Approved sources for AI: how to choose them
6 minutes read

The open web holds everything at once, good and bad. Approved sources are the specific material a team has chosen to trust, and choosing them well is most of the work.
An approved source is external material a team has deliberately chosen to trust for AI-assisted work: a specific report, standard, or reference site someone reviewed and added on purpose. The open web is different. It contains everything at once, good and bad, current and stale, and nobody has decided whether any of it belongs in this firm's work.
The difference seems small until someone sends a client a draft with a claim that traces back to a content farm article that happened to rank for the right search term. Approved sources prevent that. A good set comes less from finding more material than from deciding in advance what counts.
Why an AI default to the open web causes problems
The open web is a poor default source because nobody chose any of it for this firm's work. A general web search returns whatever ranks, and search engines rank pages on volume and recency signals, not accuracy or relevance to a specific engagement. A model drafting from that pool cannot tell a regulator's own guidance page from a blog post summarising it three steps removed, where each retelling changes more of the original's details.
At first nobody notices. A draft built this way still reads fluently. The problem is traceability: when a reviewer asks where a claim came from, "the model searched the web" is not an answer anyone can check in the time a review allows. The reviewer either trusts it or restarts the research, and restarting is slower than if the draft had given no source at all.
What qualifies a source
A source belongs in an approved set if it meets a few specific criteria.
| Criterion | What it means | Example |
|---|---|---|
| Traceable | Someone can open the exact document and check it | A named report or standard, not a search result |
| Chosen for this work | Reviewed and added for a specific kind of task | The regulator's own guidance, not a summary of it |
| Access-controlled | Only what the firm added is in scope by default | A maintained list, not an open web search |
| Kept current | Someone owns refreshing it before it goes stale | A quarterly check on anything tied to a live standard |
Teams skip the second row most often. It is tempting to approve a source because it looks credible in isolation, without asking whether it fits the work it will be used for. A well-regarded industry outlet is a fine source for a market trend and a poor one for a legal obligation. Build the list task by task, not as one generic "trusted sites" folder.
A worked example
Say a firm is drafting a compliance summary for a client, and the question is what a specific regulation requires. The approved source is the regulator's own published guidance: a document with a version number and a date, which the reviewer can open and match against a sentence in the draft.
The unapproved default is a general web search for the regulation's name. It returns the regulator's page, three law firms summarising it with their own framing, and a two-year-old forum post that predates an amendment. A model pulling from that mix cannot weight the regulator's page above the outdated post. Both look equally citable in the results, and whichever one the model drew on first shapes the sentence in the draft.
The fix is a decision made before drafting starts: this kind of work uses only the regulator's own material, and the reviewer brings in anything else deliberately instead of leaving the model to find it.
How to build and maintain an approved source set
Start with what the team already checks by hand: the reports, standards and reference sites an experienced person would open without thinking. That list is usually shorter than you expect, and it is the right starting point because it rests on judgement the firm already trusts.
- Write down the sources a senior person on the team relies on for this kind of work, not a generic "good sites" list.
- Group them by the kind of work they support. A source that is right for market context is often wrong for a compliance claim.
- Assign an owner for each group, someone whose job includes noticing when a source goes stale or a newer edition replaces it.
- Set the default for that kind of work to draft only from the approved set, and make pulling in anything else a visible, deliberate step instead of something the model does quietly.
- Review the set on a fixed cadence. A source that was current a year ago may not be now, especially for anything tied to a regulation or a fast-moving market.
You do not need an engineer for any of this. The person who owns delivery already decides informally which sources the firm trusts. Writing that down and keeping the AI within it makes the judgement repeatable.
Where approved sources fit in the rest of the work
Choosing approved sources is the input side of a process that continues into review. With the right sources, reviewers question claims less often, but they still check them, because even a claim from an approved source can be stretched beyond what the source says. The guide to telling a sourced claim from an assumption covers that check in detail and assumes the source was a reasonable one to draw from. The approved list is where that decision gets made in advance.
Approved sources are also one piece of a firm's connected knowledge, alongside its own prior work, its methods, and permitted client material. The connected knowledge guide covers how all four fit together and connect to the tools and data a firm already uses, under the access rules those tools already have.
The short version
An approved source is something a team chose on purpose, for a specific kind of work, and can point to directly. The open web is no substitute for that list, however good any single result looks. Building the set takes effort up front, but a firm that spends an hour deciding what counts spends far less time later tracing a claim back to something nobody would have chosen.
