The US and UK have refused to sign a declaration on AI safety at the Paris Summit. Is there a ‘third way’ for regulation, asks Russ Shaw in today’s Notebook

Is a ‘third way’ on AI regulation possible?

Discussions have concluded at the two-day AI Action Summit in Paris, which sought to be a landmark moment for global governance efforts.

The geopolitical backdrop to the talks felt more polarised than ever. On the one hand, the EU has doubled down on consumer protection across all sectors in recent times – most notably with the passage of the AI Act last year, which implemented a risk-based approach to AI regulation and established a finite list of limits for its development.

On the other hand, it’s clear that the new administration in the US is committed to a programme of de-regulation for the AI sector. Look no further than President Trump’s decision to revoke Joe Biden’s executive order on AI safety for evidence that the US has set sail on a regulatory journey significantly at odds with the landscape in Europe.

And although it was encouraging to see Vice President JD Vance and Vice Premier Zhang Guoqing in Paris this week, ultimately the US (along with the UK) refused to sign a declaration to ensure that the technology is “safe, secure and trustworthy”.

While not surprising, this was nonetheless a disappointing outcome. Ultimately, AI is not an issue which any single nation or bloc can regulate alone. The very nature of AI is inherently cross-border – its impacts and potential risks, like data privacy breaches or algorithmic bias,........

© City A.M.