AmanTrue Information

Deploying AI puts autonomous entities inside your systems: AI agents that read, decide and act, with access to records of every kind. Verified or not, each one is a path for sensitive data to leave custody, an actor nobody is accountable for, and an exposure your auditor cannot measure. Access is decided at the level of the system. Agents ask at the level of the fact. That gap is where control is lost. AmanTrue is the control layer for companies, institutions and governments deploying AI. It sits between the agent and the record and answers what today’s controls cannot: who the agent is, who is accountable for it, and what it can see and do. Every agent carries an identity issued by an authority, tied to the party accountable for it, revocable at any time. Every request is decided by your policy and enforced on the data: the agent gets the fact it needs, signed by the source. The record stays in your custody. Every answer carries a proof your auditor can check. Everything runs under your control: sovereign cloud, a national data centre, or your own servers, under your law. Built in Abu Dhabi, at Hub71, by a team that has run identity infrastructure for banks and governments since 2019. Its cryptographic libraries run inside products used by Microsoft, TikTok, HSBC and Telefónica Tech, across more than 9,000 projects. Selected for a European Commission regulatory sandbox on privacy-preserving verification.