The UK’s AI Security Institute has launched a new international initiative to strengthen the safety and security of artificial intelligence systems. Backed by over £15 million in funding, the AI Alignment Project brings together a global coalition of research institutions, technology companies, and civil society to accelerate progress on one of the most critical technical challenges in AI development: ensuring AI systems behave as intended.
A global partnership to tackle AI alignment
The coalition includes: the Canadian AI Safety Institute, Canadian Institute for Advanced Research (CIFAR), Amazon Web Services (AWS), Anthropic, Schmidt Sciences, Halcyon Futures, Safe AI Fund, UK Research and Innovation, and the Advanced Research and Invention Agency (ARIA).
The collaboration reflects growing international recognition of the urgent need for AI alignment research, focused on making advanced AI systems reliable, predictable, and responsive to human oversight.
The Goal: make AI systems behave as designed
AI alignment research aims to ensure that as AI systems become more capable, they continue to act in accordance with human goals and values. Misaligned AI behaviours, even if unintentional, could present serious risks to safety, national security, and public trust.
The Alignment Project is designed to address these challenges by supporting research and development in three key areas:
- Grant funding: Up to £1 million for researchers across disciplines, including computer science, ethics, and cognitive science.
- Cloud compute access: Up to £5 million in cloud computing credits from AWS, enabling experiments at scale.
- Venture capital: Private investment to support commercial solutions focused on alignment technologies.
These resources aim to overcome historical barriers to progress in the field, such as limited access to computational infrastructure and fragmented research efforts.
Expert guidance and international momentum
The project is guided by a high-level advisory board featuring global AI leaders including Shafi Goldwasser, Yoshua Bengio, Zico Kolter, and Andrea Lincoln. Their expertise will help shape the project’s research direction and ensure its work aligns with global best practices.
The launch comes at a time when the 2025 International AI Safety Report has highlighted the rapid improvement of AI systems, with some demonstrating capabilities equivalent to PhD-level performance. Current control methods may soon fall short as AI continues to evolve, making international collaboration on AI safety more critical than ever.
Strengthening national security and driving innovation
According to Science, Innovation and Technology Secretary Peter Kyle, the initiative reflects the UK’s commitment to safe, secure, and trustworthy AI development. “Advanced AI systems are already exceeding human performance in some areas, so it’s crucial we’re driving forward research to ensure this transformative technology is behaving in our interests,” he said.
By addressing the technical challenges of AI alignment, the project supports the UK government’s Plan for Change—a strategy aimed at delivering economic growth, high-skilled jobs, and improved public services. At the same time, it reinforces the UK’s position as a global leader in responsible AI development.
A call for global participation
The AI Security Institute is inviting additional partners—governments, philanthropies, research labs, and private investors—to join the Alignment Project. Contributions can take the form of research grants, cloud computing support, or investment in alignment-focused startups.
This multi-sector approach is expected to accelerate the development of tools and frameworks that will allow AI to be deployed safely and at scale.
AI alignment and the future of trustworthy AI
As AI systems become more complex and autonomous, alignment is increasingly seen as foundational to ensuring public trust and broad adoption. By building infrastructure and incentives to advance this field, the UK’s AI Security Institute aims to ensure the technology delivers long-term benefits while safeguarding national and international security.
Read next: Construction’s digital transformation in the age of AI – Build in Digital
Are you a building professional? Sign up for a FREE MEMBERSHIP to upload news stories, post job vacancies, and connect with colleagues on our secure social feed.
