What happens at an in-person AI research study? A session walkthrough

What happens at an in-person AI research study? A session walkthrough

Ways to earn

Article by

Mindrift Team

Mindrift Team

An in-person AI research study is a short, paid session where you take part in a recorded activity, usually a conversation, at a study site. You register online, verify your ID, sign a consent form, and book a slot. A session may last as little as 90 minutes or occur over a few days.

If you’ve ever heard about focus groups, you understand the basic premise of an in-person study. But instead of providing feedback on a new product or idea, contributors are filmed doing ordinary things, including having conversations, using electronics, cooking, or doing simple tasks. These recordings are then used to train AI models on how people operate in the real world, from using different accents and slang to the various unique ways a human can make a cup of tea.

Traditionally, AI training involved clicking buttons or typing responses on a computer. In-person AI training studies are relatively new, thanks to the growth of the robotics field. Because this is an emerging project type, contributors might not know where to find these opportunities or how they work.

How an AI research study differs from in-person focus groups

The format looks familiar, but three differences change what you should expect.


Traditional focus group

In-person AI research study

What's being studied

Your opinions about a product or idea

How people naturally talk, interact, complete tasks, or use items

What's recorded

Usually notes, sometimes audio

Typically both video and audio of the session

What you say

Your honest views

Conversational studies focus on ordinary topics

Preparation

Sometimes a screener survey

Online onboarding, ID check, and consent before you join

The topics are deliberately ordinary. For example, in one research study in Santa Clara Valley, CA, contributors are asked to talk about everyday things like having lunch with coworkers or hanging out with friends. There are no sensitive or personal topics.

Before the session: Sign-up, screening, and consent

Most of the screening happens before you get to the study site. While the process can differ depending on the type of project, a typical registration timeline might look like this:

  • Register: Create a free Mindrift account, or log in if you already have one. Make sure your profile shows the country you live in, or the project won't be visible to you.

  • Pass screening: Answer a short questionnaire or take a brief online interview so we can check you're a good fit.

  • Complete onboarding: Learn what the study involves and sign the consent form. Some projects also ask you to verify your ID online, others check it at the study site.

  • Book your slot: Once you're approved, we'll send you a booking link. Pick a date, time and location that work for you. Some projects offer several options, others only one.

  • Set up payouts: Your payment goes to your Mindrift balance after the session. To withdraw it, you'll need a connected Payoneer, PayPal or bank account. Set it up in advance: verifying a new account can take a few days.

Read the consent form properly – it's the one document that tells you exactly what happens to your audio and recordings. Before signing any study's consent, check that you’re provided with information on four key aspects:

  • What will be recorded? Video, audio, or both.

  • Who will use the recordings? The research team, a named client, or an unnamed company.

  • Will anything be published? Recordings can be internal use only or may be published, depending on the project.

  • Can you change your mind? Look for how to withdraw and what happens to data already collected.

If the form doesn't answer these, ask the project coordinator before you join.

On the day: What happens in the room

Study sessions typically follow a fixed sequence, run by a small team:

  • Arrival: A coordinator meets you and confirms your identity. Bring a government-issued ID that matches the one you verified online.

  • The studio: You're shown to a small studio. A session director guides the process, and a moderator might be available to help you.

  • A short intro video: You typically watch a brief video explaining what to expect before anything is recorded.

  • The task: You chat with other participants, follow step-by-step instructions, cook a meal, fix a bike — whatever the study calls for.

You don't need acting skills or preparation. A moderator may give simple prompts, such as speaking in a certain mood or being a bit more assertive, but the aim is natural, casual task completion.

Learn more about different in-person studies currently open on Mindrift!

After the session: How and when you get paid

Payment is typically fixed and doesn't depend on how the task goes. In the Santa Clara Valley study, $110 is added to your Mindrift balance within 48 hours of the session. You withdraw it on the regular payout schedule. Other studies, like the multi-day language study in New York, pay higher, with contributors receiving up to $1,600 on completion.

Two practical points to keep in mind:

  • Travel isn't reimbursed: Factor parking or transit into whether the session is worth it for you.

  • Provider fees may apply: Mindrift charges no fees, but your payout provider may.

