Students being trained through the UKRI AI CDTs will be using computational and data techniques in their projects, and some may have projects aimed specifically at software development.
It is essential that they are given appropriate training so that they can confidently undertake such research in a manner that represents high professional standards and good practice in software development, data management and ethical use of data. For example, to ensure reproducible research (this may need to include data protection and regulation).
Students should be trained in the principles of open data in accordance with the Concordat on Open Research Data. Centres should ensure research data gathered and generated by students is, wherever possible, made openly available for use by others.
Centres requiring students to undertake computational research should set out a programme of training, tailored to meet the needs of the centre students, and explain how this training will be provided. For students who are required to adapt, extend, or develop software as part of their research we expect them to receive training in relevant programming and software engineering skills, including working collaboratively on code, testing, automation, and revision control.
There is a significant amount of training available and centres should contact potential providers, as they may be able either to provide the training required, or to help with ‘training the trainers’ so that material can be delivered locally and at the most appropriate time.
A list of training courses is provided in the additional information guidance (PDF, 136KB).
Computational research training would be expected to include at least several of the following:
- fundamentals of computing
- basic data analysis and curation
- numerical analysis and algorithm development
- how to apply computational techniques and data analytics as research tools, in particular the design of experiments and the interpretation of results
- targeted training in applying and using the standard codes for the particular research area of the CDT
- matching problems with available and new hardware (desktop, cloud, high performance computing, graphics processing units) and scaling up beyond the desktop
Other research staff
UKRI recognises the importance of research software engineers and support staff in the development and deployment of AI technologies and ensuring that AI is developed in a way which is open and provides broad access to the technology in the public good.
As such, you should consider how you will embed the principle of software sustainability into training and research projects. In addition, you should consider where training packages may be available in the CDT which will support associated and aligned research software engineers, technicians and other support staff, or elsewhere in the academic or user training environment.
Up to £117 million is available for this opportunity, subject to approval of a business case. Once indexation is applied to successful awards, we expect to support 10 to 15 CDTs. UKRI will fund 100% of the full economic cost.
Estates and indirect costs will not be funded on these awards.
The allocation of funding to UKRI is subject to business case approval by the Department for Business, Energy, and Industrial Strategy (BEIS) and HM Treasury. Applicants should proceed on the understanding that UKRI’s ability to fund CDTs recommended for funding through this opportunity will be dependent on that approval being secured.
The duration of award must be 102 months, with a start date between 1 April 2024 to 1 October 2024.
Costs that may be requested from UKRI
It is strongly expected that each centre supports a minimum of 50 students over 5 cohorts. Smaller cohorts may exceptionally be permitted where a strong rationale can be provided.
Studentship costs (fees, stipends and appropriate research training support). It is strongly expected that additional support will be provided from non-UKRI sources to contribute to these costs. UKRI funding may be used flexibly but must support students at a minimum of 50% of their studentship costs. The Research Training Support Grant (RTSG) covers items such as travel and consumables.
Centre delivery, coordination (including between a centre and other parties if justified) and management staff costs can be requested. Costs associated with student supervision may not be included.
Tuition fees and stipend above the minimum rates published by UKRI may be requested.
Get a studentship to fund your doctorate.
However, UKRI will not cover additional college fees. Fees cannot be higher than the fee charged by the university for UK non-research council funded students on similar programmes. Any stipend enhancement should be fully justified in the context of the area of training and UK skills need.
Start up costs should only be included where necessary and should not duplicate existing provision such as where existing centres already have necessary infrastructure in place.
All costs (including stipends and fees) requested in applications should be calculated at current rates with no addition made to consider inflation over the length of the funding period. UKRI will include this indexation at the final funding stage.
Costs should not be included to support students outside the CDT cohorts already supported by funding from other sources. Where a central cost is incurred by the CDT (for example in developing a new training course principally for the CDT students) these ‘aligned ‘students may benefit from these.
Additional support or leverage
In recognition of the diversity of potential partners across UKRI’s remit, no minimum leverage requirement has been set for this funding opportunity. However, both cash and in-kind support from non-UKRI sources is strongly expected.
HM Treasury have set an expectation of UKRI achieving significant leverage for this investment, and that universities and their partners will work together to give the best outputs for the UK. As such, all CDTs will be expected to contribute to this to an extent that is appropriate for the scope of the centre. UKRI will work with CDTs to meet leverage expectations over the lifetime of the awards.
Typically, it is expected that leverage of UKRI funds will be achieved through support from the applying organisations or project partners. You may use additional support flexibly to contribute to studentship costs (stipends, fees and RTSG). For example, dedicating full support for some studentships each year, or spreading funding to partially support all the students.
