RelayRelay
Back to app →

Acceptable Use Policy

This Policy sets out the things you must not do with Relay. It applies in addition to, and is incorporated by reference into, the Terms of Service. A breach of this Policy is a material breach of the Terms.

1. Purpose and scope

This Acceptable Use Policy (the "AUP") sets out the things you must not do with Relay (the "Service"). It applies in addition to, and is incorporated by reference into, the Terms of Service at /terms. A breach of this AUP is a material breach of the Terms of Service.

The prohibitions below are written using the construction "includes but is not limited to". They are illustrative, not exhaustive. The general principle is the one that controls: you must not use the Service for anything that is illegal, that we reasonably consider abusive or harmful to a third party, that exposes Relay or its upstream providers to material legal, regulatory, reputational or financial risk, or that we determine, in our sole reasonable discretion, breaches the spirit of this AUP.

2. Illegal use

You must not use the Service for any purpose that is unlawful in the jurisdiction in which the call originates, the jurisdiction in which the destination party is located, or the jurisdictions whose laws otherwise apply to the call. This prohibition includes but is not limited to use of the Service to:

2.1 Commit or facilitate a criminal offence.

Including without limitation fraud (by false representation, by failing to disclose information, or by abuse of position); obtaining services dishonestly; identity theft; computer-misuse offences; offences against the person; offences against property; offences relating to controlled drugs; or any offence under the Bribery Act 2010 or the equivalent law of another jurisdiction.

2.2 Make threats, harass, stalk or abuse a natural person.

Including without limitation threats of violence, intimidation, stalking, doxing, hate speech, content that depicts or solicits child sexual abuse material, or content that incites violence against a person or group on the basis of a protected characteristic.

2.3 Defraud, deceive or mislead a third party.

Including without limitation phishing, vishing, business-email-compromise or invoice-fraud calling, impersonation of a person or organisation, false claims of authority (for example, falsely claiming to be law enforcement, tax authority, bank, healthcare provider, or family member of the destination party), and any communication that is calculated to mislead the destination party into acting against their own interests.

2.4 Infringe intellectual-property rights or confidentiality.

Including without limitation soliciting trade secrets or other confidential information, infringing copyright or trademark, or disclosing information you are bound to keep confidential.

2.5 Breach sanctions or export-control law.

You must not use the Service to place a call to a person, entity or country subject to sanctions imposed by the United Kingdom, the European Union, the United Nations Security Council, or the United States Department of the Treasury's Office of Foreign Assets Control (OFAC), and you must not use the Service while located in a country subject to a comprehensive trade embargo. You represent and warrant that you and (so far as you reasonably know) the destination party are not on any such list.

2.6 Breach data-protection or privacy law.

Including without limitation calls placed in breach of the UK GDPR, the EU GDPR, the Privacy and Electronic Communications Regulations (PECR), or the equivalent law of another jurisdiction; recording a call without consent in a jurisdiction that requires all-party consent (see Terms §4.4); and processing biometric data of the destination party (including voice biometrics) without an appropriate lawful basis.

3. Improper or inappropriate use

You must not use the Service in any of the following ways, even where the use is not, on its own, unlawful. This list is illustrative, not exhaustive.

3.1 Emergency services.

You must not use the Service to contact emergency services, including any number that connects to a public-safety answering point (999, 911, 112, 110, and equivalent), or to contact lines that are not designed to handle calls of the kind Relay places (including suicide-prevention and crisis-support lines, which depend on a human-to-human connection that the AI agent cannot provide).

3.2 Synthetic-voice misuse.

You must not use, attempt to use, or instruct Relay in any manner that: (a) clones or generates a synthetic version of a specific natural person's voice, with or without that person's consent; (b) misrepresents the agent's nature as AI to the destination party (the agent's opening AI disclosure is required by Terms §13.1 and may not be suppressed); (c) attempts to make the agent adopt your name or personal identity (see Terms §13.2); or (d) processes biometric voice data of any person without an appropriate lawful basis and the person's prior consent where required.

3.3 Deepfake or impersonation calls.

You must not use the Service to impersonate any natural person, business, public authority, or other organisation. You must not use the Service to make a call that, taken as a whole, is likely to deceive the destination party as to your identity, the agent's identity, or the purpose of the call.

3.4 Election interference.

You must not use the Service to influence the outcome of an election, referendum, or other vote, including without limitation: automated calls to voters or election officials; calls designed to suppress or discourage voter turnout; calls that misrepresent polling-station information, election dates, voting procedures, or candidate identity; or calls that promote or denigrate a candidate, political party, or campaign. This prohibition is independent of, and stricter than, the general law on political communications.

3.5 Bulk or unsolicited commercial calling.

The Service is not designed for and may not be used for bulk outbound calling, telemarketing, market research, robocalling, predictive or auto-dialling, or any form of unsolicited commercial communication. You must not use the Service in a manner that breaches the US Telephone Consumer Protection Act (TCPA), the US Do-Not-Call Registry, the UK Telephone Preference Service or Corporate Telephone Preference Service, the UK PECR, the EU ePrivacy regime, or the equivalent law of another jurisdiction.

3.6 Premium-rate and high-cost numbers.

You must not use the Service to call numbers whose connection or per-minute charge is significantly above standard rates (including premium-rate numbers, international special-services numbers, and revenue-share traffic-pumping destinations). The destination-safety gate is configured to refuse such numbers; you must not attempt to circumvent it.

3.7 Adult content.

You must not use the Service to place calls to adult-services lines, sexually explicit lines, or any line whose primary purpose is the provision of pornographic content; and you must not use the Service to solicit sexually explicit content from the destination party.

