Mistral CEO warns closed AI models give providers ‘immense leverage’ over your business

The business move

Arthur Mensch, CEO and cofounder of the French AI lab Mistral, publicly urged companies to move away from closed AI models. In a LinkedIn post, he warned that closed AI providers are increasingly forcing businesses to give up control over their own data. According to Mensch, these providers now gain what he called “immense leverage” by retaining the internal data companies feed into their AI systems.

Why it matters

When businesses use closed AI models, especially in enterprise settings, providers see the context and data connected to those models. This visibility lets providers learn from proprietary company information, potentially using it to improve their own offerings or shape future contract terms. It also means companies have less control over their sensitive business data and face risks of vendor lock-in. This dynamic tightens provider power, reduces negotiation leverage for buyers, and raises privacy concerns.

For operators, this shifts the AI vendor relationship from a simple service to a strategic risk. The more internal business context and data are integrated with closed models, the more companies expose themselves to dependency and potential loss of control. This can increase costs and limit future AI strategy options while ceding valuable insights to third parties.

Who gains and who gets squeezed

Closed AI model providers gain power by collecting and leveraging client data. This gives them an advantage over competitors who do not have such direct access or control over business inputs. On the flip side, enterprise customers, especially those handling sensitive or proprietary information, are squeezed. Their internal data becomes a source of value for providers without clear reciprocity or ownership. Smaller businesses and startups locked into closed ecosystems face higher switching costs and competitive disadvantages.

What to watch next

Keep an eye on how enterprise AI contracts evolve around data ownership and retention clauses. Watch for whether companies start pulling back toward open-source or more transparent AI solutions to regain control. Also monitor regulatory responses that may arise around data privacy and vendor dominance in AI-driven business workflows. Finally, see if this pressure drives innovation and market shifts favoring open models that do not require wholesale data sharing.

AI Quick Briefs Editorial Desk