TrueTheorem

AI Integrations

Put AI to workon the boring parts.

We add AI to the jobs that eat your team’s week: answering the same questions, typing up documents, sorting the inbox. Built into the tools you already use, and tested on your own examples before anyone relies on it.

Sound familiar? Here’s what we’d build.

Useful, and safe to rely on.

  • Tested on your examples

    Before launch we run it on a sample of your real questions or documents and show you where it gets things right and wrong.

  • A person stays in charge

    It suggests, a person decides. It never emails a customer, moves money or changes a record on its own: someone checks each one first.

  • Your data stays yours

    We use business AI services that do not train on your data by default, and we tell you which ones before anything is sent.

  • Running costs up front

    AI services charge per use. We estimate the monthly cost on your volumes before you commit, and the bills go to your own account.

Where it starts, fixed before we begin.

  • One focused job, agreed in writing before we start
  • Tested on a sample of your real documents or questions
  • Connected to one tool you already use
  • A walkthrough for your team when it goes live

Every project gets its own fixed price in writing before work starts, paid 30, 40, 30. Bigger jobs cost more, and the first reply includes a rough price.

AI Integrations

AI Quick-Start

from A$1,500

A chatbot that answers from your own documents, or an extractor that reads invoice data for your team to check.

Optional care plan after launch: monitoring, fixes and small changes for A$150 to A$300 a month.

Questions we get asked.

Will it make things up?

It can, which is why we build it to answer only from your own material, show its source, and say when it does not know. We test for this before launch and tell you what we found.

Do we need to change our software?

Usually not. We connect to what you already use, through its API or a simple export, and tell you in the first reply if something cannot be connected.

Which AI do you use?

Whichever suits the job: models from OpenAI, Anthropic or Google, or an open model on your own server if the data must not leave it. The proposal explains the choice.

What does it cost to run?

AI services charge by usage, so it depends on your volumes. We work out an estimate before you commit, and you pay the provider directly, with no markup from us.

Other services

Tell us what you’re stuck on.

A few lines is enough. We reply in writing within one business day, with a first read and a rough price.