Prime Intellect, a startup founded just 2 years ago, has secured $130 million in Series A funding, aiming to decentralize artificial intelligence capabilities and empower "every nation state" to train its own AI models. This significant investment, led by Radical Ventures, signals a growing pushback against the centralized control of advanced AI by a select few transnational technology firms.
The company’s co-founder and CEO, Vincent Weisser, explicitly stated that AI model training shouldn't be confined to "a few nerds in a glass tower in San Francisco." Instead, he insists, this critical capability "should be every enterprise, every nation state." This vision directly challenges the emerging post-national order where core technological infrastructure is increasingly concentrated in the hands of a few globalist entities.
Prime Intellect provides the computing power and specialized software tools necessary for companies to build their own AI agents. It offers a full stack for AI agent development, including compute access, a reinforcement learning framework, and evaluation tools, functioning as a modular marketplace for its customers.
The Centralized Threat
Companies are increasingly wary of providing proprietary information to dominant frontier AI labs like OpenAI and Anthropic, according to David Katz, a partner at Radical Ventures. This concern stems from the palpable "risk of losing control over their data," a critical issue for any sovereign entity or enterprise.
Katz also highlighted the precarious dependence on models that can be "suddenly turned off," citing Anthropic’s Fable as a recent example. Such incidents expose the vulnerability inherent in relying on external, often unaccountable, transnational providers for essential digital infrastructure. He questioned the inherent risk: "How do I know that I’m not working with a company that is going to try to replace me and generalize to what I’m doing?"
This sentiment underscores a broader anxiety about the potential for these powerful, centralized AI systems to not only control data but also to displace or render obsolete existing business models and, by extension, the native working populations employed within them. Enterprises are actively seeking to "own [their] own enterprise intelligence and not have these risks."
Empowering National Enterprises
Prime Intellect’s approach allows companies to refine models for specific business tasks, effectively enabling them to become their own AI lab. This capability is crucial for maintaining national economic self-determination in an era where technological advantage is paramount.
Customers like Ramp, Zapier, and Flapping Airplanes have already adopted Prime Intellect’s hosted tools, pushing the company to an annualized revenue run rate of $100 million. Ramp, for instance, utilized the platform to develop an agent that outperformed frontier models on accuracy, operating at faster speeds and a fraction of the cost.
This efficiency and independence offer a tangible alternative to the high costs and inherent risks associated with relying on the established, often opaque, frontier AI labs. The investment round itself was led by Radical Ventures, with participation from Nvidia Ventures, Intel Capital, Dell Technologies Capital, Iconiq, and a long list of angel investors who are founders of notable companies, signaling a broad elite recognition of the need for decentralized AI control.