Question 1

Does the AI take actions on its own?

Accepted Answer

Only if you tell it to. Autonomy is a three-level dial per action type: Recommend (AI suggests, you execute), Approve (AI prepares, one click ships it), Autonomous (low-risk, reversible actions run under policy). Everything lands in an audit log.

Question 2

What happens when the AI is not sure?

Accepted Answer

It says so. Every prediction ships with a confidence score and the factors behind it. Below threshold, nothing is asserted and nothing executes — the item is surfaced to a human as a normal task.

Question 3

Do you train models on our data?

Accepted Answer

No. Your data never trains foundation models. The system gets better through your organization's own revenue memory — patterns, outcomes and intervention history that stay inside your tenant.

Question 4

Do we have to migrate off our CRM or helpdesk?

Accepted Answer

No. The OS sits on top of the stack you run. Salesforce and HubSpot connect via OAuth; support, billing and warehouse adapters ride the same contract. Swap a tool later, keep the brain.

Question 5

Can we adopt just one module?

Accepted Answer

Yes — that's the design. Most teams start with Pipeline Management, Deal Intelligence or Customer Experience, prove value, then widen. Every module runs on the same core, so each addition compounds the last.

Question 6

How is this different from our BI dashboards?

Accepted Answer

A dashboard shows you a chart and stops. The OS states the finding, explains why with evidence, recommends the action, and executes it with approval — then measures whether it worked.

Question 7

How do you handle security?

Accepted Answer

Encryption in transit and at rest, OAuth-only connections, strict tenant isolation, role-based access from CEO to rep, full audit logging, and SOC 2 / GDPR-ready architecture from day one. Documentation available under NDA.

Straight answers.