Acceptable Use Policy

Dialgem places real phone calls to real people with a synthetic voice. That makes misuse unusually harmful and unusually easy, so this policy is stricter than a typical software one. It forms part of the Terms of Service; breaching it can lead to suspension or termination.

You must not use the Service to:

• call or message anyone without the consent or lawful basis the law requires, or ignore a request to stop — honour opt-outs and do-not-call registrations immediately;
• make unsolicited marketing, “robocalls” or mass calls in breach of laws such as the US TCPA and Telemarketing Sales Rule, the UK PECR, or the rules of any country you call into;
• spoof or hide caller ID, use numbers you are not entitled to use, or evade carrier, STIR/SHAKEN or number-registration controls;
• call outside permitted hours, or at volumes or patterns that look like a harassment or denial-of-service attack on a number or network;
• place or route emergency-service calls, or operate in a way that could prevent a person reaching them.

You must not:

• deceive anyone about who is calling or why — including falsely claiming to be a human, a government body, a bank, a charity, a named person or any organisation you do not represent;
• deny being an AI when someone sincerely asks, or where the law requires you to disclose it;
• clone or imitate a real person’s voice without their documented consent, or use a voice in a way designed to make the listener believe it is someone else;
• run scams, phishing, vishing, fake prizes, fake debt collection, investment or romance fraud, or any attempt to obtain money, credentials or personal data by deception.

Agents must not be built or prompted to produce, or to facilitate:

• threats, harassment, stalking, hate or abuse directed at a person or group;
• sexual content involving minors in any form, or sexual content aimed at anyone who has not consented;
• incitement to or instructions for violence, self-harm, terrorism or serious crime;
• unlicensed regulated advice or services presented as professional, such as medical diagnosis, legal advice or financial advice, without a qualified person and the required licences;
• political persuasion campaigns that deceive about the speaker or the facts, or that breach election law;
• content that infringes others’ intellectual property or privacy.

Collecting payment-card, health or other sensitive data through an agent is your responsibility and must meet the standards that apply (for example PCI DSS and HIPAA). We do not currently hold those certifications for you to rely on; do not route such data through the Service unless you have agreed that in writing with us.

You must not:

• probe, scan or test the Service’s security without our written permission (see below for good-faith reporting);
• circumvent rate limits, concurrency caps, billing or access controls;
• share or leak API keys, or use another customer’s account or data;
• send malware, or interfere with or overload the Service, our networks or our providers;
• scrape the Service, or use its outputs to train a competing speech or voice model in violation of our Terms or the providers’ terms;
• resell access in a way that hides the real end customer from us.

We may investigate suspected breaches, review call metadata (and, where the customer’s settings and the law allow, recordings) to do so, remove content, block numbers, pause campaigns, suspend or terminate accounts, report unlawful conduct to carriers, regulators or law enforcement, and cooperate with their requests. We will try to give notice and a chance to fix a problem first, but where there is risk of harm, fraud or legal exposure we may act at once. Credit on an account terminated for breach is not refundable.

If you suspect misuse, or received an unwanted call from an agent running on Dialgem, report it to abuse@dialgem.com with the number that called you and the time. To report a security vulnerability, write to security@dialgem.com; we welcome good-faith research that avoids harming users or accessing data beyond what is needed to show the issue, and will not pursue it in return.

This policy may be updated as Achroweb Solutions learns of new abuse patterns; the date on this page shows the latest version.