Anthropic's Chief Executive Officer Dario Amodei compared artificial intelligence chips to nuclear weapons. Google DeepMind's CEO said China's technology is just months behind the US. Signal's Meredith Whittaker called new adaptations of AI uniquely "perilous".
The message from tech executives at the World Economic Forum is that AI is a powerful geopolitical force, and one politicians and business leaders need to wield as such. It's effectively one more piece of firepower in the global struggle between major economies like the US and China.
The idea that AI is more than just a buzzy next-gen tech but a crucial tool for whoever wants to be a 21st century superpower, means it's high on the agenda at Davos. It's holding its own in the attention stakes, even if the buildup to the annual gathering of the world's global elite has been dominated by Donald Trump and tensions over Greenland.
In a sign of the AI push at Davos, Anthropic and tech venture capital investors Lightspeed and General Catalyst planned exclusive parties. Even with a busy schedule ahead for the week, there was a line outside the VC event at 10:30pm. Palantir Technologies Inc. set up its house across the street from the USA House, a church just outside the Davos perimeter that's the main venue for the Trump administration this week.
AI is becoming a major driver of the global economy – with companies and nations committing billions to ensure dominance – and has implications for everything to the future of work and robotics to how wars are fought and space travel.
With so much at stake, countries can't afford to slow down when it comes to investment and innovation. And, according to Amodei, the US even risks giving up its edge over China by selling advanced AI chips to its rival.
"If you think about the incredible national security implications of building models that are essentially cognition, that are essentially intelligence," Amodei said. "It's like selling nuclear weapons to North Korea."
It was almost a year ago that the Chinese large-language model DeepSeek spooked the industry – and the western world more broadly – with a new model that performed similarly to leading models like OpenAI's ChatGPT on some metrics at a fraction of the cost.
It drove a rout of nearly $1 trillion in US and European technology stocks and – temporarily – wiped out hundreds of billions of dollars from Nvidia's market value. The technology has yet to meaningfully supplant its American rivals outside of China, but it was quite the wake-up call.
"They may be only six months behind, not one or two years behind, the frontier. That's what DeepSeek showed," Google DeepMind chief Demis Hassabis said. "But they have yet to show they can innovate beyond the frontier."
Europe Lags
In this tech race between the US and China, Europe is falling behind. Microsoft Corp.'s Satya Nadella said the continent needs a change in mindset.
In his view, Europe is too focused on regulation and not enough on encouraging homegrown tech.
"You're only going to be competitive if the products coming out of Europe are globally competitive," he said. Europe has led in privacy and in safety around AI, Nadella added, "but you also have to complement it by building locally and thinking globally."
Europe is losing out as many of its most promising companies get absorbed by larger tech companies abroad. While governments have been marshalling resources to build out data centers and deploy AI, the scale is significantly smaller than the US and Asia.
France's Mistral AI is Europe's leading AI startup. But valued at 11.7 billion euros ($13.7 billion) in a funding round last year, it's a minnow compared with OpenAI, valued at north of $500 billion.
Eric Schmidt, the former Google CEO and a tech investor, said that Europe must ramp up investment in open source AI or it will be left with few options but to rely on China. US companies are moving toward a "closed source" AI.
"Unless Europe is willing to spend lots of money for European models, Europe will end up using the Chinese models," Schmidt said. "It's probably not a good outcome for Europe."
'Intimidation'
While economies and companies are working out how to safely integrate AI without losing their edge, the technology continues to become more powerful.
For businesses thinking about adopting it, Whittaker said executives need to ask tough questions about what they need and what will work for them. She says there's sometimes an "intimidation factor" at play when it comes to the tech.
"Ask the questions, and ask the questions selfishly in relation to what your business actually needs," she said. "Get as specific as possible."
The technology is also advancing in ways that will force governments to reconsider how they structure their economies and deploy resources.
"We could have this very unusual combination of very fast GDP growth and high unemployment or at least underemployment or a lot of low wage jobs, high inequality. I don't think that's a macroeconomic combination we've ever seen before," Amodei said. "That's really going to be a problem we need to solve."
Hassabis said he'd like to see the world's best minds collaborate in a scientific way - combining philosophers, social scientists, economists and technologists, to work out how best to deploy the technology.
"Unfortunately it kind of needs international collaboration."
