We welcome the opportunity to respond to the Future of Compute Review’s call for evidence. Our response focuses on the future of compute used for Artificial Intelligence (AI). In particular, we emphasise the risks posed by advanced AI systems that are enabled by access to large amounts of compute.
We recommend that the Future of Compute Review:
- Considers how best to support AI research in academia
- Encourages the prevention of misuse of AI capabilities while still preserving safe access, via structured access to ML systems
- Aims to enable research to scrutinise and understand high-compute and impactful models by researchers outside AI labs
- Considers a tiered access approach to compute provision, where access to larger amounts of compute comes with additional review requirements, if the Review recommends the creation of a compute fund
- Recommends cooperating with allies in securing and stabilising the semiconductor supply chain