Your students are already using AI. The question is which one.
Quester is the research workspace built on decades of dissertation advising, and the only one whose refusals are written down. It absorbs the executional cost of scholarship between your meetings, then declines, in a sentence, to do the work your student has to defend.
The choice was never AI or no AI.
Every student you advise can reach a tool that will produce a plausible literature review in ninety seconds. Some are using one, and some are using one and not telling you, while a blanket prohibition cannot be enforced and mostly selects for the students willing to say they complied.
So the real question is which tool, and what that tool does when a student asks it to do the thinking for them. Most oblige, whereas Quester declines and says why, and not as a content filter you can argue your way around, since the refusals sit underneath every setting in the product.
It will not draft chapters, sections, abstracts, or extended prose to be used as the student's own writing. It will not cite works outside their library. It will not assert conclusions their data cannot support. It will not flatter them or compare them to anyone else. And it never acts without their approval.
Those five hold whatever the student configures, and apply even on direct request. They are the identity of the product, not a preference inside it. The reasoning is in the manifesto, and the exact behaviour is in the manual.
The part of advising that does not scale.
You have a fixed number of hours and a variable number of students. A meaningful share of those hours goes on work that is necessary but is not why you were hired: re-establishing where you got to, finding out what has been read since, and discovering mid-conversation that the question has quietly changed.
They arrive with a record, not a recollection
Quester logs what actually happened, across sources added, notes written, questions framed and models revised, day by day, and exports it, so a student hands you the last eight weeks instead of reconstructing them and you both start at the substance.
The methods conversation starts earlier
Research questions get scored against the project's own design rubric across research, sampling, measurement and analytic design, and the weak dimension is named, so students bring you a sharpened question instead of finding the problem at the proposal defence.
The corpus is inspectable
You can see what a library is made of in a single page: how much predates 2015, whether it is all one method, and where the authors sit, which beats reading ninety abstracts and is a quicker way to spot the review that has fallen a decade behind the field.
Drift becomes visible
Every revision of every research question is kept. When a project has moved, you can see what it moved from, and that makes for a far better conversation than "I thought we agreed on something else."
It is not a monitoring tool, on purpose.
No dashboard shows you a student's private project. Work belongs to the projects it sits in, and you see a project because you are a member of it. Share a workspace and you get a roster, roles, and a period view of what moved and who was active, but not their notes, their conversations or their drafts.
That is a decision, and it is also practical. A tool students experience as surveillance is a tool students keep their real work outside of, and you end up with a tidy dashboard and no insight. What you gain here is a student who thinks better between meetings, not the ability to watch them do it.
Equally honest: Quester does not replace you and is not trying to. It has read their corpus and knows nothing about their committee's politics, their funding cliff, or whether this is the year to push. That is the job it is explicitly not doing.
It does not stop being useful after the doctorate.
The structure suits dissertation-scale work, which is also the shape of a monograph, a grant-funded project, or a review you agreed to write in a weaker moment. Postdocs and faculty use it for exactly that, and several of Quester's longest-running users finished their PhDs years ago and never closed the workspace.
Same product and same plan, since nothing about it assumes you are a student, and the voice treats you as a serious researcher because the people who wrote it were tired of tools that do not.
A shared workspace changes the unit.
Departments, methodology labs and doctoral programs can run Quester as a shared workspace, with projects grouped by department, a managed roster, and a common structure every student learns once instead of inventing badly.
See what it refuses to do.
Free to open, with no time limit. Bring a paper from your own work, or send a student the link and read what it tells them.
Open Quester