Financial Adviser Product Research Reports: Use case and AI Confidentiality

A licensee has to be able to show why a fund is on its approved list, and an adviser has to show she investigated the products that might suit the client. Both end up in a product research report: the issuer, the Product Disclosure Statement and Target Market Determination dates, how the fees are built, the strengths and risks, any benefit the issuer has offered, and a recommendation. Staff paste PDS extracts and fact sheets into ChatGPT or Claude to summarise them, and the report carries more than the extract does. This page marks every item in an invented product research report and shows the same note prepared on Ermos inside the practice.

The product research report

Written by an adviser or paraplanner and reviewed by the compliance manager when a fund is added to the licensee’s approved list or compared for a client. Section 961B(2)(e) of the Corporations Act lists a reasonable investigation into the financial products that might meet the client’s needs among the steps by which a provider can prove the best interests duty was met, and section 961D(1) says that does not require an investigation into every product available. Neither says what form the record of the research takes.

Product research note: Australian share funds on the platform
Harbourview Wealth Partners · prepared by · reviewed by · Thursday 19 November 2026

Why this note exists.

Scope and sources.

The funds.

Fees and costs, structure only.

Target market. The three determinations were read for a client seeking growth over seven years with a medium to high risk tolerance.

Conflicts and benefits.

Recommendation.

Limits.

An invented research note. The names, funds, dates and rankings are made up, and Ilsley Funds Management, Halberd, Halstead and the client are invented. Hover or tap any highlighted item for the analysis; the full list is below.

Product research report drafter

An Agent+ that prepares the first draft of a research report. It reads the product disclosure and fee documents the practice collects, fills your research template, and sets out the comparison plainly. Gaps in the material, and any provider relationship on file, are flagged.

What it does

  1. Reads the product documents and fee schedules collected
  2. Drafts the comparison in your research template
  3. Flags gaps and provider relationships to disclose

What stays with your people. The adviser or paraplanner writes the conclusions and the compliance manager reviews them. The approved list committee decides what goes on it.

An Agent+ is built once and anyone in the practice can run it. It works from your own documents and runs on Ermos Dominion, the AI unit in your office, or on Ermos Edge, your own instance in an Australian data centre.

What leaves the practice

  1. 1
    The practice’s own finding

    The licensee’s account of a shortfall on a file. Section 961B(2)(e) lists a reasonable investigation into the financial products that might achieve the client’s objectives and meet their needs among the steps by which a provider can prove the best interests duty was satisfied, and section 961D(1) says it does not require every product to be investigated. A candid admission about the quality of an adviser’s work, written for the licensee.

  2. 2
    A named client in a product document

    A name, an age, an amount and a date. A product note should hold no client detail, and each of these turns a document about funds into a document about a person. Personal information under the Privacy Act where the Act applies to the practice. Privacy Act guide.

  3. 3
    Sources and method

    Public documents and the practice’s way of collecting them. Nothing in it is about a client.

  4. 4
    The fund-by-fund findings

    Facts from public documents arranged into a judgement the licensee will rely on when it justifies a recommendation. An AI summary of a PDS can drop or misstate a detail, and the Code of Ethics says the duty of competence is ultimately personal and cannot be outsourced to others.

  5. 5
    Fees and costs, in structure

    The shape of each fund’s fees and a ranking. The figures are in the PDS and are public, and the ranking is the practice’s own comparison. Standard 7 of the Code of Ethics asks an adviser to satisfy themselves that the fees and charges a client pays are fair and reasonable and represent value for money.

  6. 6
    Conflicts and benefits

    A named sales director, a client event, a declined research offer and a register check. Standard 3 of the Code of Ethics says an adviser must not advise, refer or act where there is a conflict of interest or duty, section 912A(1)(aa) requires the licensee to have adequate arrangements for managing conflicts, and section 961J requires a provider to give priority to the client’s interests when they conflict with the provider’s or the licensee’s. A named person’s commercial dealings are also personal information about her. Privacy Act guide.

  7. 7
    The recommendation and the decision date

    What the practice intends to put on its approved list before the committee has decided, and when. Commercial information about the practice, and a recommendation that is not yet a decision.

  8. 8
    The limits paragraph

    Standard wording that research is not advice to a client and that section 961B(2) applies to each client’s advice. No client data.

  9. 9
    The named staff

    The preparer, the paraplanner and the reviewer. Personal information about employees, and it ties the note to the people responsible for it. Privacy Act guide.

  10. 10
    The target market reading

    A statement of how the design and distribution rules apply. Section 994A(1) lists providing personal advice and arranging to implement it as excluded conduct, and section 994E(3) does not apply to excluded conduct. It is a public rule with no client data in it, and a reader still has to check the reading against the Act.

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 Use cases in financial advice practices.

Common questions

Do ChatGPT and Claude use the information in a product research report?
Everything pasted or uploaded from a research note 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. This note names Dean Jolliffe and his $60,000, a finding about a client file and a recommendation the product committee has not yet decided.
Is it safe to paste a product research report into ChatGPT or Claude?
Only the public parts, and this note mixes them with private ones. 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. The Product Disclosure Statement and Target Market Determination extracts are public, but a named client, a benefit offered by Ilsley's sales director and an undecided approved list recommendation are not.
Does a ChatGPT Business or Claude Team plan make it safe for a product research report?
Not on its own. A business plan changes the default on training, but it does not change who receives the content. On a personal plan the user cannot choose where conversations are stored or processed. Some business plans let an organisation choose to store data at rest in Australia (OpenAI), which still leaves the content with the AI company. The OAIC's guidance also tells organisations to review carefully the terms and settings that will apply to their use of any AI product. For this note, a compliance finding about a named file and a recommendation made before the committee decides would still arrive at the AI company. Our comparisons with ChatGPT Enterprise and Claude weigh the trade-offs.
How can a licensee use AI for a product research report 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 note is drafted from the Product Disclosure Statement extracts and the practice's own research template. The adviser and the compliance manager check each summary against the document before the committee decides.
Does an adviser have to investigate every fund before recommending one?
No. Section 961D(1) of the Corporations Act says a reasonable investigation into the financial products that might meet the client's needs does not require an investigation into every product available. If the client asks the provider to consider a specified financial product, section 961D(2) says the investigation includes that product. The adviser still has to be able to show the investigation that was done.
Does a target market determination apply when an adviser gives personal advice?
Providing personal advice, and arranging for a retail client to apply for a product to implement that advice, are excluded conduct under section 994A(1) of the Corporations Act, so the distribution duty on regulated persons in section 994E(3) does not apply to that conduct. The determination can still be a useful input to the adviser's research.
Can an AI summary of a product disclosure statement go into the research file?
The Code of Ethics says an adviser may supplement their competence by accessing the expertise of others, but the duty of competence is ultimately personal and cannot be outsourced. A summary of a PDS is therefore a draft to be checked against the document. On an Ermos unit the extracts are summarised on hardware in the office, and the adviser still checks them.

See Dominion answering from your own documents.

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

Book a demo