At a glance
StressLens turns an AI-research-ethics worry into one concrete, fixable change. Here's what's inside.
AI research ethics can be complex. Let's take it one step at a time.
No perfect plan needed. Tell me about your study in your own words — I'll ask a few questions, then we'll narrow it down to things you can actually fix.
Upload your study material
Have a protocol, ethics form, or paper? Upload a PDF (or .txt) and I'll read it, interpret the problem, and show you where it could lead.
Your document is only read to interpret it in this session. Files are not stored.
How it works
- Talk it throughI ask one question at a time — no forms, no jargon.
- Pick one of five issuesI turn what you said into five concrete things you could fix.
- See where it leadsGet a dated timeline for that one point, and the smallest change that bends it.
What are you working on?
Mirror asks one question at a time. Answer in your own words — no right answers.
I think I've got enough to name some concrete things you could fix. Want to see them?
Five things you could actually fix
Pick one to work on first. Each is a concrete step in your study — not a general category. Tap a card to see where it leads.
You don't know which role you'll get
Pick up to 3 fixes
Your team has limited time. Choose the three changes that best balance effort and risk reduction.
0 of 3 selected
One last step
Thank you
Your response has been recorded. You can close this page, or start again with a different issue.