Points test spreadsheets and AI data leakage: what a skilled migration scorecard gives away in ChatGPT or Claude.

Skilled migration agents keep a working spreadsheet of every client’s points: age, English, work experience, qualifications, partner and nomination, with a total beside each name. Staff upload it or paste it into ChatGPT or Claude to check the arithmetic, rank clients for the next invitation round or draft a status email. The sheet holds more than the scores. The notes column carries visa expiry dates, refusals and health remarks. This page marks every cell in a routine points sheet that the Code of Conduct and the Privacy Act bear on, and shows how the same sheet is checked on Ermos inside the practice.

The points test spreadsheet

Kept open all year and updated when a client’s circumstances change. It supports the practice’s advice on whether an expression of interest is worth lodging. Home Affairs gives 65 as the minimum points to be invited for subclass 189, 190 and 491; meeting it does not guarantee an invitation.

fx=SUM(C2:H2)
ABCDEFGHIJ
1ClientDOBAgeEnglishSkilled emp.QualsPartnerNom.TotalNotes
230101020105; ; ;
32510151550; ; ; ;
425101515015; ; ; ;
515202015100; ; ; ; ;
63010020100; ; ; ;
7010151050; ; ; ;
83020015100; ;
92510151055; ; ; ;
10Pass mark
PointsRoundsNotes

An invented points spreadsheet. The clients, employers and dates do not exist. The point values are the invented practice’s working figures and are not a statement of the current Home Affairs table, which changes. Hover or tap any highlighted cell for the analysis; the full list is below.

What leaves the practice

  1. 1
    Client names

    Eight named people in one file, with every score and note tied to them. Personal information, and the key to everything else in the sheet. Privacy Act guide.

  2. 2
    Dates of birth

    Age is derived from the date of birth, so the full date sits in the sheet. An identifier used for identity verification and fraud. Privacy Act guide.

  3. 3
    English test results

    Test names and scores. Personal information a client supplies to prove a points claim. Privacy Act guide.

  4. 4
    Employers and work history

    Named employers and years of experience, including a company that is itself the subject of an employer nomination. Employment information, and confidential to the employer in some cases. Privacy Act guide.

  5. 5
    Qualifications and institutions

    Degrees, trade certificates, universities and years. Personal information, and the evidence behind the points the practice claims. Privacy Act guide.

  6. 6
    Partner and family details

    Spouses and partners named, with their skills and English level, and a note about children. Personal information about people who are not the client. Privacy Act guide.

  7. 7
    Visa status, expiry dates and refusals

    The deadlines that drive each file, a bridging visa, an earlier visitor refusal, a refusal now before the Administrative Review Tribunal. A refusal history and an expiry date are what an adverse party would most want to know. Privacy Act guide.

  8. 8
    A spouse’s treatment

    A note that a client’s wife is in treatment. Health information, sensitive information where the Privacy Act applies, and bearing on the health requirement a visa may carry. Patient data guide.

  9. 9
    The total score

    How competitive each client is. The practice’s assessment, and a figure it must check against the current Home Affairs points table rather than trust a model to add.

  10. 10
    The pass mark and the sum formula

    A public threshold and a standard addition. No client data. Spreadsheet files can also carry hidden sheets and columns, cell comments, formulas that pull from other sheets and file properties, all of which travel with an uploaded file.

What happens in Frontier Models (ChatGPT, Claude)

  • The whole document is transmitted to the provider’s servers. On personal plans the user cannot choose where it is stored or processed, and for the major providers that has generally meant outside Australia, under that country’s legal process. Some business plans offer storage in Australia, but the provider still receives the content.
  • On consumer tiers, prompts can be retained and used to train future models.
  • The provider’s staff and systems can reach the content under the provider’s own policies.
  • Once sent, the practice no longer controls it. Retention and deletion happen on the provider’s terms.

The same document on Ermos

Ermos Dominion

The document is drafted on a unit in your office, from your own templates and the client file. Every marked item stays on the unit, which keeps working with the internet switched off. Nothing in it is transmitted to an AI provider, so there is no disclosure to analyse: the confidentiality work is the same as for your file server.

