US government announces 'most significant actions ever taken' on AI safety

President Joe Biden speaks about artificial intelligence in the Roosevelt Room of the White House in July 2023. Credit: AP

The White House has announced a new landmark directive on AI safety, making it compulsory for developers to share the results of safety tests with the government.

The executive order will establish new industry standards for security and privacy, and promises to protect against the risks of using artificial intelligence to create "dangerous biological materials".

White House Deputy Chief of Staff Bruce Reed has called it "the most significant actions ever taken by any government to advance the field of AI safety".

As well as keeping Americans safe, the directive will aim to stop AI from deepening discrimination and ensure the technology is used fairly in the justice system.

White House chief of staff Jeff Zients said Biden had told them to move urgently on the issue.

“We can’t move at a normal government pace,” Mr Zients said the president told him. “We have to move as fast, if not faster than the technology itself.”

The order will also create guidance to label and watermark AI-generated content to help people tell the difference between authentic interactions and those generated by software. 

The White House is also looking to increase its AI workforce, with a new website advertising job openings to "build and govern" artificial intelligence in line with the White House's priorities.

This comes as the UK government is set to hold a summit on the dangers of artificial intelligence at Bletchley Park this Wednesday.

The meeting, which Prime Minister Rishi Sunak and US Vice President Kamala Harris plan to attend, will address the risks of the technology and how international coordination can reduce them.

Joe Biden has previously stressed harnessing the potential and preventing the dangers of AI as a priority, meeting with a group of industry leaders in June.

In July, the White House secured an agreement from several tech companies, including Amazon, Google, Meta and Microsoft to make sure their AI products are safe before they release them.

Want a quick and expert briefing on the biggest news stories? Listen to our latest podcasts to find out What You Need To Know...