AI Agent Governance

About TenetGraph

Q: How does TenetGraph work?

TenetGraph follows four connected steps: Derive, Evaluate & Refine, Enforce and Prove. It derives the intended boundary from the specific agent and its context, tests that boundary against the agent, turns the refined boundary into runtime policy and records the resulting decisions.

TenetGraph is an AI agent governance platform that determines what each AI agent should be allowed to do, tests and refines that boundary, turns it into enforceable runtime policy, and records evidence of each decision made in production.

TenetGraph is built for two places where AI agents are taking action: Product Agent Governance for agents embedded in customer-facing products, and Employee Agent Governance for agents acting on behalf of employees across systems, tools and workflows.

Organizations can use TenetGraph directly or integrate its control loop into their product or runtime workflow.

The control loop

What TenetGraph does

01

Derive the intended boundary

TenetGraph starts with the specific agent and its intended purpose. It uses the agent's artifacts, capabilities, tools, permissions and governing context to derive the limits it is supposed to operate within.

02

Evaluate and refine it

TenetGraph challenges the agent against that boundary with an adaptive adversary that learns from the agent's behavior and decides where to probe next. Each finding shows where the agent can still operate outside the intended boundary and is used to tighten it.

03

Enforce it at runtime

TenetGraph turns the refined boundary into deterministic runtime policy. That policy can be enforced through TenetGraph or supported external enforcement points before the action executes.

04

Prove what happened

TenetGraph records the boundary, how it was tested, the policy in effect and each runtime decision. This creates evidence of what the agent was allowed to do and what happened when it acted.

Why TenetGraph

What makes TenetGraph different

Determine the right boundary, not just enforce a rule

Authorization and policy infrastructure can enforce a rule once it exists. TenetGraph helps determine and validate the operating boundary for the specific agent, then turns that boundary into the policy that should be enforced.

Evaluation that learns where to push next

TenetGraph adapts its testing based on how the agent behaves. It probes further where testing shows the boundary may not hold, then uses those findings to tighten the boundary before it is enforced.

Prove what was enforced in production

TenetGraph connects the agent's intended purpose, its boundary, how that boundary was tested, the policy in effect and the resulting runtime decisions into one evidence chain.

Who it is for

Who uses TenetGraph

TenetGraph is built for organizations governing AI agents that take action in production.

Product Agent Governance

For AI agents embedded in customer-facing products.

TenetGraph gives Product, Security, Governance, Trust and other stakeholders a way to determine what each agent should be allowed to do, test that boundary and show what happened in production.

Employee Agent Governance

For AI agents acting on behalf of employees across company systems, tools and workflows.

New tools, permissions, workflows, integrations, instructions and employee access can change what an agent is capable of doing. TenetGraph provides an agent-specific boundary that can be tested, enforced and re-established as those conditions change.

The team

The founding team

Chris Finan

Co-founder & CEO

Chris Finan has built and led security companies through scale and acquisition. Before TenetGraph, he was CEO at Anitian, President and CRO at ActZero, and Mid-Market GM at Shape Security, which was acquired by F5.

Jay Christopherson

Co-founder & CTO

Jay Christopherson has led engineering teams across complex technical systems. Before TenetGraph, he held engineering leadership roles at Mercedes-Benz R&D, Spaceflight Industries, and Apptio. He holds degrees in Computer Science and Business from the University of Washington.

Alexis Yelton

Co-founder & CPO/CAIO

Alexis Yelton has worked across AI, model training, and agentic systems. Before TenetGraph, she was AI Lead at ActZero and VP of Data and Machine Learning at Genvid Technologies. She holds a PhD and completed postdoctoral work in bioinformatics at UC Berkeley and MIT.

Emily Bertrand

Co-founder & CMO

Emily Bertrand has built and led go-to-market functions spanning revenue, sales, marketing, positioning, and product marketing. Before TenetGraph, she was Head of Marketing at Anitian and Mid-Market GM at ActZero. Earlier in her career, she worked in product marketing at Uber and Instacart.

TenetGraph is headquartered in Menlo Park, California.

Deployment

How TenetGraph fits your architecture

Direct

Use TenetGraph directly when you want to assess and govern an agent and produce evidence without first integrating TenetGraph into your product.

Integrated

Integrate TenetGraph when the boundary and control loop need to become part of your product or runtime workflow. Exact implementation is customer-specific.

Existing enforcement infrastructure can remain in place while TenetGraph helps determine and validate the boundary that should be enforced.

Reference

Key facts

Company name: TenetGraph
Category: AI Agent Governance
Company type: Privately held software company
Headquarters: Menlo Park, California, United States
Website: tenetgraph.ai
Co-founder & CEO: Chris Finan
Core offering: Determines what each AI agent should be allowed to do, tests and refines that boundary, turns it into enforceable runtime policy, and records evidence of each decision made in production
Primary use cases: Product Agent Governance, Employee Agent Governance
Core product loop: Derive, Evaluate & Refine, Enforce, Prove
Deployment: Direct or integrated into a product or runtime workflow

FAQ

Frequently asked questions

What is TenetGraph?

TenetGraph is an AI agent governance platform. It determines what each AI agent should be allowed to do, tests and refines that boundary, turns it into enforceable runtime policy, and records evidence of what happened when the agent acted.

What is Product Agent Governance?

Product Agent Governance is the governance of AI agents embedded in customer-facing products. TenetGraph determines the operating boundary for each agent, tests whether that boundary holds and turns it into policy that can be enforced in production.

What is Employee Agent Governance?

Employee Agent Governance is the governance of AI agents acting on behalf of employees across company systems, tools and workflows. TenetGraph applies the same core control loop to determine, test, enforce and prove the boundary for those agents.

How does TenetGraph work?

TenetGraph follows four connected steps: Derive, Evaluate & Refine, Enforce and Prove. It derives the intended boundary from the specific agent and its context, tests that boundary against the agent, turns the refined boundary into runtime policy and records the resulting decisions.

How is TenetGraph different from authorization and policy tools?

Authorization and policy infrastructure can enforce a rule once that rule has been defined. TenetGraph addresses the earlier problem of determining and validating what the specific agent should be allowed to do, then turns that boundary into policy that can be enforced.

Can TenetGraph work with existing enforcement infrastructure?

Yes. Runtime policy can be enforced through TenetGraph or supported external enforcement points. Teams can keep existing enforcement infrastructure while using TenetGraph to help determine and validate the agent-specific boundary that should be enforced.

Define the boundary. Authorize the action.

See TenetGraph on your own agents.

Bring an agent that's already running. See how TenetGraph determines its boundary, tests it and turns it into a control you can enforce and prove.

See it on your own agents Email a founder