IntelXview

IntelXview AI Control Plane

Control what AI agents can do. Prove what they did.

ACP checks requests routed through it against policy and records the decision for review.

Open AI Control PlaneSee it running

Opens the ACP sign-in page. Activity outside its configured execution path is not controlled by ACP.

A proof run, with its limits visible

ACP public evidence screen · 4 October 2026

ACP public proof-run status, dated 4 October 2026, with the evidence-chain result and assurance note
This capture shows one automated proof run. It does not establish the outcome of your workflow. Inspect the current evidence and limits.

Before an AI agent acts, what is it authorised to do?

Follow one governed action.

Start with an action in your workflow. Follow who requested it, the authority they hold, the policy decision and the record it leaves.

  1. 01

    Identity

    Establish who is making the request.

  2. 02

    Authority

    Check the authority delegated to that actor.

  3. 03

    Mandate

    Define the permitted purpose and scope of the action.

  4. 04

    Policy decision

    Evaluate the routed request against the configured policy.

  5. 05

    Enforcement

    Apply the decision in the supported execution path.

  6. 06

    Evidence

    Keep the decision record available for review.

This sequence explains the control path. Coverage depends on the integration and policy configuration; it does not extend to tools or calls that bypass ACP.

What would you show for one governed action?

Inspect a published decision record, then check its chain or the separate signed evidence pack. Each evidence page sets out the artefact, verification steps and limits.

View the example decision record

A public sample and a recipe for checking its hash chain. Internal consistency is not proof that every action was captured.

Inspect the signed evidence pack

Download the pack and follow its signature-verification instructions. A signature does not establish regulatory compliance.

Start with the workflow you want to govern.

Explore ACP, or describe the agent action, systems and review requirements you want to evaluate. A free initial call helps establish the integration boundary and what evidence you need.

Open AI Control PlaneTalk through your workflow

Product access opens sign-in. An initial conversation about one workflow, to see whether ACP could help. No charge, no pressure. Booking or attending the call does not commit you to buying anything; any paid work needs a separate written scope and price that you choose to accept. To change or cancel, reply to your calendar invitation.

Questions about evaluating ACP

Testing, with a clear boundary

IntelXview Limited has been selected to participate in Cohort 2 of the FCA AI Supercharged Sandbox. Participation does not imply FCA endorsement, authorisation, registration, approval, or validation of IntelXview Limited or its products or services.

AI Control Plane demonstration

90 seconds · Captioned · No audio · Recorded August 2026

A fictional organisation and synthetic data show an allowed action, a refused action and a decision-chain check. The example covers the demonstrated ACP workflow. It does not establish production-key custody.

Read the demonstration steps and limitations
Open AI Control PlaneTalk through your workflow

Product access opens sign-in. An initial conversation about one workflow, to see whether ACP could help. No charge, no pressure. Booking or attending the call does not commit you to buying anything; any paid work needs a separate written scope and price that you choose to accept. To change or cancel, reply to your calendar invitation.