Interactive demos

See it work

Governed AI, security response and an AI helpdesk agent, running on sample data. Click through each one instead of reading about it.

Sample data. Real workflow.

A prompt with sensitive data meets the policy engine

Pick a sample prompt. Watch what the engine detects, which policy fires, what the model is allowed to see, and what gets logged. Then switch enforcement off.

No customer data is shown. Every name, record and value below is synthetic. This follows the documented AiT AI Gateway and Hosted Agents flow: policy check first, approved model second, audit log always.

1. Prompt as typed

Summarize this intake note for the care team: Jane Q. Sample, DOB 01/02/1970, MRN TEST-000001, reports poor sleep for three weeks.

2. Detection

  • Patient name Jane Q. Sample
  • Date of birth 01/02/1970
  • Medical record number TEST-000001

3. Policy that fired

POL-PHI-01 PHI minimization

Protected health information is removed before any prompt leaves the tenant boundary. Action: redact.

4. Redacted prompt

Summarize this intake note for the care team: [PATIENT_NAME], DOB [DOB], MRN [MRN], reports poor sleep for three weeks.

5. Model routing

Approved model only

Approved model via the AI gateway, de-identified payload only

6. Audit-log entry

event
sample-0001
user
user@example.test
policy
POL-PHI-01
detections
3
action
redact
stored
redacted copy only
entry
sha256:3b1f...c0de (sample)

Sample data. Real workflow.

A security alert, from detection to resolution

Step through one sample incident to see who or what acts at each stage. The service-level promise sits underneath.

No customer data is shown. Every name, record and value below is synthetic. The stages mirror the published Managed SOC service: monitoring, automated containment, senior-engineer escalation, forensic preservation and incident reporting.

  1. Detection Automated monitoring, 24/7

    Identity and email telemetry are correlated. A sign-in from a new country is followed by a new mailbox forwarding rule on the sample account jordan@example.test. An alert is raised without waiting for anyone to notice.

  2. Triage AI triage first

    The alert is checked against the account's normal sign-in pattern, scored for severity, and the context is attached to the ticket so the next person starts informed.

  3. Containment Automated response

    Active sessions are revoked and the forwarding rule is removed. If an endpoint is involved, automated isolation runs in under 30 seconds.

  4. Escalation US-based senior engineer

    A senior engineer takes the ticket with the evidence already attached, confirms the scope, and the client contact is told in writing what happened and what was done.

  5. Resolution Engineer and SOC

    The account is re-secured and credentials are reset. Forensic evidence is preserved and an incident report is written. The ticket closes with a record of every action.

The promise behind the timeline

Two-minute average response, from 2025 SOC data.

Two-minute average response, 24/7, in writing.

These are published averages across real incidents. This sample is an illustration of the sequence, not a timed record.

Sample data. Real workflow.

An AI agent handles a ticket, with a person in the loop

Pick a sample ticket and step through the agent: what triggers it, what it decides, where a person must approve, and what it logs.

No customer data is shown. Every name, record and value below is synthetic. The agent is Helpdesk Tier-1 Triage, published as: AI-powered ticket classification and auto-resolution with a full audit trail for every decision. Sensitive requests are flagged before escalation.

Incoming sample support request

“Can you pull up the chart for patient Jane Q. Sample, MRN TEST-000001? She needs her results today.”

Illustration of “sensitive requests are flagged before escalation”. Not a product screen.

  1. Trigger Support request

    A sample user, sam@example.test, submits a support request. The agent picks it up.

  2. Classify Agent

    Category: records request. Priority: high. This is not something the agent should resolve on its own.

  3. Sensitivity check Agent

    Protected health information found: a patient name and a medical record number. The request is flagged before anything is escalated.

  4. Review step (illustrative) Person

    In this illustration the flagged request is held for a person, who approves or declines before it moves on.

  5. Result Agent

    After approval, the request is routed to the right role (role-aware routing), the PHI stays out of uncontrolled channels, and the requester is told it is in progress.

Audit-log entry (written for every decision)

agent
helpdesk-tier1-triage (sample)
category
records request
sensitive
yes, PHI flagged
review step
required, pending
entry
sha256:5e07...b3a2 (sample)

Want to see it on your own workflows?

Every demo above uses sample data. In 30 minutes we will walk through the same flows against your environment and your policies. You can also browse recorded examples by department on AI in Action.

Book 30 minutes