Professionals are increasingly asked to research a topic as part of their job: a market scan before a proposal, a regulatory check before a decision, the evidence behind a recommendation. AI makes that quick, but a quick answer is not the same as a sound one. The difference is method: the things a careful researcher does that an AI does not do on its own. AIGroundwork writes those methods as prompts, organised into four steps, and you use them at whatever scale the work needs. Reach for a single prompt in a chat to fix one thing, or install the full set as an instruction for a project or an agent, so the method runs on every task without being asked for.
Four common situations, each leading to the step that deals with it.
S1 I don't know what I'm really asking to Frame S2 I'm drowning in sources to Gather and verify S3 I've got material but no findings to Analyse S4 I need to check this holds up to ConcludeIt helps to see research as four steps, each resting on the one before. They are how AIGroundwork is organised, so finding your way around is also how you learn the shape of the work.
Turn a topic into a question you can actually research, before any searching starts. An AI will not question a badly framed topic. A researcher reframes it first.
Collect material and work out what can be trusted, one item at a time. This is where an AI on its own falls furthest short: it reads the top few results and stops.
Produce findings that are not sitting in any single source. The core move is combination: putting two things side by side and reading what comes out.
Test the finding and put it in a form you can trust. Search for evidence against it, state what the checked material supports, and show what it rests on.