Mistral CEO Says New AI Model Beats Chinese Rivals on Cybersecurity

French artificial intelligence company Mistral has unveiled a new AI model that it claims outperforms Chinese rivals in cybersecurity, with chief executive Arthur Mensch announcing the launch on stage at the Ai Everything conference in Abu Dhabi on Tuesday.

Speaking at the sold-out event, Mensch said the newest model — the company's first major release since April — sits above Chinese models "on certain aspects, including cyber," directly challenging the narrative that Europe cannot compete with either the United States or China in frontier AI development.

Mensch stopped short of naming which Chinese models Mistral's latest system beats, and did not specify the benchmarks or measures behind the claim.

But the positioning was unmistakable: the 3-year-old startup is pitching itself as a safer, European alternative in a market increasingly dominated by lower-cost Chinese open models and the massive US labs.

The announcement comes at a moment of intense competition among open-weight model makers.

Chinese models such as DeepSeek and Z.ai's GLM series have won global adoption by being cheaper and highly customizable, while Nvidia-backed Reflection AI launched its own open-weight model, Beam, just a day earlier to take on the same Chinese rivals in coding and agentic tasks.

By leading with cybersecurity — a domain where trust and safety assurances matter as much as raw performance — Mistral appears to be aiming at enterprise and government buyers who are cautious about relying on models built in China or the United States.

Mistral, valued at 21 billion euros after raising 3 billion euros in September, is widely seen as Europe's best contender in the global AI race, though it remains far smaller than US giants Anthropic and OpenAI. Its pitch rests on independence: technology that belongs to neither the American nor the Chinese camp.

Mensch argued on stage that this independence is now paying off. "There's an enormous desire and an enormous need for alternative technology suppliers," he said, pointing to the company's rapid growth in the Gulf and Asia-Pacific regions as evidence that buyers want options beyond the usual two poles.

The funding round last month gave the Paris-based company fresh firepower to scale up its models, infrastructure, and enterprise sales at a time when European policymakers are pushing hard for technological sovereignty in AI.

The choice of Abu Dhabi for the launch is itself a statement. Ai Everything Abu Dhabi, running October 6–7 at the ADNEC Centre, has drawn more than 400 AI infrastructure builders, 200-plus investors managing 91 billion dollars, and delegations from over 60 countries, with 180 speakers on the program.

Mensch appeared alongside figures including Boston Dynamics founder Marc Raibert to discuss sovereign AI and industrial intelligence, underscoring how the Gulf has become a battleground for AI influence — and a hunting ground for investment and cloud partnerships.

For Mistral, the Gulf offers both capital and customers eager for non-aligned AI suppliers. For its rivals, the message from Abu Dhabi was blunt: Europe intends to compete, and it is starting with the domain where trust matters most.