Separating Internal AI Tools From Customer AI Features
AI shows up in two very different places at Jag Journey. Our own team uses AI tools to run the platform: drafting replies to support tickets, preparing reports, reviewing code and helping new organizations get set up. Customers use AI features inside their own accounts: drafting pages, writing posts, getting help answers and more.
Those two uses look similar from a distance, since both involve sending a request to a model and getting text back. Underneath, they differ in who is allowed to use them, who pays and how they are audited. We keep them deliberately separate.
Different users, different access
Our internal AI tools are available only to authorized central administrators, and only in our central administration area. Customers, including organization administrators, never get access to them. Customer accounts use the customer AI features built into their admin area, which work within that account's data and permissions.
That separation protects both sides. Customers cannot reach tools that might expose information across accounts, and our internal tools do not wander into customer workflows unannounced.
Different billing
Customer AI features are paid with the account's AI credits, using measured usage and clear estimates, as described in showing AI costs before customers spend credits. Our internal AI usage is our own cost. It is never charged to a customer.
A narrow, documented exception
There is one carefully bounded place where the two worlds meet. When our team helps a new organization set up their website during onboarding, an authorized team member can use internal AI capacity inside that organization's account to draft their site. That help is not charged to the customer.
The exception has strict conditions. It applies only to onboarding work, only while a verified team member is helping that specific organization, it rechecks the team member's access before queued work runs and every request is recorded. Ordinary customer use of AI in the same account still uses the customer's own credits.
Everything is audited
Every prompt and reply from our internal AI tools is recorded in an audit log, along with who used it and for what. That record exists so we can always answer how a piece of internal AI work was produced and by whom. Customer AI requests are recorded with their measured usage and charges, so organizations can see what they spent and why.
Drafts stay private until a person acts
When our support team uses internal AI to draft a reply to a customer's ticket, the draft is private to the team member. It is sent only when that person reviews it and chooses to send it as their own reply. Customers receive answers that a real person stands behind.
Registered like everything else
Both internal and customer AI uses are listed in our AI usage registry, with their purpose, users and billing rules. An unregistered use is refused before it runs. The registry is what makes the separation enforceable rather than a matter of good intentions. You can read more in every AI call is registered before it runs.
Public visitors get a third, separate path
There is also a third audience: visitors to our public website who ask a question before they are customers. Their help path is separate again. It can select a reviewed public help article and show it with a citation, or create a ticket for a person to answer, and it never gives visitors access to internal tools. A failed answer hands off to a person rather than quietly switching to a different AI service.
Three audiences, three paths, each with its own rules. Keeping them distinct is more work than building one shared chat box, and it is the only way to be confident about who can see what and who pays for what.
Signed in, and accountable
Internal AI tools can only be used from a team member's own signed-in administrator account, and every request is recorded with that person's identity. When something was produced with AI help, the record shows exactly which person was responsible.
Why the separation matters
Blurring internal tools and customer features creates confusing bills, unclear responsibility and potential privacy problems. Keeping them separate means customers pay only for what they use, our team can work efficiently without touching customer budgets and everyone can see who did what. It is a structural choice that supports trust in everything else we build.
Frequently asked questions
Can customers use Jag Journey's internal AI tools?
No. Internal AI tools are limited to authorized central administrators. Customers use the AI features built into their own accounts.
Am I charged when your team uses AI to help me?
Onboarding help from our team is not charged to your account. Your own use of AI features is billed with your AI credits.
Are AI-drafted support replies sent automatically?
No. A team member reviews each draft and chooses whether to send it as their own reply.
Related Posts
Every AI Call Is Registered Before It Runs
AI features cost money and touch customer data, so we keep a registry of every one. Here is how a call that is not on the list is refused before anything is spent.
Letting Customers Get Paid Directly With Connected Accounts
When an organization sells through our platform, the money should land in its own account. Here is how connected accounts, platform fees, refunds and disputes work.
Failing Closed: When the Safe Answer Is No
When a check cannot be completed, software has to choose between allowing and refusing. Here is why we design money, access and AI features to refuse until they are sure.
Using AI Moderation as a Signal, Not a Judge
AI can spot harmful content quickly, but it also makes mistakes. Here is how we use moderation results to alert people, and where narrow automatic action is justified.
How We Use AI Coding Assistants With Human Review
AI coding assistants help us move faster, but they never decide on their own what ships. Here is how we pair them with written rules and human review.
Comments (0)
No comments yet. Be the first to share your thoughts!