Skip to content

Disclosure

How ASV makes money.

ASV is free, and it stays free because of referral bonuses. Here is exactly how they work, along with the things they never change.

Last updated
On this page7 sections

The short version

  • ASV is free to use, and so is applying to any legitimate platform.
  • Apply buttons use a referral link wherever we have one. If you join a platform through one, ASV may earn a bonus.
  • Bonuses are paid by the platform. They never come out of your pay.
  • No one can pay to be listed, ranked or reviewed on ASV.

How referral bonuses work

Many AI training platforms run referral programs. A member shares a personal link, and when someone joins through it and is accepted (or goes on to complete paid work), the platform pays the member a bonus. The referral links on ASV come from these programs.

Each platform decides whether a bonus is paid, how much it is and what earns it, and it changes those terms from time to time. We do not publish amounts: they vary, and they are not ours to promise.

The bonus is paid by the platform, from its own budget. It is never deducted from your pay, and joining through ASV costs you nothing.

What an Apply button does

A platform’s Apply button points to a short address on ASV (/go/ followed by the platform’s name, such as /go/outlier), which forwards you to the platform. Where we have a referral link for that platform, that is where it sends you. Where we do not, it sends you to the platform’s ordinary sign-up page. Either way, you apply exactly as you would have without us.

Apply buttons on job openings use /go/jobs/ followed by the opening’s name. They lead to our referral link for that role, or for its platform where one link covers every role, which opens the platform’s sign-up page rather than the role; for Mercor, micro1 and Handshake AI, our referral code is added to the listing’s own address, so you still land on the role itself. Without a referral link, they open the public listing. Each opening’s page says which, under its Apply button.

These links open in a new tab and are marked rel="sponsored" in the page’s code. That is the label search engines ask websites to put on links that can earn them money. The same goes for any sign-up link inside a guide that passes through the same address, so anyone can check which links are which.

ASV Projects involves no referral links at all. It is our own work, organized and paid for by ASV.

What referrals never change

  • No paid placements. No platform can pay to be listed, featured, ranked higher or described more kindly, and we do not publish sponsored posts or paid reviews.
  • One standard for every platform. A platform we have no referral link for is researched with the same care, and appears on the same terms, as one we do.
  • The drawbacks, too. Alongside the strengths, every profile sets out the downsides we can source: uneven work, demanding assessments, slow support.
  • Independence. ASV is not owned by, affiliated with or endorsed by any platform it covers.

How we choose platforms

A platform earns a place on ASV when it passes all four of these tests:

  • It is real. An identifiable company stands behind it, its terms are public, and it has a record of paying the people who work for it.
  • The work is AI training or research. It pays individuals to write, rate, review, code or apply expert judgment to train and evaluate AI models, or to take part in paid research studies.
  • It is free to join. Anything that charges you to apply, to train or to unlock work is left out.
  • It is relevant to our readers. Most platforms we list accept contributors from at least some African countries. Where one does not, as with DataAnnotation and Handshake AI for now, we still profile it because readers ask about it, and the profile says so plainly. Every profile sets out what we know about applying from Africa.

The platforms we feature first are the largest and most active, not the ones with the most generous referral terms.

How profiles are researched

Each profile starts from the platform’s own public material: its website, help center, terms, job listings and payout documentation. Where those are silent, we look for independent reporting we can check, and we leave out what we cannot.

  • Sourced. Every profile lists the sources it draws on.
  • Dated. Every profile shows when its facts were last checked, and every guide shows when it was published, and when it was last updated if it has been.
  • Careful with numbers. Pay appears as a range only where a reliable figure exists. Where it does not, we say it varies rather than guess.

We revisit profiles regularly, and sooner when a platform changes something that matters: pay, eligibility or how it pays out.

Corrections

Platforms change quickly, and we will sometimes be behind or simply wrong. When we find an error, or someone points one out, we check it against the platform’s own documentation, correct the page and update the date it was last checked. If a correction changes the substance of a guide, such as a pay figure, an eligibility rule or a payout method, we note it in the guide.

To report an error, email support@asvalley.com with the page and what is wrong.

If what you read here differs from what a platform tells you, trust the platform.