UK signs AI agreement with EU and USA

'The first agreement with real teeth globally'

The three major Western regions developing AI technology have signed the world's first legally binding international treaty.

The USA and UK have signed the European Union's Convention on AI today, which aims to ensure the use of AI systems is fully consistent with human rights, democracy and the rule of law.

Officially known as the Council of Europe Framework Convention on artificial intelligence and human rights, democracy, and the rule of law (aka CETS No. 225), the agreement has been signed by Andorra, Georgia, Iceland, Norway, the Republic of Moldova, San Marino and Israel, in addition to the three major powers.

The agreement was drafted over two years by a coalition of 57 countries, with academia, civil society and private industry stakeholders contributing as observers.

The treaty provides a legal framework covering the entire lifecycle of AI systems. It promotes AI progress and innovation, while managing the risks it may pose to human rights, democracy and the rule of law. It is also technology-neutral, in the hope that this will future-proof the agreement.

Marija Pejčinović Burić, the Council of Europe's secretary general, said: "We must ensure that the rise of AI upholds our standards, rather than undermining them. The Framework Convention is designed to ensure just that. It is a strong and balanced text - the result of the open and inclusive approach by which it was drafted and which ensured that it benefits from multiple and expert perspectives.

"The Framework Convention is an open treaty with a potentially global reach. I hope that these will be the first of many signatures and that they will be followed quickly by ratifications, so that the treaty can enter into force as soon as possible."

Peter Kyle, the UK's minister for science, innovation and technology, said the convention is "the first [agreement] with real teeth globally."

However, while the convention is legally binding, critics point out that it includes no strict enforcements such as sanctions or fines. Instead, compliance is measured primarily through monitoring.

Brett Lambe, commercial lawyer, at Ashfords, says monitoring is "a fairly weak form of enforcement," and tech businesses especially "tend to take more notice where there is a threat of clear sanctions, such as fines."

He added, "As ever with regulation of fast-moving technologies, the key to [the convention's] success will be ensuring effective monitoring and, crucially, enforcement of breaches and non-compliance."

Several new AI laws and agreements have come into force around the world lately, including the Bletchley Declaration and the EU's AI Act, but they are either not legally enforceable or only apply in one region. The new agreement is both legally binding and cross-region.

Kyle pointed out to the Financial Times that the AI Act has been divisive. "Companies like Meta, for example, are refusing to roll out their latest Llama product in the EU because of it. So it's really good to have a baseline which goes beyond just individual territories."

The convention will enter into force on the first day of the month, three months after five signatories (including at least three Council of Europe member states) have ratified it.