Skip to content
Features

How to control what AI agents can access and do

In BARITA AI the crew works inside the same permissions as the person it works for, so it can only read and change what that person is allowed to. What goes live, and anything that would leave your company, waits for a person to approve it.

Permissions for each person

The workspace owner invites people and sets what each one can see and do. Every request is checked against your workspace and the person’s permissions before any data is read or changed.

Approvals

Every app starts as a draft that you approve before it goes live. Anything that would leave the company, such as a message to a customer, waits in the Inbox for a person to approve it.

Companies are kept apart

Each company’s data is kept apart from every other company’s. Each AI task is given only the content it needs.

A record of what was done

The crew works through tools that check permissions and record what was done, so you can see what happened. Sign-in is protected against automated abuse, people can add a second step to their sign-in, and sessions can be ended.

How it works

  1. Invite people The workspace owner invites each person.
  2. Set what each person can see and do Permissions are set in People and access.
  3. Approve what matters Drafts and anything leaving the company wait for a person.

Common questions

Can the AI see everything in my company?

No. The crew works within the permissions of the person it works for.

Does the AI act without my approval?

What goes live is approved by you, and anything that would leave the company waits for a person.

Where can I read more about security?

The Security page describes how BARITA AI protects your company and how to report an issue.

Related answers

Tell the crew what your business needs

Describe it in your own words. You approve what goes live.

Get startedAll features