# AI Agent Terms

URL: https://promu.ai/ai-agent-terms
Last updated: 2026-10-05

> Promu’s agents research, write and reply on your behalf. These terms explain who is responsible for what.

**AI Agent terms.**

Promu’s agents research, write and reply on your behalf. These terms explain who is responsible for what.

Last updated 5 October 2026.

## 1. What these terms cover

These terms apply to Promu’s AI features, including the Sales Agent, Reply Agent, Lead Finder Agent, AI SDR and automations (together, “Agents”). They form part of our [Terms of Service](https://promu.ai/terms).

## 2. You stay in control

Agents act on the instructions, limits and approval rules you set. By default, every email an Agent writes waits for your approval. You decide if and when an Agent may send or reply on its own, and you can switch that off at any time.

## 3. You are the sender

Emails written or sent by an Agent are sent in your name, from your inboxes. Legally, they are your emails, and our [Sending Policy](https://promu.ai/sending-policy) applies to them in full.

## 4. Review what Agents produce

AI can be inaccurate, out of date or inappropriate for the situation. Before you approve an email or let an Agent send automatically, check that research is correct, claims are true and the tone fits your brand. Promu shows the sources behind personal details so you can check them quickly.

## 5. Give clear instructions

Tell Agents what they may and may not say: for example topics to avoid, claims they must not make, prices they may not quote and when to hand a conversation to a person. Agents can only follow instructions you have given them.

## 6. Uses that are not allowed

- Using Agents to make or support decisions about people’s access to jobs, credit, insurance, housing, education or public services.
- Making Agents pretend to be a specific real person without their permission, or denying that a message was written with AI when someone sincerely asks.
- Instructing Agents to infer or use sensitive characteristics about recipients.
- Any use that breaches the law, including the EU AI Act, or our Sending Policy.

## 7. Safeguards are not a guarantee

Promu includes safeguards such as spam checks, sending limits, blocked topics and hand-over rules. They reduce risk, but they cannot catch every mistake, and they do not replace your own review.

## 8. Your data and the models

Agents use Your Data to work for you. We do not use Your Data to train AI models for other customers. How we process personal data for Agents is described in our [Data Processing Agreement](https://promu.ai/dpa).

## 9. Responsibility

Promu provides the tool; you decide how it is used. You are responsible for claims that result from how you configured or used Agents, and you will compensate Promu for third-party claims that result from use in breach of these terms. Our liability is limited as set out in the Terms of Service.