TRUSTPASS AI

The Human Approval Layer for Autonomous AI- The pause button for autonomous agents — so every action your AI takes has a human signature behind it. --- Don't let your agents YOLO it

Comment

Inspiration

The future of AI is not about generating text. The future of AI is about taking action. Today, AI can find the best flight, negotiate with suppliers, review invoices, compare vendors, and make recommendations faster than any human. But when it comes time to actually spend money, approve a payment, or make an important decision, everything stops. The AI knows what to do. The human knows the AI is probably right. Yet nobody feels comfortable letting the AI act alone. We realized the real problem isn't intelligence. It's trust. Imagine an operations manager trying to coordinate travel for dozens of employees. AI can do the work. But organizations still need confidence that critical decisions remain under human control. That's why we built TrustPass AI. A trust layer that allows AI to move fast while ensuring humans stay accountable.

What it does

TrustPass AI sits between AI decisions and real-world actions. Before AI performs a critical action, TrustPass ensures: The action is reviewed Risk is evaluated The correct person approves it Identity is verified Approval is recorded The action is safely executed

TrustPass transforms AI from an assistant that recommends into a system that can safely act. Example: Flight Booking An AI travel assistant finds the best flight: Singapore → Tokyo Price: SGD 850 Departure: Friday 8 PM Instead of forcing the user to manually repeat the entire booking process, TrustPass requests approval. The user reviews the request, approves it, and the booking is completed. Fast. Simple. Accountable.

How we built it

We combined AI reasoning, workflow orchestration, and approval management into a single platform.

Frontend-> Next.js, React, TypeScript, Tailwind CSS,

Backend -> Java, Springboot

Database -> PostgreSQL

AI Layer -> OpenAI

Used for: -> Risk evaluation, Decision reasoning, Action classification, Workflow orchestration

Voice Layer -> ElevenLabs

Used for:

  1. Human interaction
  2. Approval workflows
  3. Natural language confirmations

Challenges we ran into

One of the biggest challenges was defining the balance between autonomy and control. If every AI action requires approval, users become overwhelmed. If AI can act without approval, organizations lose trust. Finding the right balance between automation and accountability became the core challenge. Another challenge was simplifying a complex problem. Governance, compliance, approvals, and trust are difficult concepts.

Accomplishments that we're proud of

We are proud that TrustPass AI addresses a real barrier preventing AI adoption. Many AI products focus on what AI can do. We focused on what organizations need before they allow AI to do it. Our biggest accomplishment was creating a system that enables AI to move faster while preserving human accountability. We also built a framework that can work across multiple industries instead of being limited to a single use case.

What we learned

Building TrustPass taught us that trust is not a feature. Trust is infrastructure. Organizations do not hesitate because AI lacks intelligence. Organizations hesitate because they lack confidence. It will be determined by how safely humans can delegate responsibility to them.

We learned that successful AI systems must balance:

  1. Automation
  2. Transparency
  3. Accountability
  4. Human agency

What's next for TRUSTPASS AI

Today TrustPass focuses on approval workflows. Tomorrow we envision something larger. A universal trust layer for autonomous systems. Autonomous Business Workflows

  1. Supporting travel,
  2. procurement,
  3. healthcare administration,
  4. Operation
  5. Others

Built With

Share this project:

Updates

Submission history