Responsible AI
AI that shows its work.
GradsHQ uses AI to research and analyse — never to invent. Here's what that means in practice, not just in principle.
We never fabricate data
If a tuition figure, deadline, or eligibility detail can't be confirmed from an actual source, GradsHQ labels it "Not currently verified" instead of guessing. This is enforced in the product itself, not just stated here.
Fit scores are signals, not predictions
A programme fit score or scholarship eligibility score reflects alignment with your profile and published criteria. It is never presented as a probability of admission or funding — that language is deliberately excluded from every score we show.
You stay the author
AI coaching (as it rolls out) asks questions and gives feedback. It doesn't write your personal statement, invent achievements, or answer application questions on your behalf.
Sources are shown, not hidden
Every programme and scholarship we surface carries a source link and a verification status, so you can check our work rather than take it on faith.
Scoped research, not open-ended scraping
AI research is deliberately restricted to your chosen target countries — enforced twice, once in what we ask the model and again by discarding anything outside that scope before it's saved, regardless of what comes back.
Minimal data, short retention
Resumes are processed to help complete your profile and then deleted within 48 hours. We don't hold onto sensitive documents longer than they're useful.