Venice AI: This “uncensored” AI platform is growing, and privacy is the reason why

Venice AI has turned privacy into a business model. Ram Iyer reports for TechCrunch that the two-year-old startup now counts more than 850,000 unique website visitors, 3 million active users, and 1.7 million API calls per day. This week it raised a $65 million Series A at a $1 billion valuation, but the real story is how the platform itself works.

How the platform works

Venice gives users access to more than 200 AI models covering text, image, audio, and video generation. It hosts “uncensored” open source models on its own data centers and routes requests for closed-source models, such as those from OpenAI or Anthropic, through an external proxy. According to the company, no user data is stored on its systems, and paying subscribers can get end-to-end encryption on select models.

CEO Erik Voorhees, an early bitcoin advocate who previously founded ShapeShift, describes the approach as treating AI as a “neutral tool.” Asked about risks like AI-related psychological harm, he told TechCrunch that constant surveillance poses a bigger danger than occasional misuse. “We’re optimizing for freedom and actually respecting users as adults,” he said.

The company also lets users customize AI “characters” to chat with, and says it adjusts open models’ system prompts to make them more responsive rather than adding restrictions.

Venice is already profitable, with annualized revenue above $70 million, and plans to use its new funding to build its own data centers and reduce reliance on leased GPUs.