The US And UK Establish New Partnerships In Artificial Intelligence Security

US Trade Minister Gina Raimondo and UK Technology Minister Michelle Donelan (right) (photo: x @michelledonelan)

JAKARTA - The US and Britain on Monday April 1, announced a new partnership in artificial intelligence (AI) security science, amid growing concerns about the upcoming version of the next generation.

US Commerce Secretary Gina Raimondo and UK Technology Minister Michelle Donelan signed a memorandum of understanding in Washington to develop a joint follow-up AI model testing following a commitment announced at the AI Safety Summit at Bletchley Park in November.

"We all know AI is the determining technology of our generation," Raimondo said. This partnership will accelerate the work of our institutions across the spectrum to address the risks of our national security concerns and the concerns of our wider society.

Britain and the US are among the countries that established the government-led AI security institute. Britain said in October that its institute would examine and test new AI types, while the US said in November that it was launching its own security institute to evaluate the risks of the border AI model and is now working with 200 companies and entities.

In formal partnerships, Britain and the US plan to conduct at least one joint testing exercise on publicly accessible models and are considering exploring personnel exchanges between institutes. The two countries are working to develop similar partnerships with other countries to promote AI security.

"This is the first agreement of its kind anywhere in the world," Donelan said. "AI has become an extraordinary force for good in our society, and has great potential to overcome some of the world's biggest challenges, but only if we are able to overcome these risks."

Generative AIs - which can create text, photos, and videos in response to open clues - have sparked the excitement and fear that it could leave some work obsolete, disrupting elections, and potentially controlling humans and catastrophic effects.

Raimondo and Donelan urged the joint action needed to address AI risks. "Time is very important as the next set of models will be released soon, which will be much more capable," Donelan said. We have a focus on the area we share and share and really specialize.

Raimondo said he would discuss AI issues at a meeting of the US-EU Council on Thursday, April 4. President Joe Biden's government plans to announce the addition of its AI team soon. "We are withdrawing the full resources of the US government," Raimondo said.

The two countries plan to share key information about the capabilities and risks associated with AI models and systems as well as technical research on AI security and security. In October, Biden signed an executive order aimed at reducing AI risk.

In January, the Commerce Department said it proposed requiring US cloud companies to determine whether foreign entities access US data centers to train AI models.

Britain said in February that it would spend more than 100 million pounds (Rp2 trillion) to launch nine new research centers and train AI regulators on the technology. Raimondo said he was deeply concerned about the AI threat applied to bioterrorism or nuclear war simulations.

"Those are things where the consequences could be catastrophic and so we really have to have zero tolerance for some of these models used for those abilities," he said.