"The UK, the US and 28 other countries have agreed to strengthen AI safety cooperation to prevent catastrophic injuries"

Countries including the UK, US and China have agreed to a consensus on the risks posed by advanced artificial intelligence, pledging to ensure the technology is developed and deployed safely.

At the British government's two-day "Global Artificial Intelligence Security Summit" held this week, 28 countries including Brazil, India, Nigeria and Saudi Arabia, as well as the European Union, signed the "Bletchley Declaration" on AI. protocol.

The UK government said the declaration achieves the summit’s key aim of establishing common agreement and responsibilities on the risks, opportunities and international cooperation moving forward in advanced AI safety and research, particularly through wider scientific collaboration.

Participating countries agreed that potential intentional misuse could pose serious risks and highlighted concerns about cybersecurity, biotechnology, disinformation, bias and privacy risks

The declaration quotes that AI models' "most important functions may cause serious or even catastrophic harm, whether deliberate or intentional."

In addition, relevant countries agree to encourage developers of advanced AI to maintain transparency and accountability in measuring, monitoring and mitigating harmful capabilities.

British Prime Minister Rishi Sunak called the agreement a landmark achievement in which "the world's most powerful countries unanimously recognize the urgency of understanding the risks of AI." C114 Communication Network Jiang Junmu

The above is the detailed content of "The UK, the US and 28 other countries have agreed to strengthen AI safety cooperation to prevent catastrophic injuries". For more information, please follow other related articles on the PHP Chinese website!