UK and US among first signatories of global AI treaty

Drawn up by the Council of Europe, the Framework Convention on Artificial Intelligence – referred to simply as the AI Convention – is intended to ensure that activities encompassing the entire lifecycle of AI systems are consistent with human rights, democracy and the rule of law. In parallel, it also aims to support technological progress and innovation.

In practical terms, it requires signatories to document and share relevant information regarding AI systems and their usage, and give anyone adversely affected by them sufficient means of recourse.

States must also agree to carry out risk and impact assessments in respect of actual and potential impacts on human rights, democracy and the rule of law, using these as the foundation for implementing effective prevention and mitigation techniques.

It also leaves the door open for authorities to introduce bans or moratoria on certain applications of AI systems.

The treaty does not seek to replace current ongoing efforts to police the AI sector, like the Bletchley Declaration or the EU's AI Act, but rather to provide additional legal scaffolding to support these endeavours.

The convention was hammered out by no fewer than 68 international representatives from civil society, academia and industry, and included contributions from multiple other international organisations.

"We must ensure that the rise of AI upholds our standards, rather than undermining them. The Framework Convention is designed to ensure just that," said Council of Europe secretary general Marija Pejčinović Burić. "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 US, UK, and EU have all signed up, as have Andorra, Georgia, Iceland, Israel, Norway, the Republic of Moldova, and San Marino. Countries from all over the world have been invited to join in and comply with its provisions.

The AI Convention will become effective three months after five signatories – including at least three member states of the Council of Europe – have ratified it.

"The Framework Convention is an open treaty with a potentially global reach," said Burić. "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."

This time last year, when the AI hype train seemed unstoppable, the thought of a global, legally binding AI treaty would in all probability have been considered too little too late.

But the train's momentum is showing signs of slowing, as more questions are raised about AI's usefulness in its current form, and whether the technology stands a realistic chance of generating a return on the massive investments in AI infrastructure and development.

So far, those pushing AI have yet to come up with concrete answers, beyond the oft-repeated and ambiguous refrain that it promises to revolutionise various sectors of the economy and everyday life.

If it does, then the world will be grateful for guardrails like the Council of Europe's AI Convention. But the longer AI takes to realise its potential, the more these globally-coordinated efforts at containment begin to look like overkill.