Privacy
This site is a signpost to AI-training platforms. We don't ask for your name, email address or any other personal details, and we don't use cookies. We do count how far visitors get through the check, which is explained below.
The suitability check
The four-question check runs entirely in your browser. Your answers are held in memory only while the page is open. They are never sent to us or to anyone else, and they are gone as soon as you close or reload the page.
Counting how far people get
So we can tell whether the check is worth running, we record how far each visit gets: that a visit started, which of the four questions it reached, which of the two result screens it ended on, and whether a platform link was clicked.
Each record holds a random identifier created when you open the tab and thrown away when you close it, the advert that brought you here if there was one, and the two-letter country code your connection resolves to. It is not a cookie and it cannot follow you to another site or another visit. Your answers themselves are never included — only how far through you got — and nothing in a record identifies you. These counts are stored in our own database, hosted by Supabase in London. If you arrived from a Meta advert, three of these steps are also reported to Meta, which is explained below.
If you arrived from a ChatGPT advert
We advertise on ChatGPT. When you arrive from one of those adverts, the link carries a click identifier from OpenAI. We keep that identifier in your browser's local storage for 30 days and use it to tell OpenAI two things: that you viewed our page, and whether you clicked through to a platform. Each report is a small image request to OpenAI, which will see your IP address as part of it. We don't send your quiz answers or anything else. If you didn't arrive from an advert, nothing is sent to OpenAI. OpenAI's own privacy policy covers what they do with it.
If you arrived from a Meta advert
We also advertise on Facebook and Instagram. When you arrive from one of those adverts, the link carries a click identifier from Meta. We use it to tell Meta three things: that you started the check, that you reached the "You're a good fit" result, and that you clicked through to a platform. The "Not quite yet" result is never reported. Each report is sent from our server rather than your browser: there is no Meta pixel on this site and no Meta cookie is set. A report carries the click identifier, the time, the address of the page you were on, and the IP address and browser user-agent string your request arrived with. Those last two are read from the request and passed straight to Meta; we don't keep them, and our own database still holds only the per-tab identifier, the step reached, the advert that brought you here and the country code. We don't send your quiz answers or anything else. If you didn't arrive from a Meta advert, nothing is sent to Meta. Meta's own privacy policy covers what they do with it.
Links to other platforms
The "See open roles" buttons take you to Mercor and micro1. Once you're there, their own privacy policies apply. If you're hired through one of our links, the platform may pay us a referral fee. That doesn't change what you're paid.
Hosting
The site is hosted on Vercel, which may keep standard server logs (such as IP address and request time) for security and operational purposes.
Questions
Email hello@aitrainingroles.com.