The short version
Shravan.ai places and answers real phone calls to real people. That is a privilege with rules attached, and most of them come down to one thing: the person on the other end should always know who is calling, why, what they are speaking to, and how to make it stop.
What the platform will not be used for
You may not use Shravan.ai to impersonate a person, a company or a government body; to run a scam, a phishing campaign or any attempt to obtain money, credentials or personal details by deception; to harass, threaten or repeatedly call someone who has asked you to stop; to make debt-collection calls that misrepresent who is calling or what is owed; to call emergency numbers; to distribute malware; or to break the law of the place you are calling into.
You may not configure an agent to claim it is human when a caller asks, to conceal that a call is recorded where it is being recorded, or to promise something your business does not actually offer.
What the platform does about it
These are not only terms. The product enforces the following, and a workspace cannot turn them off:
- The agent identifies itself as an AI at the start of every call, and where the workspace records calls, it says that too. A call script that omits either is refused at publish time, and the runtime says the missing sentence itself if one gets through.
- “Don’t call me again” is honoured automatically. A caller who asks not to be called — in English, Hindi or Hinglish — is added to that workspace’s do-not-call list after the call, with the words they used recorded against the entry. Numbers on that list are refused before any call is created.
- Campaign calling is confined to daytime hours, 09:00–21:00 IST by default, in the workspace’s own timezone. Outside the window a campaign simply does not dial.
- Asking the agent to reveal its instructions, or to hand over a credential, is refused, and a reply that would repeat a credential is redacted before it is spoken.
Calling rules are yours to satisfy
The controls above help; they do not make a campaign lawful. You are responsible for having a lawful basis to call each number you upload, for honouring national do-not-call registrations that apply to you, and for whatever registration, consent or content rules your regulator imposes on the kind of calls you are making. Promotional calling in India in particular carries obligations that depend on how you are registered, and they sit with you as the caller. If you are not sure which rules apply to your campaign, get advice before you dial, not after.
Keep a record of why each number was called. If a complaint arrives, that record is what answers it — ours will show what was said, not why you were entitled to say it.
Recording
Recording is off unless you switch it on. If you switch it on, the agent discloses it at the start of the call and you are responsible for the consent rules of the places you call into, which are not the same everywhere. Recordings are kept for a set period and then removed; tell us if your own obligations require a different period.
Children
Shravan.ai is built for a business to speak with the adult who contacts it — a parent, a guardian, an adult student. We do not provide, and do not claim to provide, a mechanism for verifying parental consent. Do not configure a workspace to collect children’s personal data until your organisation has the consent it needs, and do not rely on us to have obtained it. Where a caller appears to be a child speaking for themselves, the platform flags the call and keeps it out of quality sampling.
We do not use anyone’s call data for advertising, and we do not profile callers for any purpose other than producing the call summary and analytics your own workspace sees.
Fair use of the API
Do not attempt to bypass rate limits, probe other customers’ workspaces, or use the sandbox to avoid paying for real calls. Automated testing of your own workspace is welcome; load-testing the shared platform without arranging it with us first is not.
If something goes wrong
We may suspend a workspace that is causing harm to people being called, to the platform, or to our carriers’ standing — we will tell you why, and we will restore access when the cause is fixed. Persistent or deliberate breaches end the contract.
To report misuse — a call you received from a business using Shravan.ai, or a workspace breaking these rules — write to hello@shravan-ai.tech with the number that called you and roughly when. Where the matter is a security report rather than a misuse report, write to security@shravan-ai.tech instead.