AI Race: The United States Demands to Choose a Side - The French Mistral Turns to China

AI Race: It's Time to Choose a Side

The current race for artificial intelligence is actively focused on not giving an inch to the competition, without truly concentrating on the potential loss of control that this could imply in the short or long term, or the now much more real risk that quantum computing poses to cryptocurrencies.

This reality is compounded by increased competition from governments, prompting the United States to consider requiring the 35 signatory countries of the AI Opportunity Statement - created last June - to choose their side against China, which is making a strong entry with its open-source models like the now-famous Kimi K3.

Indeed, Reuters reports the preparation of an official letter intended to inform the members of this coalition led by the United States of their possible exclusion if they also adhere to the competing framework proposed by Beijing, in direct line with the Pax Silica plan implemented last year to secure supply chains related to AI models against the possible takeover by "concerned countries" like China.

In fact, this warning is aimed more specifically at Kazakhstan, due to its status as a potentially significant source of critical minerals and (especially) its problematic adherence to the Chinese coalition World Artificial Intelligence Cooperation Organization founded in July.

Being part of everything means being part of nothing. Signing the Pax Silica Declaration is not a mere membership, but a commitment. It is not compatible with joining initiatives that duplicate efforts and whose requirements contradict ours. Excerpt from the American letter

Has the French Company Mistral AI Already Chosen China?

A firm stance from the United States that does not directly concern France, since our government has not adhered to the AI Opportunity Statement and is also not among the signatories of the Pax Silica coalition, even though the European Union officially joined it last June.

However, this threat issued by Washington raises certain questions regarding the recent strategy implemented since the beginning of August by one of Europe's AI champions, the French company Mistral AI, which now hosts the Chinese model GLM-5.2 developed by Z.ai on its platform.

This evolution has been sharply criticized by French entrepreneur and AI engineer Pierre-Louis Biojout, who explains on X how Mistral AI seems to be transitioning from a cutting-edge European AI lab project to a mere "integrator of models designed by others, notably Chinese ones."

But before too quickly burying this French gem, China and AI specialist on X, MadMaks, explains how Mistral AI's strategy fits into a competitive context he considers already lost, facing industry giants like OpenAI and Anthropic.

As a result, and in order to repay its creditors, Mistral AI has reportedly decided to provide its French computing capabilities to other models. MadMaks sees this as a two-step strategy: monetizing its computing infrastructure to generate more predictable revenues while simultaneously developing a potentially more strategic activity in "physical AI," an artificial intelligence capable of interacting with the physical world and applicable to industries like Airbus, BMW, or EDF.

A case to follow...

Sources: Reuters, Pierre-Louis Biojout, MadMaks

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.