trainer

software engineer

writer

prompt engineer

freelancer

Outlier AI

The positive with this company is that you have a lot of freedom with regard to when, from where, and how much you work. However, this freedom has been pretty much all but cancelled out by managerial and corporate incompetence. 

I have been with Outlier for over a year and worked my way up to what they consider an "elite status" contributor. That has gotten me better support (which is often needed), and the offering of bonuses whose requirements simply aren't feasible for me to meet, so I never get them. Here are the main drawbacks of working for Outlier:

1. The pay is opaque and unreliable. Last summer, after contributing 7 months of quality work and loyalty, Outlier cut my pay by 25% overnight, without notice, warning, or explanation. This was after recently being praised as a top contributor and being onboarded as an early bird onto high-priority projects. The only way I knew my pay had been cut was by noticing that the previous day's pay had been reduced. There was no advanced notice. Support wouldn't even acknowledge that my pay had been cut. The project leaders told us who had experienced a pay cut that this was a mistake and would be remedied by the next day. Then, after two weeks, we realized that this was a lie and that the reduced pay would never be addressed.

2. Work availability can vanish without notice. My most recent project would shutdown for entire weekends without so much as a second of notice. This has made planning weekly hours and meeting quotas impossible. You can be out a couple hundred bucks of work for the week with no notice. Projects can be discontinued or shutdown overnight. Outlier can be a suitable side-hustle for spending money, but no one should ever rely on it for necessary income.

3. They occasionally offer special bonuses for working a certain amount, but I mostly ignore these, as they are more of a hassle than anything. The bonuses are nice in theory, however, for some of my bonuses, I spent weeks arguing with support in order to get them. They will find any reason not to pay your bonus. For one bonus, I met all the specified requirements, but wasn't paid. Over the course of 6 weeks, I sent 8+ emails to support and couldn't get anyone to acknowledge why my bonus was being denied. They would just have me send screenshots and then say that I didn't meet the requirements. Finally, I got one support agent who actually took the time to count my hours, and then realized that I met the requirements. I was paid the next week. However, I probably would have made more if I had just ignored the bonus and spent the hours it took arguing with support working on a project instead.

4. Their communication is terrible. I've been sent mass emails telling me that I've been matched to a project, and then rejected for a project, and then re-matched with it without ever interacting with the project at all. I've been sent quality warnings telling me that I was going against Outlier community guidelines and risked removal from projects, only to have them tell me hours later that the email was sent out to me in error. As noted above, they won't communicate with you about important decisions that affect contributors' livelihoods. They don't consider that you have a life and bills to pay. You're a cog in a wheel and they will cut your pay, reduce your hours, or end your project whenever they want, without telling you.

5. Project pay structures are becoming increasingly unfavorable to contributors. When I onboarded a year ago, I was paid for training for projects, and paid the full amount for all the time I spent on a project. Now, most projects don't pay you for the hours of training it takes to get onboarded. They pay you half-time for the assessment tasks they give you on a regular basis for quality assurance. And they expect meticulous, flawless, high-quality work, but also reduce your pay by half if it takes you longer than 30 minutes to evaluate, fact check, and research a 5-paragraph model response, for which you have to provide 3-5 fact-checking sources complete with quotes and rate the response according to 10 categories and write flawless and accurate 200-word explanations for. 

All that to say that there's a reason Outlier has such low ratings on Indeed. I'm one of many, many people who have been yanked around by this company.



I started working for Outlier earlier this year, and from the beginning it was a confusing mess. After the onboarding assessments, it was unclear when the work would start or whether I would actually be compensated. But soon, I was heartened to see that I was actually on a project and being paid at a decent rate ($40/hour).

The work is somewhat interesting, but the guidelines are constantly changing, and the platform regularly experiences errors that make it difficult or impossible to work. Some of this is understandable, since processes are being refined by client feedback and tech issues ~happen~. But the way employees are treated is completely unacceptable. We're fully disposable and nobody cares if we can actually work or if we're happy. My pay was cut by $5/hour without any prior notice--I found out when I was paid less for a whole week's worth of work, and was basically told "that's your rate now, sorry we didn't tell you in advance, deal with it!"

