Key facts
- Ex-Meta executive predicts US and EU firms will abandon OpenAI and Anthropic
- Called resistance to Chinese AI 'outright stupid'
- Reflects growing competitive parity of Chinese AI models
A former Meta executive has fired a sharp salvo into the global AI debate, declaring that enterprises across the United States and Europe will increasingly ditch OpenAI and Anthropic for Chinese AI alternatives — and that resisting such a move would be 'outright stupid'. The remarks, reported by NDTV Profit, represent one of the starkest public endorsements yet of Chinese AI competitiveness from someone with direct knowledge of the frontier model landscape.
The ex-exec's argument appears to rest on a straightforward commercial calculus: Chinese AI models have closed much of the capability gap with their American counterparts, while often being available at significantly lower cost or with fewer access restrictions. For large enterprises managing sprawling AI deployments, those differences can translate into meaningful savings and operational flexibility.
The comments land at a delicate moment. Governments in both the US and Europe are actively trying to limit Chinese AI adoption on national security grounds, warning that models developed under Chinese jurisdiction could expose sensitive enterprise data to state surveillance. The former executive's dismissal of such concerns as 'outright stupid' is likely to provoke pushback from policymakers who see the issue as far more than a cost-benefit calculation.
For India, already reported to be exploring Chinese AI models as OpenAI and Anthropic access remains uncertain, the ex-exec's remarks add an unexpected dimension: the shift away from US platforms is not just an emerging-market pragmatism story — it may be a global enterprise trend that even mature Western markets are beginning to follow.