Ermos Edge

Your documents, the index and your access boundaries stay on your own instance in an Australian data centre. When a staff member asks for this draft, only their question and the excerpts they are permitted to see go to a named Australian provider, who is contractually barred from keeping the content or training any model on it. Nothing is processed outside Australia.

Frontier Models (ChatGPT, Claude)Ermos EdgeErmos Dominion
Where the document is processedProvider servers, generally offshore on personal plansYour instance in an Australian data centre; answers generated in AustraliaThe unit in your office
Who can read itProvider staff and systems, under policyYour staff, within the boundaries you set. The named Australian provider processes only permitted excerpts, under contract, and keeps nothingYour staff only
Is it keptOn the provider’s termsKept on your instance; the provider is barred from retaining itStays on the unit
Can it train a modelOn consumer tiers, yesNo, by contractOnly your own unit learns from it, on your hardware

Part of AI data leakage in migration practices.

Common questions

Do ChatGPT and Claude use the information in a points test spreadsheet?
Yes, whatever is pasted or uploaded from the sheet is sent to the AI company's servers and processed there to produce a reply. As at October 2026, whether a conversation on a personal plan (ChatGPT Free, Plus or Pro; Claude Free, Pro or Max) can be used to train the models depends on a setting each user controls. Business plans (ChatGPT Business and Enterprise, Claude Team and Enterprise) and the developer APIs sit under commercial terms, and OpenAI and Anthropic each say they do not train on that content by default (OpenAI, Anthropic). On either product, the thumbs up and thumbs down buttons can send a whole conversation back for training. The sheet lists eight named clients with birth dates, employers and visa deadlines, and one note records a spouse in treatment.
Is it safe to paste a points test spreadsheet into ChatGPT or Claude with the names removed?
No, because the other columns still identify each client. Removing names rarely makes a document anonymous. Information stays personal information while the person is reasonably identifiable, and the OAIC warns that information can be at risk of re-identification even when it has been de-identified or anonymised. A birth date, a named employer, a university and a visa expiry fit one person together, and a refusal now before the Administrative Review Tribunal narrows it further. The Office of the Australian Information Commissioner (OAIC) recommends in its guidance on commercially available AI products that organisations do not enter personal information, and particularly sensitive information, into publicly available generative AI tools, because of the significant and complex privacy risks.
Can I upload a points test spreadsheet to ChatGPT or Claude instead of pasting it?
An uploaded file leaves the practice as a whole file. This workbook has Rounds and Notes tabs beside Points, and a spreadsheet can also carry hidden columns, comments, formulas and file properties as well as the cells you meant to share. The OAIC's guidance also tells organisations to review carefully the terms and settings that will apply to their use of any AI product. One upload sends every client in the sheet at once.
How can a migration practice use AI for a points test spreadsheet without sending it to ChatGPT or Claude?
Ermos is built so the document stays under your control. On the Dominion AI unit it is drafted inside your office and none of it is sent to an AI provider. On Ermos Edge your documents and index stay on your own instance in an Australian data centre, and only the question and the excerpts the user is permitted to see go to a named Australian provider that is contractually barred from keeping them or training any model on them. The status emails and rankings are produced from the client file and the practice's own records. A registered migration agent checks every total against the current Home Affairs points table before relying on it.
Does an uploaded spreadsheet carry more than the cells I can see?
It can. A spreadsheet file may hold hidden sheets and columns, cell comments, formulas that point to other files, and properties such as the author. Uploading the file sends all of it. Pasting only the visible cells sends less, but a points sheet is still a list of named people with birth dates, employers and visa deadlines.
Can a model be trusted to total the points?
No. The agent remains responsible for the advice, and a model can add up wrongly or apply an out-of-date points table. Home Affairs publishes the points table and a calculator, and the threshold for invitations. Check every total against the current source, whichever tool produced the arithmetic.
Is the whole caseload a bigger risk than one client's file?
Yes. A single paste of the sheet discloses every client in it rather than one. Section 35 of the Code of Conduct stops an agent disclosing a client's personal information to a third person without that client's written consent, so one upload can engage the duty for eight clients at once.

See Dominion answering from your own documents.

A demonstration takes 30 minutes, uses no client data, and comes with no obligation.

Book a demo
Keep reading
AI data leakage in migration practices Resources The Privacy Act, the APPs and AI All professions