3.8 Use against the destination party's wishes.

If the destination party asks the agent to stop calling, asks not to be called again, or asks for their number to be added to a do-not-call list, you must not authorise a subsequent call to that destination from the same account. You must respect a clear opt-out.

3.9 Use that interferes with the Service or third parties.

You must not: attempt to gain unauthorised access to the Service or to another user's account or data; probe, scan, or test the vulnerability of the Service or its upstream providers; reverse engineer, decompile, or disassemble the Service or any of its components; introduce or transmit malware, viruses, worms, ransomware, or other harmful code; mount a denial-of-service attack; scrape, harvest, or otherwise extract data from the Service by automated means; bypass any rate-limit, balance-gate, allowlist, or other technical control; or use the Service in any other manner that interferes with the Service's operation, its security, or the experience of other users.

3.10 Jailbreaking the safety guardrails.

Relay is configured with safety guardrails described in the "Relay will never" surface (the "Guardrails"). You must not author, share, or use prompts, Plan inputs, or instructions designed to cause the AI agent to act outside those Guardrails (commonly called "jailbreaking"). This includes attempts to cause the agent to make a binding commitment on your behalf, to authorise a payment, to share information you have not approved on the current Plan, to adopt a non-AI identity, or to take any action listed in Terms §13.

3.11 Resale or sub-tier use.

You must not resell, sub-licence, or make the Service available to a third party as a service of your own, and you must not use the Service to place calls on behalf of a third party other than yourself in a way that is concealed from the destination party. If the Service is at any time made available for team or business use, the account owner will be responsible for the compliance of every authorised user with this AUP.

3.12 Use in breach of upstream provider policies.

The Service depends on third-party providers identified in the Privacy Policy. Your use of the Service must comply with the published acceptable-use policies of those providers (including Telnyx's AUP, OpenAI's Usage Policies, and the equivalent documents of any other identified provider) to the extent that those policies apply to end users of services built on the provider's platform. Where this AUP and an upstream policy differ, the stricter of the two applies to your use of the Service.

4. Recording-consent specifics

Telephone-call recording is regulated differently across jurisdictions. The Service records every Call Run and discloses the recording at the start of the call. You are responsible for ensuring that recording the call is lawful in both your jurisdiction and the destination party's jurisdiction, including obtaining any additional consents required.

4.1 United Kingdom.

Recording a call to which you are a party for personal use is generally lawful in the UK, but recording a call in the context of an activity covered by the UK GDPR or PECR is subject to those regimes. Where a call falls within UK GDPR / PECR, your basis for processing the recording, and the destination party's right to be informed, must be addressed in accordance with the Privacy Policy and applicable law.

4.2 European Union.

The position varies by Member State. In some Member States, recording a call requires the consent of all parties; the agent's opening disclosure is one part of meeting that requirement, but the destination party's continued participation in the call is the operative consent. You must end the call if the destination party objects to being recorded.

4.3 United States — two-party-consent states.

In the following US states it is unlawful to record a call without the consent of all participants: California, Florida, Illinois, Maryland, Massachusetts, Montana, Nevada, New Hampshire, Pennsylvania and Washington (the precise rules vary; this list is a non-exhaustive guide). When calling a destination in one of these states, the agent's opening disclosure plus the destination party's continued participation is the operative all-party consent; you must end the call if the destination party objects.

4.4 Elsewhere.

For any other jurisdiction, you are responsible for confirming that recording the call is lawful and that the agent's standard opening disclosure is sufficient to obtain the necessary consent. Where it is not, you must not authorise the call.

5. Reporting abuse

If you believe a user of the Service is breaching this AUP, or if you are a third party who has received an unwanted call you believe was placed via the Service, please email abuse@relaycall.app with as much detail as you can provide (the date and time of the call, the calling number if you have it, the substance of the call, and any other information that may help us investigate). We aim to acknowledge abuse reports within two working days and to investigate promptly. Where a report identifies a serious or imminent risk to a person, we will treat the report as urgent.

6. Enforcement

6.1 Our enforcement rights.

We may, in our sole reasonable discretion and without prior notice, take any of the following actions on a single breach or a pattern of breaches of this AUP: (a) refuse to place a particular Call Run; (b) suspend your account, your token balance, or your access to specific Service features for a period we determine; (c) terminate your account with immediate effect and delete its contents in accordance with the Privacy Policy; (d) report the breach to law enforcement, an upstream provider, or another competent authority where we are required to do so or where we reasonably believe it is necessary to prevent serious harm; and (e) take any other action that we reasonably consider proportionate to the breach. We may also recover from you, as a debt, any reasonable cost we incur as a result of the breach.

6.2 No waiver.

Our failure to take enforcement action in respect of one breach does not waive our right to take enforcement action in respect of another, whether by the same user or any other.

6.3 No liability for enforcement.

To the maximum extent permitted by applicable law, we will not be liable to you for any loss arising from any action taken in good faith under this §6, including loss of access to the Service, loss of data, loss of profit or business, or any indirect, incidental, special or consequential loss. This is reflected in, and is subject to, the limitation of liability in Terms §10.

7. Changes to this AUP

We may update this AUP from time to time to address new patterns of abuse, new legal requirements, or new features of the Service. The current version is always at /acceptable-use with a last-updated date. Material changes will be notified to signed-in users in accordance with Terms §15. Minor clarifications that do not alter the substance of the prohibitions may take effect immediately on publication.

Last updated: 29 May 2026.