Researchers focused on improving society’s resilience against AI risks such as deepfakes, misinformation, and cyber-attacks, can now access government grants to drive their work and help ensure the safety of AI.

Launching the formal opening of its Systemic Safety Grants Programme, the UK’s AI Safety Institute is looking to back around 20 projects with funding of up to £200,000 each over the course of its first phase, worth £4 million.

The scheme, launched today, will focus on how society can be protected from the potential risks of AI, and support research to tackle the threat of AI systems failing unexpectedly in critical areas, for example in the financial sector. 

The fund, first announced at May’s AI Seoul Summit, will eventually allocate £8.5 million, with the additional cash becoming available as further phases are launched. 

The first phase of the AI Safety Institute scheme, launched in partnership between the Engineering and Physical Sciences Research Council (EPSRC) and Innovate UK, is dedicated to developing ‘systemic AI safety’ to build public trust as the technology is rolled out in an effort to boost the economy.

The government hopes that tackling these risks head on will build public confidence in the technology, spark long-term growth and aid in the improvement of public services, all while keeping the UK at the cutting edge of research into responsible AI development.

The programme is intended to spark a broad range of research identifying the critical risks at the frontier of AI adoption in sectors like healthcare and energy services, identifying potential solutions which can be transformed into tools which tackle AI risks in specific areas.

The government has also committed to introducing highly-targeted legislation for the handful of companies developing the most powerful AI models, aiming for a proportionate approach to regulation rather than new blanket rules on its use.

The technology secretary, Peter Kyle, said: “My focus is on speeding up the adoption of AI across the country so that we can kickstart growth and improve public services. Central to that plan though is boosting public trust in the innovations which are already delivering real change.

“That’s where this grants programme comes in. By tapping into a wide range of expertise from industry to academia, we are supporting the research which will make sure that as we roll AI systems out across our economy, they can be safe and trustworthy at the point of delivery.”

The UK has made considerable efforts to remain at the forefront of ethical AI development, being among the first countries to launch a dedicated AI safety institution last year, followed up by the release of the open-source Inspect evaluation platform that allows testers to assess specific capabilities of individual AI models.

Recommended reading

• UK AI Safety Institute to Expand to San Francisco
• UK and South Korea Prepare for Next Global AI Safety Summit
• UK AI Safety Institute Releases Open Source Testing Platform

The country is also playing an increasingly important role in fostering responsible AI development internationally, too. Last month, the UK hosted an AI safety conference in San Francisco, following a similar summit in Bletchley Park and a subsequent event in Seoul last May. 

These events are aimed to provide an opportunity for AI developers, nation states, and technology and media companies to share ideas and insights to support the development of AI safety frameworks, with attendees urged to share their thoughts on potential areas of concern.

The UK, however, has neglected to implement formal legislation regulating the deployment and development of AI, and has yet to follow the footsteps of the EU which introduced a stringent set of regulations in its AI Act.