TLDR: The UK’s AI Security Institute has launched a £15 million international coalition, ‘The Alignment Project,’ to accelerate research into AI alignment and safety. This initiative, backed by major industry players and research institutions, aims to ensure AI systems are predictable, controllable, and beneficial, addressing critical risks as AI capabilities rapidly advance.
The UK’s AI Security Institute (formerly AI Safety Institute) has spearheaded the formation of a significant international coalition, ‘The Alignment Project,’ dedicating over £15 million (approximately $20 million USD) to safeguard the development of artificial intelligence. Launched on July 30, 2025, this initiative seeks to enhance AI safety and alignment, ensuring that advanced AI systems behave as intended and remain under human control.
The coalition includes prominent organizations such as the Canadian AI Safety Institute, Amazon Web Services (AWS), Anthropic, CIFAR, Schmidt Sciences, Halcyon Futures, the Safe AI Fund, UK Research and Innovation (UKRI), and the Advanced Research and Invention Agency (ARIA). The primary objective is to fund and accelerate cutting-edge research into AI alignment, focusing on making AI systems more predictable, transparent, and responsive to human oversight.
Peter Kyle, the UK’s Science, Innovation and Technology Secretary, emphasized the urgency of this endeavor, stating, ‘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.’ Geoffrey Irving, Chief Scientist at the AI Security Institute, further highlighted the challenge, noting, ‘AI alignment is one of the most urgent and under-resourced challenges of our time. … Misaligned, highly capable systems could act in ways beyond our ability to control, with profound global implications.’
‘The Alignment Project’ operates through a three-pronged approach to remove barriers to alignment research:
Grant Funding: Up to £1 million per project will be available for researchers across various disciplines, from computer sciences to cognitive science.
Compute Access: £5 million in dedicated cloud computing credits, primarily from AWS, will enable large-scale technical experiments beyond typical academic reach.
Venture Capital: Investment support from private funders will accelerate the development of commercial alignment solutions.
The project’s advisory board boasts world-class experts, including Turing Award winners Yoshua Bengio and Shafi Goldwasser, and OpenAI board member Zico Kolter. This initiative builds on previous international collaborations, such as the agreement at the Seoul AI Summit in May 2024 to establish a network of AI safety institutes, and a strategic agreement signed with OpenAI in July 2025 to boost AI security and infrastructure collaboration.
Also Read:
- Understanding Sovereign AI: A National Imperative for Security and Innovation
- AI Pioneer Geoffrey Hinton Criticizes Tech Giants for Downplaying Risks
This ambitious undertaking positions the UK as a leader in responsible AI development, fostering collaboration between government entities, industry leaders, academia, and startups worldwide. By addressing the critical issue of AI alignment, the project aims to unlock the benefits of advanced AI while safeguarding national security and ensuring public protection from potential risks, ultimately making AI more reliable and trustworthy for deployment across various sectors, including financial services and public services.
