The UK’s AI Security Institute has launched a joint initiative with Amazon, Anthropic and its Canadian government equivalent to study AI alignment.
The Alignment Project, backed by £15m, will see the state-owned institute work with industry and international partners to determine how to ensure, as the technology becomes more advanced, that AI systems behave predictably and as designed.
The field of AI alignment research is essentially concerned with making sure the technology acts in the way it is supposed and will continue to do so in a predictable manner.
“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,” said Technology Secretary Peter Kyle.
“This is at the heart of the work the Institute has been leading since day one – safeguarding our national security and ensuring the British public are protected from the most serious risks AI could pose as the technology becomes more and more advanced.”
The AI Security Institute said that due to the rapidly developing capabilities of the technology, existing methods for controlling it will likely soon be insufficient.
“AI alignment is one of the most urgent and under-resourced challenges of our time. Progress is essential, but it’s not happening fast enough relative to the rapid pace of AI development. Misaligned, highly capable systems could act in ways beyond our ability to control, with profound global implications,” said Geoffrey Irving, chief scientist at the AI Security Institute.
“By providing funding, compute resources, and interdisciplinary collaboration to bring more ideas to bear on the problem, we hope to increase the chance that transformative AI systems serve humanity reliably, safely, and in ways we can trust.”
Other backers of the project include Halcyon Futures, the Safe AI Fund, UK Research and Innovation, Schmidt Sciences and the Advanced Research and Invention Agency.
Jack Clark, co-Founder and head of policy at Anthropic, added: “As AI systems become increasingly intelligent, it is urgent that we improve our understanding of how they work.
“Anthropic is delighted to work with the UK AI Security Institute and other partners on the Alignment Project, which will bring greater focus to these issues.”
Read more: UK government strikes deal with Anthropic to bring AI to public services
Register for Free
Bookmark your favorite posts, get daily updates, and enjoy an ad-reduced experience.
Already have an account? Log in
