Credal is a company building the security and governance control plane for enterprise AI. Its founders, Jack Fischer and Ravin Thambapillai, came from Palantir, where data access control and auditability are central, and they built Credal to solve the equivalent problem for generative AI: how can a large or regulated organization safely connect its internal data to LLMs and deploy AI assistants and agents without leaking sensitive information or violating existing permissions?
The platform sits between company data sources and AI models, acting as a secure gateway. It connects to enterprise systems and enforces the same access controls that govern the underlying data, so an AI assistant only ever surfaces information a given user is actually permitted to see. Credal layers on capabilities critical for regulated environments: automatic redaction of personally identifiable information, comprehensive audit logging of every AI interaction, data-loss prevention controls, and policy enforcement over which models and data can be combined. This lets security and compliance teams approve AI deployments with confidence rather than blocking them.
On top of this secure foundation, Credal lets organizations build AI assistants and increasingly autonomous agents that handle multi-step workflows across connected systems. Because governance is built into the control plane, these agents inherit the right permissions and leave a full audit trail, which is essential in sectors like healthcare, financial services, and government.
Credal raised a seed round of $4.8M in October 2023 led by Spark Capital, and is backed by Y Combinator and Drive Capital among others. Its customers include the U.S. Department of Health and Human Services, MongoDB, Comcast NBCUniversal, and Wise. Credal targets enterprises that want to adopt AI quickly while keeping data security, permissions, and compliance firmly under control.