Payments for studies count as income. US participants can check reporting rules in the IRS gig economy tax center.

Who can and can't take part

Eligibility rules for AI studies are stricter than for most focus groups, because the recordings need to reflect specific criteria. For example, the Santa Clara Valley study requires participants to be:

  • 18 to 64 years old

  • Able to attend in person at one of four sites in Santa Clara Valley, California

  • Have the United States set as your country in your Mindrift profile

  • Not work at major tech and AI companies, or as bloggers or content creators

  • Not be a resident of Illinois, Texas, Washington, Colorado, and Maryland

The study page doesn't explain each exclusion, but the pattern is common across AI data studies. Tech-industry participants and content creators can skew "natural" conversation data, and several US states have specific laws on recording and biometric data, which often affects where studies can enroll participants.

Every study is different and asks for specific requirements that include location, age, language knowledge, field or work, etc. It’s critical to read all requirements thoroughly before joining a study.

How to tell an in-person study is legitimate

A real study pays you and never asks you to pay. Before registering for any study, check that the sign-up runs through the organizer's own website, that joining is free, and that the consent form clarifies who will use the data. The checklist for spotting fake AI training platforms applies the same tests to remote projects.

If a study recruiter contacts you through a messaging app and asks for a deposit, an equipment fee, or bank details before you've signed anything, it isn't a real study. Not sure if it’s really Mindrift contacting you? Check out our LinkedIn post or Reddit post for details on how to spot impersonators.

Frequently asked questions

Do I need any experience to join an AI research study?

No. Studies are usually open to people without special skills. Some studies do have specific requirements like language skills or knowledge of a certain field.

What do you talk about in an AI conversation study?

Everyday topics, such as celebrating a birthday or how a job interview went. Studies of this kind avoid sensitive and personal subjects.

Are the recordings published online?

Sometimes. Most studies on Mindrift are for our clients’ internal use and are not published. Always check the consent form to know for sure.

Can I take part more than once?

Some studies restrict previous participants from joining a second time. All studies set their own rules, so it’s best to check on the project description or landing page.

Are there remote alternatives if I can't attend in person?

Yes. Remote data projects pay for things like recording everyday activities on your phone. The guide to AI data collection projects covers what's available.

Explore in-person AI training

Ready to try a new wave of AI training projects? In-person studies can be a great way to make extra income, use your knowledge to train better models, and level up your resume. Studies are paid, guided, and interesting.

Check out open opportunities to see if there’s an in-person study near you:

Explore projects

Want more great reads? Check these out:

Our picks for the best AI newsletters to fill your inbox with

Debunked: My accent isn't standard enough for voice data projects

Multiple income streams: How freelancers stabilize income across platforms

Frequently asked questions

Do I need any experience to join an AI research study?

No. Studies are usually open to people without special skills. Some studies do have specific requirements like language skills or knowledge of a certain field.

What do you talk about in an AI conversation study?

Everyday topics, such as celebrating a birthday or how a job interview went. Studies of this kind avoid sensitive and personal subjects.

Are the recordings published online?

Sometimes. Most studies on Mindrift are for our clients’ internal use and are not published. Always check the consent form to know for sure.

Can I take part more than once?

Some studies restrict previous participants from joining a second time. All studies set their own rules, so it’s best to check on the project description or landing page.

Are there remote alternatives if I can't attend in person?

Yes. Remote data projects pay for things like recording everyday activities on your phone. The guide to AI data collection projects covers what's available.

Explore in-person AI training

Ready to try a new wave of AI training projects? In-person studies can be a great way to make extra income, use your knowledge to train better models, and level up your resume. Studies are paid, guided, and interesting.

Check out open opportunities to see if there’s an in-person study near you:

Explore projects

Want more great reads? Check these out:

Our picks for the best AI newsletters to fill your inbox with

Debunked: My accent isn't standard enough for voice data projects

Multiple income streams: How freelancers stabilize income across platforms

Article by

Mindrift Team

Explore AI opportunities in your field

Explore AI opportunities in your field

Browse domains, apply, and join our talent pool. Get paid when projects in your expertise arise.

Browse domains, apply, and join our talent pool. Get paid when projects in your expertise arise.