AI Exposure · Accounts
What Xero actually does with your data
Your accounts package now has an assistant called JAX. Xero says your data is not used to train its AI, and names the AI companies your questions travel to.
- Trains on your data
- Xero says no, on its JAX page
- Processed
- Your questions go to the US
- Can you turn it off
- No off switch, only roles
- Verified
- 20 July 2026, against Xero's published documents
What AI is in it?
- JAX, the Xero assistant. An assistant in your Xero navigation bar that you can ask about your accounts in plain English, and that can act on invoices, quotes and chasing payments. Xero's help articles label it beta. Its UK product page says JAX chat is available to all Xero subscribers and users, with nothing to switch on.
- Predictions in everyday bookkeeping. Bank reconciliation suggestions and similar features that learn patterns to speed up routine entry. These have been in Xero for years and are AI even though nobody called them that at the time.
- Newer features arriving by plan. Things like auto reconcile and smart document capture appear from the middle plans upward, several still labelled beta. They arrive in the product rather than as a decision anyone at your organisation takes.
Does your data train AI models?
Xero says no, on its JAX page.
Xero answers this directly. Its JAX page says: "No, JAX chat doesn't use your data to train the LLMs it uses." It adds that the outside AI companies it sends your data to do not retain it or train on it either, as part of its commercial arrangements with them.
Where that answer sits is worth knowing. It appears on Xero's JAX page and in its help articles rather than in the terms of use you agreed to, and Xero says it may change JAX's terms. Its privacy notice separately lists AI and machine learning among the ways it analyses collected data to develop new products and services.
So you have a clear published answer, and it is a published answer rather than a contractual one. If you are relying on it, keep a dated copy.
Where is your data processed?
Your questions go to the US.
Xero publishes the list of companies it passes data to, and the AI companies are named on it: Anthropic and OpenAI, both listed for conversational artificial intelligence services, both in the United States.
Xero hosts on Amazon Web Services on servers in the US, with standard contractual clauses covering the transfer, and processes some data in New Zealand, which the EU treats as an adequate country. I have not found a published statement covering JAX questions specifically, and there is no UK or European data residency option for them.
The summary for a UK charity: your accounts data already leaves the UK under safeguards, and a question asked of JAX goes to a US AI company.
Free versus paid: Included, but no free tier. JAX chat costs nothing on top of a Xero subscription, and there is no free tier of Xero at all, only a trial. Which AI features have reached your plan and your region is not published in one place, so the only reliable way to know is to look at your own Xero.
Can your organisation turn it off?
No off switch, only roles.
Xero's published answer to the opt out question is that you can leave JAX unopened if you would rather not use it. JAX does not run in the background: it does something when somebody asks it a question.
I have not found an organisation level switch documented anywhere.
The lever Xero does offer is user roles, because JAX can only reach what that person's permissions already allow. Roles are blunt here, since the same roles gate ordinary bookkeeping. You cannot take JAX away from your finance officer without taking Xero away from them too.
What does this mean for the people you support?
The thing worth noticing about an accounts assistant is that the question travels, not just the ledger.
"Which of our regular donors stopped giving this year?" names people in the asking, and that sentence is what goes to a US AI company. Your books are already a record of named donors, salaries and supplier contacts, but until now nobody was sending questions about them anywhere.
What to check this week
- Look at what has already arrived. Open Xero and see whether JAX is in the navigation bar and which AI features are on your plan. Whoever does your bookkeeping will know what has appeared recently.
- Save Xero's no training answer, with today's date on it. It is published on a web page rather than in your agreement, and web pages change without telling you. Screenshot the JAX FAQ and keep it with your policy, so you know what you were told and when.
- Decide who should have JAX, and accept the trade. The only real control is user roles, and those also gate ordinary bookkeeping. Work out who genuinely needs it, remembering that an assistant makes the books much easier to browse than they used to be.