An estimate of the total cash support for each CDT is requested at the outline stage. The purpose is to have a view of the overall cost of a CDT. This financial information will not be seen by the panel but is purely for information for UKRI. Please indicate:
- funding being requested by the centre from UKRI
- institutional funding secured
- additional funding secured from project partners
Studentship costs consist of 3 elements:
- appropriate RTSG
If you are using the UKRI published rates then you should use the 2022 to 2023 rates without any allowance for inflation over the lifetime of the grant.
For more information on costs, read the additional information guidance (PDF, 136KB).
Equipment over £10,000 in value (including VAT) is not available through this funding opportunity. At the full proposal stage, smaller items of equipment (individually under £10,000) should be included in the ‘directly incurred – other costs’ heading.
Where possible researchers are asked to make use of existing facilities and equipment, including those hosted at other organisations.
Investigators and supervision
The investigators named on the Joint Electronic Submission (Je-S) system application form should represent the core management team of the centre. We would generally expect no more than 10 investigators to be named.
A strong justification will need to be provided for a larger core management team. Any requested funding for investigator time should reflect commitments to centre delivery and should not include individual student supervision related to research projects.
In order to maintain a cohort of this size, it is necessary to have access to a suitable pool of potential supervisors. Experience of current centres demonstrates a need for 20 to 40 supervisors, the majority of whom should have internationally recognised research excellence and a track record of doctoral supervision.
Applications will need to provide evidence of a suitable pool of potential supervisors, taking into account the interdisciplinary focus of the CDT and wider considerations such as equality, diversity and inclusion (EDI). You should not record supervisors on the Je-S application form.
Due to the scale of these awards, significant collaboration and leverage (cash or in-kind) will be expected from project partners (for example, business, public sector, third sector). This may include models such as funding studentships, industrial placements, co-created workshops.
We expect collaborations to build a mutually beneficial two-way relationship based on:
- career development opportunities for students
- increased depth of understanding of sector by students
- regional strengths
To ensure the awards are inclusive of a variety of approaches and research fields, no specific leverage expectations are being set for this funding opportunity.
Clear plans for engaging with new and existing collaborators over the duration of the CDT should be detailed in the case for support.
Involvement of The Alan Turing Institute
As the UK’s national institute for AI and data science The Alan Turing Institute is well positioned to engage with the UKRI AI CDTs. They will be taking a neutral stance towards all applicants as they intend to work openly and proactively with all successful UKRI AI CDTs. This means they will not be offering specific support to individual centres, for example acting as project partners on any UKRI AI CDT application. They will not offer letters of support to any proposed centres.
Successful CDTs will be brought together after awards are made to discuss opportunities to engage with The Alan Turing Institute but engagement will not be mandated.
Find out more about The Alan Turing Institute.
We welcome applications which include elements of international engagement where they add value to the proposed centre. Support requested might include travel, subsistence and consumable costs for UK-based students undertaking training or research visits to overseas centres of excellence, or for leading researchers to visit the UK to contribute to the students’ training experience.
Where a formal, joint training partnership is proposed, the UK component must be able to stand on its own merits. Students registered at international institutions will not count towards the minimum cohorts.
Applicants planning to include international collaborators on their proposal should visit trusted research and innovation for guidance on getting the most out of international collaboration whilst protecting intellectual property, sensitive research and personal information. Centres will be expected to engage with the relevant regulatory bodies where concerns may arise under the National Security and Investment Act.
Find out more about UKRI’s work around international engagement and partnerships.
The UKRI-RCN Money Follow Cooperation Agreement does not apply to this funding opportunity. As such CDT grants cannot include a Norway-based co-Investigator.
Equality, diversity and inclusion
UKRI aims to support a diverse and inclusive research environment where there is equal access to opportunities.
We are committed to supporting through our investments a diverse range of flexible approaches to ensure we support the diverse needs, backgrounds and potential careers of doctoral students as well as the requirements of the research and innovation communities.
It is therefore a requirement of all UKRI AI CDTs that EDI best practices are embedded into all aspects of the of the doctoral recruitment and training process throughout the lifetime of a training grant including:
- improved access and diversity of entry points to doctorial education
- project design, advertisement and applicant support
- applicant shortlisting and interviews
- management, training, supervision and tailored student support
- monitoring and evaluation
Read the EDI expectations guide to help you identify and overcome local barriers and to be used alongside other toolkits provided by organisations and your local institution.
Applications that support job shares, part time contracts and flexible working arrangements for CDT staff, academics and students are welcomed. Part-time students must undertake study for a minimum of 50% full time equivalent.
Successful CDTs will be expected to work together as a cohort to share best practice, maximise the value of the investment, and engage with other key actors in the UK AI landscape. This may include subsets of the CDTs where appropriate.
UKRI’s environmental sustainability strategy lays out our ambition to actively lead environmental sustainability across all our investments. CDTs must also seek opportunities to influence others and leave a legacy of environmental sustainability within the broader operations of their academic and industry partners.