Paris AI Summit hangs in the balance

Shrugging off some winter showers, France’s Emmanuel Macron (co-hosting with India’s Narendra Modi) has rolled out the red carpet for heads of state, executives (hi Jamie Dimon), and AI power players at the latest AI Action Summit, which wraps today (Tuesday).

And a quick glance at the headlines might suggest that oh, Macron is announcing $110B in new AI investments, the EU’s von der Leyen is telling CEOs and anyone who looks like a CEO that the bloc is open for business, and JD Vance has landed with young family in tow, making the first transatlantic trip for the Trump 2.0 administration.

But trouble is brewing in the City of Light as the summit’s final declaration hangs in the balance. And that’s kinda the whole point of these AI summits — to agree on some basic rules of the road as AI keeps gathering pace. So with our very own Kristen there on the ground, here’s what folks are whispering by the coffee cart.

1) Red pens out

Let’s cut right to the chase – the summit declaration is in trouble. The early drafts focussed on AI safety, inclusion, and sustainability, which to the American guests in town sounded a lot like the exact things Donald Trump just got elected to rein in (ie, over-regulation). Even though Macron amusingly paraphrased the American adage, pledging Europe is ready to “plug, baby, plug”, the US doesn’t want to get boxed into Europe’s rules.

And that divide has put others in an awkward spot — the UK’s Keir Starmer, for example, pulled out citing domestic priorities, but for just a quick 40-minute plane ride over to Paris? The rumours swirled. For example, he cancelled after an early draft of the Paris text leaked, laying the above US-EU divide out there for all to see. So rather than get caught up in another transatlantic spat, he sent his tech secretary, who’s divided his time between sharing AI videos of himself speaking French and dodging questions from the press.

Macron’s team is now downplaying the importance of the declaration itself, which is not something diplomats do when they’re hopeful everyone’s about to sign a declaration.

Anyway, speaking of those Americans in town…

2) Vance does Europe 

Making his first international appearance as VP, JD Vance is coming in hot after dunking on Europe’s “evil trend towards censorship” last week. So, many at the summit were eager to hear from the young VP, particularly given his own stint in Silicon Valley, where he once showed sympathy for those wanting to break up the “Big Tech oligarchy”. 

But the Trump administration is unflinchingly clear it now wants to level-up “America’s AI dominance”, whether by rolling back regulations or announcing the $500B Stargate project. And that’s the message Vance delivered during last night’s address: the US wants fewer restrictions, fewer regulations, and more cash to flow.

Others here have sought to nuance any narrative that less regulation necessarily equals more progress — Kate Crawford, author of Atlas of AI, for example, has pointed out that highly regulated industries like banking and mining are still major innovation hubs.

But either way, it’s all left Brussels looking a bit isolated, just as its own new AI Act kicked into effect on February 2nd — even in its final watered-down form, that new law still looms large as a source of friction. Why, you ask? Well, much of that hinges on…

3) The money

JP Morgan’s Jamie Dimon, Google’s Sundar Pichai, OpenAI’s Sam Altman, Anthropic’s Dario Amodei, and pretty much every other major financier and tech pioneer all trekked to Paris. That’s because, with CEOs and world leaders in a sprint for any early-mover AI advantage, this summit becomes a marketplace not just for investments, but ideas:

• Cash-strapped leaders want these tech tycoons to invest in their countries, and
• The tycoons dangle that carrot in return for a more sympathetic ear on regulation.

So against that backdrop, any real mention of regulation or even the risks of AI now starts to look like a total Debbie Downer. And EU folks tell us it’s all pushing Europe to reconsider its regulatory approach.

Oh, and if you’re wondering what happened to all the buzz around China’s Deepseek, which rattled Wall St, Silicon Valley, and DC alike…?

4) China, wya

China has been relatively quiet at this summit — while President Xi has sent Vice Premier Zhang Guoqing this time around (a step-up from the relative junior-burger he sent in 2023), Zhang isn’t scheduled to deliver any grand remarks. From his perspective, if the West can’t even agree among itself on AI guardrails, why enter the fray?

INTRIGUE’S TAKE

This is just the third AI Action summit (the UK hosted the first in 2023), but each one has served as a useful straw in the wind:

• The EU has gone from promising rules that deal with AI’s ethical implications, to now focusing more on the need to compete and attract investment
• The US has gone from seeing cooperation as a way to project US power, to now fearing it as a recipe for ceding that power, and
• That’s shaped the way the US has gone from fearing Big Tech as something to break, to now cherishing Big Tech as a champion to back.

Meanwhile, initial shared AI fears around job losses, misinformation, and algorithmic bias are all giving way to a bigger fear: not winning the AI race.

Also worth noting:

• The first 2023 AI summit in the UK delivered a non-binding pledge by 28 nations (including the US and China) to tackle AI risks. A follow-up meeting hosted by South Korea last year secured another pledge to set up a network of public AI safety institutes.