Sources
Everything above is my plain English reading of what Xero publishes. The originals:
- Xero: Just Ask Xero (JAX), including the AI FAQs reviewed 20 July 2026
- Xero: Subprocessors, naming Anthropic and OpenAI reviewed 20 July 2026
- Xero: Terms of use reviewed 20 July 2026
- Xero: Privacy notice reviewed 20 July 2026
- Xero: Xero and GDPR reviewed 20 July 2026
- Xero Central: Use JAX (beta) reviewed 20 July 2026
Vendors change these documents quietly and often. If you spot something out of date, email me and I'll re-verify the entry.
What if the AI reading your accounts was yours?
This is not a hypothetical for Xero. I have already built it: a private assistant running in an AWS account that reads a charity's live Xero through its API, alongside three other systems, with every figure linking back to the real record. You keep Xero, and no outside AI company sees the question or the answer.
Here it is, asked for a full update on an appeal. The answer draws on live Xero records and the other three systems at once. The full demonstration is at chat across your systems.
See your whole toolkit at once
This page covers one tool. The AI Exposure Report Generator covers your toolkit: tick the tools your charity uses and get one report showing what the AI inside each of them does with your data, as a Word document you can table at your next meeting. Free, and nothing you tick leaves your browser.
The Report Generator, start to finish: tick your tools, read your exposure, download the report.
Read the other tools
The same questions, answered for every tool in the directory.
Your office suite
Microsoft 365 Copilot
The AI inside Word, Excel, Outlook and Teams. It can reach whatever the signed in person can reach.
Verified 2 July 2026 →
Google Workspace with Gemini
Gemini inside Gmail, Docs, Drive and Meet. Google says your content is not used to train models outside your domain. What it can reach is the longer answer.
Verified 3 July 2026 →AI chatbots
ChatGPT
The one your staff are probably already using. What happens to what they type depends entirely on whose account it is.
Verified 2 July 2026 →
Google Gemini
Google's chatbot, one tap away on Android phones, in Chrome and in search. On a personal account, the training switch is on until someone finds it.
Verified 2 July 2026 →
Claude
Anthropic's chatbot. The same personal versus organisational story as the others, with the training question put to you as a setting.
Verified 2 July 2026 →Meeting notetakers
Otter.ai
The notetaker that joins your meetings, and can invite itself to the next one. What it hears may help train its AI.
Verified 2 July 2026 →
Microsoft Teams
The meeting recorder your charity already owns. Everything it captures becomes a file in OneDrive and SharePoint, where the rest of the AI can reach it.
Verified 3 July 2026 →
Zoom AI Companion
The AI inside the video calls much of the sector runs on. Zoom's no training promise is unusually specific, and there is a reason it is worded that way.
Verified 2 July 2026 →
Fathom
The free notetaker that arrives one personal install at a time. The recordings live in the US, and the training answer has two halves.
Verified 3 July 2026 →Charity CRMs
Beacon
A UK charity CRM with AI on every plan since 2024. Beacon publishes where your data sits and how to switch the AI off, which leaves one question to ask them directly.
Verified 2 July 2026 →
Donorfy
A UK charity CRM where the AI mostly arrives through the side door: optional integrations. What you connect decides where your supporter data goes.
Verified 2 July 2026 →
CiviCRM
The open source CRM ships no AI of its own. What that is worth to you depends on which CiviCRM you are running: one your organisation hosts, or a hosted product somebody sells you. Everything below splits along that line.
Verified 20 July 2026 →
Salesforce
In much of the sector through ten free licences. The AI paperwork is detailed and mostly reassuring, with one default that runs the other way.
Verified 3 July 2026 →Accounts
Xero
Your accounts package now has an assistant called JAX. Xero says your data is not used to train its AI, and names the AI companies your questions travel to.
You are here
QuickBooks
Intuit's accounts package now has an assistant throughout. Its privacy statement says your information is used to train its AI models, and it says so in as many words.
Verified 3 July 2026 →Marketing and design
Mailchimp
Your supporter mailing list lives here, and Mailchimp belongs to Intuit now, which means the same small print as QuickBooks covers it.
Verified 3 July 2026 →
Canva
The design tool half the sector lives in, free through Canva for Nonprofits. Whether your uploads help train AI depends on whose account they sit in.
Verified 3 July 2026 →