They now offer "missions" to incentivize employees to work more, usually with ridiculously high hour goals for not very much money. I laugh every time I see a new one at how bad of a deal they all are.

It seems the company basically banks on having a steady, fresh crop of new workers to put on tasks, because they cut people from projects all the time. Reviewers are typically other employees who are summarily "promoted," which creates a lot of quality issues and fosters a constant conflict between prompt writers and reviewers. Also, many "promotions" do not come with any change in pay. Just a change in task.

The people who run the webinars are nice and do their best, but it's obvious that they are also stretched pretty thin trying to juggle multiple priorities. I feel bad for them and hope they are treated better, and paid more, than the rest of us. It is unclear who actually makes decisions in the company or what their ultimate goals are.

If you need to make a quick buck here and there, and are willing to put up with a ton of bull, you can get something out of working here. But it's not steady work. Do not count on it or quit any other job you already have.

If you are working here and only making $15/hour (as I have heard some are) PLEASE HAVE SOME SELF RESPECT AND STOP. It's not worth it at all.

Outlier.AI reminds me of the next version of DataAnnotation.tech - offering flexible working conditions remotely, where you help provide, select, or generate AI prompts all in the effort of helping to train models. 

While this sounds great, maybe fun, or even an income to be generated for doing "work" it does not pan out to be this. 

There is a SEVERE lack of understanding from EVERYONE who is..'hired/contracted' -- "What do we do? Who's the boss? Where is support? What team am I on? How do I get paid? How do I use the tools? The tools aren't working, who do I contact?" 

Sadly these are the overwhelming comments I saw while in random Slack rooms. The company is not very well organized, there is no structure, the on-boarding process to start seems like it's a legitimate role you're being tasked however after "training" most Outlier.AI Trainers have zero work -  there is a reason for this, there is too much occurring, and Outlier.AI itself is providing a service to clients -  therefore, if there is no work in  your "field" of expertise, you are wasting your time.

Much like DataAnnotation (the same idea) you might yourself logging into various tools each day just to see if there's work for you to do. 

This 'job' is for you...if you have time to log into some random chat rooms/websites (where you don't know anything, or anyone) to see if there's work to do, if not you can bounce and leave. Also, if you need income, I would highly recommend staying away. 

I was compensated fairly, and paid weekly for 2 weeks without issue, and I got to help "train" AI - I made less than $50.00, and worked for a total of around 4-5 hours. 

Outlier AI is attempting something here...by providing Trainers to grade/review AI chats however they have multiple kinks to iron on, as well as...the whole on-boarding process - it is the worst feeling ever to not know what you're suppose to be doing, or where you're suppose to be in a remote environment, and I honestly felt bad (yes, I'm a feeler type of person towards others in the same boat, with zero direction, asking the same questions as me -- I had to find out a lot of things on my own time, and I didn't even get to work with AI all that much. 



This is a highly unprofessional company with inappropriate practices regarding employee treatment regardless of the fact that we are all "free lancers." 
I have worked for this company for over a year now. Work was consistent in the beginning but tremendously dipped off before Thanksgiving last year and has been sparce ever since. Communication from corporate down to us through the managers (way too many managers) has always been an issue and never was really resolved. For instance, they decided to cut contributor pay rates thrice, at least. They issued no warnings about it nor told anyone about it. We found it out for ourselves and no one really acknowledged and now act like nothing happened. They (corporate and administrators) just did it. Then there have been multiple technical issues with tasks that make it impossible to put in a full day's worth of work let alone a full week. The engineers, programmers, and coders are constantly "working" on resolutions of the problems with often very few or no results.
Now, they updated their system to automatically allocate people to different projects which sounds nice at first but turned into a holy nightmare. People are removed at whim of the system's discretion and most of the time it's the result of a really bad glitch or mistake. Yet, it's never fixed nor can you reach anyone successfully to get answers. When you do, they also have no clue and will most likely ignore you or forget about you entirely. If you approach them about the issue again after no response in days, they will either still ignore you or give you a mildly angry message about bothering them. 
I could go on, however, I will not. Instead, I will say while it seems worth it on paper this job does not live up to its hype or expectations. It's a huge let down with long processing for essentially Payless on boarding, assessments, and training for slim to no work. Keep your hopes and expectations at the bare minimum and you should not be disappointed too badly.  

