An historic international agreement signed at the AI Safety Summit at Bletchley Park has recognised the significant risks posed by artificial intelligence.

The Bletchley Declaration was signed by 28 nations including the UK, US, EU, several independent European nations and – perhaps most significantly – China.

China’s vice minister of science and technology, Wu Zhaohui, told the conference his country is “calling for global collaboration to share knowledge and make AI technologies available to the public”, adding: “We uphold the principles of mutual respect, equality and mutual benefits. Countries regardless of their size and scale have equal rights to develop and use AI.”

The document outlines how development of the riskiest forms of AI ought to be managed on a global basis, with a focus on safety.

“Particular safety risks arise at the ‘frontier’ of AI… highly capable general-purpose AI models, including foundation models, that could perform a wide variety of tasks – as well as relevant specific narrow AI that could exhibit capabilities that cause harm – which match or exceed the capabilities present in today’s most advanced models,” the declaration states.

“Substantial risks may arise from potential intentional misuse or unintended issues of control relating to alignment with human intent. These issues are in part because those capabilities are not fully understood and are therefore hard to predict. 

“We are especially concerned by such risks in domains such as cybersecurity and biotechnology, as well as where frontier AI systems may amplify risks such as disinformation. There is potential for serious, even catastrophic, harm, either deliberate or unintentional, stemming from the most significant capabilities of these AI models. 

“Given the rapid and uncertain rate of change of AI, and in the context of the acceleration of investment in technology, we affirm that deepening our understanding of these potential risks and of actions to address them is especially urgent.”

Prime Minister Rishi Sunak labelled the declaration “quite incredible” while Technology Secretary Michelle Donelan also hailed the moment: “We have always said that no single country can face down the challenges and risks posed by AI alone, and today’s landmark declaration marks the start of a new global effort to build public trust by ensuring the technology’s safe development.”

Tesla and X owner Elon Musk, attending the summit, said AI could lead to humanity’s extinction.

“For the first time, we have a situation where there’s something that is going to be far smarter than the smartest human. It’s not clear to me we can actually control such a thing,” he warned.

In a video address to the event, King Charles said the risks must be tackled with “urgency, unity and collective strength… it is no less important than the discovery of electricity”.

Paul Teather, CEO at AMPLYFI, said ensuring safety from the start with frontier AI is vital.

“Ensuring frontier AI advancements are implemented as safely as possible requires the creation of robust regulatory frameworks delivered at pace, in contrast to Sunak’s sentiment that the UK ‘won’t rush to regulate AI’,” he said, citing earlier comments from the Prime Minister. 

Founders call for ‘bold action’ ahead of AI Safety Summit

“It’s important the UK considers AI technologies present outside of its jurisdiction, working with international and domestic partners to accurately assess the risks and benefits posed by frontier AI. Equally, the UK must look beyond what specific AI applications are being used for now, and regulate under the pretence of a ‘worst-case scenario’ eventuality. 

“This will provide assurance over whether specific AI applications are being misused, intentionally or otherwise, and the risks it poses to individuals, businesses, and whole nations.” 

Khofiz Shakhidi, chairman of Alif Bank, said the summit has “sparked a much-needed conversation on how regulation can encourage good practice, enhance protections, and allow AI technologies to flourish”. 

“This is the beginning of an important, global discussion that will continue for many years to come,” he said. “I encourage an open dialogue between regulatory bodies and FinTechs making huge strides in this area.”

South Korea will host the next AI Safety Summit in six months’ time, with France agreeing to host the subsequent one a year from now.

Plans submitted for £1bn Golden Valley cyber project