Confident Security Emerges to Safeguard AI Data with $4.2M Funding
Confident Security Emerges to Safeguard AI Data with $4.2M Funding
As artificial intelligence continues to revolutionize industries, concerns about data privacy have become a major barrier to wider adoption—especially in highly regulated sectors like healthcare, finance, and government. Many companies using advanced AI models from providers such as OpenAI, Google, Anthropic, and others are often uncertain about where their sensitive data goes, who can access it, and whether it could be used for training future AI systems.
Confident Security, a San Francisco-based startup, is stepping in to address this challenge. Dubbed “the Signal for AI,” the company has developed CONFSEC, an end-to-end encryption solution designed to protect prompts and metadata when interacting with foundational AI models. This tool ensures that neither the AI provider nor any third party can view, store, or use the data—even for model improvement or monitoring purposes.
How CONFSEC Protects Your Data
CONFSEC works by encrypting user data before it ever leaves the client’s environment. The system leverages anonymization and secure routing through services like Cloudflare or Fastly, meaning the original source and content of the data remain hidden—even from the AI model’s servers. Advanced encryption protocols ensure that data can only be decrypted under strict, transparent conditions. For example, decryption is only permitted if the server agrees not to store, log, or use the information for training purposes.
Inspired by Apple’s Private Cloud Compute (PCC) architecture, CONFSEC also ensures that all AI inference software is publicly logged and auditable, allowing independent experts to verify its privacy claims.
Industry Impact and Adoption
Confident Security’s approach offers a breakthrough for organizations hesitant to embrace AI due to privacy concerns. By acting as an intermediary between AI vendors and customers—ranging from hyperscalers to governments and enterprises—the company aims to unlock new opportunities for responsible and secure AI adoption.
Even AI companies themselves may integrate CONFSEC to reassure enterprise clients. With the rise of AI-powered browsers and tools, such as Perplexity’s Comet, there’s demand for privacy-first solutions that prevent sensitive data from being misused or accessed by unauthorized parties.
Funding and Future Plans
Confident Security emerged from stealth with $4.2 million in seed funding led by Decibel, South Park Commons, Ex Ante, and Swyx. According to the company, CONFSEC has already undergone external audits and is production-ready. The team is currently in discussions with banks, browsers, and search engines to integrate CONFSEC into their infrastructure stacks.
“You bring the AI, we bring the privacy,” said Jonathan Mortensen, founder and CEO of Confident Security.
As the need for trustworthy AI infrastructure grows, solutions like CONFSEC could become essential for businesses looking to harness AI’s potential without compromising data security.
