Riskflow Oversight

AI moves fast. You make the call.

Riskflow Oversight is a standalone hub for decisions AI needs you to make. it's a central human-in-the-loop interface agents can put in their workflows which allows to triage decisions in an audit-ready way.

a simple concept:

  1. an agent that needs a decision makes a submission
  2. Riskflow Oversight provides rule-based triaging and approval workflows
  3. the agent polls continuously for a decision or receives a webhook to proceed
Agents running independent workflows; one waits with two of three human approvals received.
From a submitted request and its evidence to a recorded human approval.
A decision form with space for a justification and Approve and Reject buttons.

Human approval

There are times when a human needs to make and own a decision. This is where an approval workflow is needed. Based on rules, choose how many humans need to approve a submission.

Submission rules with a draft finance rule requiring two approvals, above the default rule.

Routing submissions

Not every submission needs two approvals. Not every submission should be reviewed by the same users. Set rules to decide who should review what.

Recent similar decisions showing an approved follow-up and a rejected renewal reminder.

Learn from human decisions, lift when feeling confident

Every decision for a submission can be used for learning. Watch metrics on how confident a learning process is and choose when to lift a HITL workflow to an automated decision workflow.

Free to use, plans ready for enterprise support

Riskflow Oversight is designed with regulated environments in mind. Integrate your identity and access management, use strong delegation methods and generate audit-related documents on demand.