ReguLattice AI Governance Platform - AI Governance Platform for AI Systems & Agents
by•
ReguLattice is an AI-native governance platform that helps organizations discover, assess, govern, monitor, and improve AI systems and AI agents. It finds shadow AI, keeps people in charge of high-risk AI decisions, and maintains live, audit-ready evidence mapped to ISO/IEC 42001, the NIST AI RMF and the EU AI Act.
Replies
Hi everyone, I’m Moazzam, the maker of ReguLattice, from Karachi.
Companies are starting to let AI agents answer customers, change records and move money. I kept seeing the same gap: most AI governance is documents and dashboards that describe risk after the fact, and nothing stops a bad action while it’s happening.
So we built a gateway that sits in front of the AI. Every prompt, agent tool call and AI payment gets an answer: ALLOW, SIGN_OFF (a named person approves) or BLOCK. The answer comes from fixed rules, not another model, and if the gateway is down, the action doesn’t run. Every decision goes into a signed record that an auditor can check offline, without trusting us. It can also find AI tools your team uses without approval, and it can run entirely on your own servers with local models.
In our own red-team test, 215 attacks on the payments gateway were all blocked or held, and all 40 genuine payments went through. That is our own test, not an independent audit, and we have no revenue yet.
The sandbox asks for a name and email before it opens, so we can follow up with people who try it.
Who it’s for: teams in banks, fintechs and other regulated companies that want to use AI agents but must show a regulator that they’re in control.
I’d really like your feedback on two things: which attack should we try next, and what would you need to see before trusting a tool like this in front of your AI agents?