Acceptable Use & AI Disclosure

Last updated: July 30, 2026

This page explains how Acumen's AI components behave, what they are and are not permitted to do, and what is expected of a business using the Service. It forms part of Acumen's Terms of Service.

1. The AI Says What It Is

Anyone speaking with the Service, whether by chat, form follow-up, or voice call, is told they are speaking with an AI assistant before they are asked anything. On a recorded call, the recording is disclosed at the same time. The Service is never presented as a member of your team, and it does not deny being AI if a customer asks directly.

2. It Answers Only From What You Approved

The Service responds using the services, pricing, policies, and process information you approved during configuration. It does not invent facts about your business. It does not guess at a price, a timeline, or a promise your business has not made. Anything outside your approved information goes to a person on your team rather than being answered.

3. You Decide the Boundaries

You decide, and confirm before launch, what the Service may answer, what it may book on its own, what it must refuse to engage with, and what must be handed to a person. These are configuration decisions you can revisit at any time, not a one-time setup step.

4. What Always Goes to a Person

Regardless of your configuration, the Service is built to route the following to a member of your team rather than resolve them itself:

• Complaints and disputes
• Expressions of distress, or any indication of a mental health or personal safety concern
• Situations involving an immediate risk to life, health, or property
• Requests falling outside your approved services or answers
• Anything the Service is not confident it can resolve correctly

The governing rule is that uncertainty escalates. The Service is built to hand over rather than force an outcome.

5. Emergencies

The Service is not an emergency service. It cannot dispatch emergency responders and may be unavailable at any time. It must never be configured or advertised as a substitute for emergency services or for a staffed emergency line.

If your trade receives enquiries that can involve immediate danger, such as gas leaks, carbon monoxide alarms, fire, flooding, structural failure, or medical distress, you are responsible for telling Acumen what those situations look like in your business and how you want them handled. Acumen will configure the Service to recognize them, direct the customer to emergency services or your emergency line, and escalate to your team.

6. Prohibited Uses

You may not use, or ask Acumen to configure, the Service to:

• Operate a business in a category Acumen does not serve, as listed in the Terms of Service.
• Present the Service as a human employee, or instruct it to deny being AI when asked directly.
• Impersonate a specific real person, or generate synthetic audio of a real person without that person's consent and clear disclosure.
• Collect, request, or store information from customers beyond what is reasonably necessary to deliver your services, including protected health information or financial account credentials.
• Contact anyone your business does not have a lawful basis to contact, or continue contacting someone who has opted out.
• Send content that is illegal, deceptive, or that promotes sex, hate, alcohol, firearms, or tobacco in violation of applicable law.
• Give legal, medical, financial, or other regulated professional advice to your customers.
• Record calls where doing so would be unlawful in the jurisdiction where the customer is located.
• Reverse engineer, extract, or misuse the underlying platform, prompts, or configuration outside your own account.
• Adversarially attack, overload, or attempt to manipulate a live, customer-facing instance of the Service in a way intended to cause malfunction or harmful output. Testing within Acumen's own demonstration environments is expressly permitted; those exist to be tested.

7. Opt-Out Must Stay Available

Where the Service sends text messages on your behalf, recipients can stop them at any time by replying STOP, receive one confirmation that they have been unsubscribed, and receive nothing further unless they reply START. Recipients may reply HELP for assistance. These mechanisms must remain enabled on every messaging channel the Service operates for you, and may not be disabled on request.

8. Human Oversight

The Service operates under your supervision. You are expected to review what it is configured to say before it goes live, and to monitor its behavior afterward. Acumen monitors the platform for faults and misuse, and may suspend a configuration that appears to be operating unsafely, unlawfully, or outside these rules.

9. Reporting a Concern

If the Service responds in a way that appears incorrect, unsafe, outside its approved scope, or inconsistent with these rules, report it to [email protected] so it can be corrected. Acumen treats reports of unsafe or non-disclosing behavior as urgent.