Delegates from 29 countries and regions wrapped up the first global AI Safety Summit on Thursday. British Prime Minister Rishi Sunak hosted the gathering at Bletchley Park in England, where early computer scientists broke codes during World War Two.

Participants also included business leaders. They discussed how to contain the danger AI poses to cybersecurity and biotechnology, as well as how they can "tip the balance in favor of humanity."

The British government invited China to the meeting, a decision that Sunak said wasn't "easy." He added that lots of people criticized him but that he thinks any serious conversation about AI safety "has to engage the leading AI nations."

Sunak said one of the achievements of the summit was that China signed a declaration on "shared responsibility" with the US and the other countries.

Sunak and US Vice President Kamala Harris also addressed threats to consumer privacy. They agreed to establish research centers to test new AI models before companies release them.

Japanese Prime Minister Kishida Fumio joined in via video. He has served this year as chair of the Group of Seven nations, and he reminded the others of codes of conduct discussed during a summit in Hiroshima to create trustworthy AI systems.