Enigmata, a Nashville startup building encryption that lets AI work with data it never actually sees, has emerged from stealth with a $6.5 million seed round led by Blockchange Ventures.

The company is led by co-founder and CEO Scott Searle.

AI on Encrypted Data

Companies in healthcare, finance, and government hold some of the most valuable data for training AI, as well as some of the most sensitive. Today that data usually has to be decrypted before a model can learn from it, creating exposure every time it moves.

Enigmata's answer is Enigmata Cipher, a patent-pending cryptographic technology that allows AI systems to train on, search, and analyze encrypted data without exposing the underlying plaintext.

"Every era of computing has forced a choice between using data and protecting it," Searle said. "Enigmata Cipher ends that trade-off."

The Speed Question

Privacy-preserving computation is not new. Techniques such as fully homomorphic encryption have existed for years, but they have struggled to go mainstream because working on encrypted data is typically far slower than working on raw data.

Enigmata says its approach avoids that penalty. In internal benchmarks, models trained on Cipher-protected data matched the accuracy of models trained on raw data and completed training 8% to 10% faster.

Those results have not yet been independently validated, and outside observers have flagged the claim as the one to watch. Remio noted that the speed claim faces a hard test as the technology moves beyond the lab.

Why Blockchange Is In

Blockchange Ventures has historically backed infrastructure in crypto and security, and sees data privacy as the next constraint on AI adoption.

"Privacy is the next great frontier for AI," said Matt Immerso, general partner at Blockchange Ventures, adding that Enigmata built its encryption for that purpose from the ground up.

Why It Matters

As regulators tighten rules on personal data and enterprises grow wary of feeding proprietary information into AI models, the ability to use data without exposing it could unlock datasets that are currently off-limits.

If Enigmata's performance numbers hold up in customer deployments, it would address the main reason encrypted computation has stayed a niche technology. A $6.5 million seed gives the Nashville company room to prove that out with early customers in regulated industries.

Source

PR Newswire — "Enigmata Emerges from Stealth with $6.5 Million to Let AI Learn Without Exposing Sensitive Data"