The new proposal, published in the journal AI and Ethics, is the result of work by Dr Kirti Prakash and Dr Ian van der Linde from the Advanced Computing Research Centre at Anglia Ruskin University (ARU). It aims to foster competition and innovation in frontier AI development while implementing safeguards against AI capabilities that could cause severe or irreversible harm.
Drawing lessons from the governance of powerful technologies such as nuclear arms control and CRISPR, the framework identifies key ethical problems. The researchers argue that merely slowing AI development is an 'incomplete' objective, as AI capabilities are globally distributed and can be replicated. They advocate for proportionate safeguards, meaningful human control, independent assessment, clear responsibility, and regular reassessment as AI evolves.
The 'Four As' framework specifically addresses: whether humans can effectively constrain consequential AI action (Action); who should have access to potentially harmful AI capabilities (Access); which consequential judgements can be delegated to AI (Authority); and how AI-related benefits, burdens, and power should be distributed (Allocation). The framework suggests that developers primarily implement Action safeguards, regulators govern Access, deploying institutions manage Authority, and governments and public bodies handle Allocation.
The researchers highlight that interventions can have unintended consequences, such as restricting access to a powerful model potentially increasing misuse risk while concentrating power among a few firms, thereby creating a new allocation concern. Dr Kirti Prakash, Senior Lecturer in AI and Applications at ARU, stated, “The debate around frontier AI is often framed as a choice between accelerating development and slowing it down. We argue that this is too blunt. Before deciding how strongly to intervene, we need to ask what exactly the ethical problem is.”
Dr Prakash further explained, “Effective AI governance will require more than a checklist of general ethical principles. The Four As provide a way of moving from identifying an ethical problem to asking what constraint is required, who should enforce it, what would signal that the safeguard has failed and what corrective action should follow.” She cited the UK's new AI Biosecurity Lab as an example of governance focusing on who can access and deploy advanced models, rather than just their existence.
This framework is released amidst growing international concern about the rapid development of frontier AI. A recent study from the Cambridge Programme on AI Science & Policy and the Centre for the Governance of AI explored the potential for AI to accelerate its own research, leading to an 'intelligence explosion'. Dr Prakash concluded, “The possibility of AI increasingly contributing to the development of its own successors makes the governance question even more urgent.”
