AI usage policy

1.What this covers

Remote Desk includes AI features. This page explains which ones send your content to an AI provider, how that data is treated, what your organisation controls, and what we ask of you when you use them.

It supplements the privacy policy and terms of service.

2.Where your content goes

These features send content to an AI model provider chosen by the operator of the service, and only when you or a feature you use makes a request:

The AI Workspace, including chat, comparing answers from two models, multi-step jobs and the routing that decides what kind of answer a message needs. Your messages, the files and images you attach and the earlier turns of the conversation are sent. The Deskly assistant works the same way.

Meeting minutes. When a recording is turned into minutes, its audio goes to a speech-to-text provider (or a transcription service run by the operator), and the transcript goes to an AI model that writes the summary, decisions and action items.

Voice dictation in the message box, deep research and web search, Studio images, files that the AI creates for you, and reading of the PDFs and images you upload so the AI can answer about them.

AI employees and agents. When someone mentions an AI employee in a room, channel or message, or assigns it a work item, the relevant conversation or item is sent to the model so it can answer.

Memory. If memory is on, short notes drawn from your chats are kept so that later answers have context. The daily briefing can also use an AI model to write its summary.

Providers are listed by category on the subprocessors page.

3.No training on your data

Remote Desk does not train AI models on your data. Your prompts, files, recordings and transcripts are used to give you the answer or output you asked for, to keep the service working, and to measure usage and cost.

The AI providers your administrators connect have their own terms. Administrators choose which providers to enable, and should review those terms for the data they will send. See subprocessors.

4.What we keep, and who can see it

Conversations are saved to your account so you can come back to them, and you can delete them. For monitoring cost and reliability we keep records of AI requests, such as which model ran, how long it took and what it cost. The text of prompts and answers is stored with those records only where the platform's settings or your organisation have turned that on.

Organisations can give some people, such as admins, an oversight role that lets them review members' AI conversations. If your organisation uses this, it is disclosed in the product.

Retention periods are on the data retention page.

5.What your organisation controls

Owners and admins can turn the AI modules on or off for the whole organisation, add the AI providers and keys they want to use, limit which models people can use, and set a budget for each member, with an alert before the limit. For AI employees, the owner sets a monthly spend cap for the organisation.

AI employees have no access of their own: they can do only what the person who asked could do, and changes wait behind an approval card before they run. Pinning a model in the AI Workspace is each member's choice.

6.Using AI well

AI output can be wrong, out of date, biased or incomplete, and it can present invented facts with confidence. You are responsible for checking it before you rely on it, send it on or act on it, especially for legal, medical, financial, employment or safety decisions. Do not use AI output as the sole basis for a decision that significantly affects a person.

Do not put into a prompt, file or recording anything you have no right to share with an AI provider, and think before including secrets such as passwords, keys or other people's sensitive data. Do not use AI features to break the law, to deceive or harm people, to create content that breaks the acceptable use policy, or to get around limits on your role, budget or models.

Read the acceptable use policy for the full rules.

7.Ownership of outputs

As between you and Remote Desk, your organisation owns the content it puts in and keeps the rights it has in the outputs the AI produces from it. AI outputs may be similar to what other people receive, and the law on ownership of AI-generated material is unsettled, so we give no assurance that an output is protectable or free of third-party rights.

8.Questions

Questions about AI features and data: hello@thehivez.com.