RAi UK has the ambition to deliver world-leading best practices for the design, regulation and operation of ‘socially beneficial’ responsible artificial intelligence developments which are both trustworthy in principle, and trusted in practice by individuals, society and government. In so doing, we are committing to the following guiding principles. These will guide everything the ecosystem does, including its internal management, and the research, engagement, and skills programmes.
1. Convening, not competing
As a funder, we bring together research, industry and civil society organisations to share ideas, research and perspectives. We will also amplify ongoing research, identify gaps, and fund new initiatives.
We are committed to doing this in an open, inclusive and transparent way, adhering to the principles of open science. By doing so, we will strengthen the ecosystem of people who develop, deploy and use AI technologies in ways that are trustworthy, secure and benefit everyday life – in the UK and beyond.
2. Multidisciplinary by design
Maximising the benefits of AI will require a range of perspectives, specialisms and experiences about it will entail to use these technologies in different contexts. The research we fund and support will therefore make a virtue of inclusion and multidisciplinarity.
When we say “multidisciplinary”, we mean Arts and Social Sciences and Computer Scientists and Engineers. All working on wicked problems together, while listening to people outside academia.
Beyond this, we strive to be an ecosystem where people feel heard and seen. Even — or perhaps especially — when those voices come from outside the academy, from early career researchers, or when they challenge long held assumptions about what makes AI trustworthy and responsible.
3. Committed to excellence
Our inclusivity and multidisciplinarity is directly related to our commitment to excellence. Many of the risks and opportunities of AI stem from the diverse contexts in which they are developed and deployed.
As an internationally respected leadership team, we see it as our job to enable excellent research across our complex, overlapping specialisms. The research and recommendations from RAi UK will therefore be characterised by rigorous research about how to mitigate risks and take advantage of the opportunities that AI technologies present.
4. Fail better — responsibly
We know we won’t get everything right the first time. But we all know the shortcomings of moving fast and breaking things. Therefore, we will do everything we can to get things right, and iterate quickly when we don’t.
Everything we fund and do therefore includes an emphasis on Responsible Research and Innovation (RRI), which includes anticipating unintended consequences from the outset.
5. Fair and transparent
We are an organisation with a public purpose. Most of RAi UK’s budget and time goes to funding, a process that creates winners and losers. We will assess responses to our calls fairly, provide feedback, and explain our decisions when asked to do so.
We will also be open about how we are funded, what we do and how we work. This in turn will allow us to build on our reputations as trusted voices in how to develop, deploy and use AI – responsibly.
6. Committed to public benefit
We are an organisation with a public purpose. Funded by the UK government, we work as good stewards of public funds, answering to our stakeholders at UK Research and Innovation.
Beyond this commitment, we are all motivated to do mission-driven work. As academics and citizens, we believe in the importance of advancing research, training the next generation, and building technologies that benefit society.
In turn, therefore, our work assumes that responsibility should be at the root of how all AI technologies are developed, deployed and used.