UK’s AI Safety Institute sets standards, not just tests AI.

TLDR: The UK should focus on setting global standards for artificial intelligence (AI) testing rather than attempting to conduct all vetting itself, according to Marc Warner, CEO of London-based Faculty AI. The newly established AI Safety Institute, led by the UK government, could end up responsible for scrutinising a range of AI models due to the country’s leading work in AI safety. Warner believes the institute should be a global leader in setting testing standards and should implement measures such as “red teaming,” where specialists simulate misuse of an AI model. By setting robust standards, other governments and companies can follow suit and independently test models to ensure safety.