This company is an absolute trainwreck on so many fronts, I can't even begin to describe the horrors of working for them— first as an expert "attempter" and then a reviewer. They treated me like expendable garbage with demeaning and dehumanizing practices from the moment I started. This treatment is the same across the board for all taskers. 

It actually felt like they wanted us to feel like little expendable piece-of-junk cogs so the tech admins could find some twisted sense of power.

Example: Performance bonuses are totally inconsistent week to week—one week I was offered a 4-figure bonus for an almost impossible amount of reviewer tasks, the next week the same amount of work was a few hundred dollars, and these were always given with the minimum amount of notice. The next week I was bumped to a new project and demoted with no explanation.  These bumps happened all the time, with no explanation. 

Not just me, they move everyone around to different projects with occasional random cryptic notices. Every new project requires training on an entirely new system/rubric without pay, while countless system bugs make the work difficult to do. The work is actually incredibly demanding to do well, it requires a high level of concentration and skill, and there is rarely enough time allotted to complete the tasks well. They know it takes longer but they limit how much time you can actually get paid for any give task, so you often end up working a lot more and getting paid for half the time you actually worked.   
 
Based on my experience I would say it's obvious that this is a company run by narcissists.  The only bright spot is some of the QM's who actually care and try to help. But they have limited power. I Highly recommend not wasting your time and energy with this awful company.  Go somewhere else where you can be appreciated for your gifts and talent and actually make a difference.   

**A typical day at work:** As a freelance worker, you are told that you will make a certain hourly rate. Then, once you go through paid/unpaid training modules, and pass, you are assigned work that is related to AI prompt and response validation, AI writing, or other AI-related tasks depending on your qualifications. Often, you work on one or two tasks, then you are told your work queue is empty. You wind up spending more time working on unpaid training tasks or attending unpaid training webinars than you do working for pay.
**What you believe you learned:** You learn a lot about how to follow Rubrics, and you do learn about the basics of AI prompts and responses.
**Management: Management is non-existent. When you interact with actual humans in the training webinars, they are usually young people who come across as smart, but it's obvious that they lack management skills. It's like being trained by brainy college students who know their product but do not know how to train adults. It's very bizarre. 
**Workplace culture:** There is no actual culture unless the Slack tool is considered a "workplace culture." And if it is, I would say the culture at Outlier is organized chaos, which at the end of the day is nothing more than an unorganized AI experiment. 
**The hardest part of the job:** Working through the maze of unpaid training and keeping up with what team you've been assigned to in Slack, as it changes every other day, and I am not kidding.
**The most enjoyable part of the job:** If you are lucky enough to get a paycheck, that would be the most enjoyable part of the job; plus, the fact that you work remotely. 

This review details my experience with Outlier's onboarding process and overall work environment. While they advertise a plethora of positions, my recent experience highlights significant issues with their onboarding, communication, and task allocation.

Unclear Onboarding Process and Lack of Communication
-Be wary of their job postings. Despite advertising numerous positions, the onboarding process is unclear and suffers from a severe lack of communication.

Delayed Onboarding & Missing Pay
-My onboarding took twice the estimated time, and despite promises of compensation for training, I received no payment.

No Assigned Tasks
-Weeks after completing onboarding, there were simply no tasks available for myself or many other new hires.

Unresponsive Support
-Attempts to reach support resulted in generic responses from the help desk, and internal communication channels offered no replies from managers or administrators, leaving hundreds of messages unanswered. You are told to submit requests, but this is the outcome.

Project Misplacement
-Disturbingly, I, along with others, were assigned projects entirely different from those we applied for. This raises serious concerns about the company's organizational structure.

Payment Delays
-Internal discussions revealed numerous reports of missed payments for completed work by new hires.

Overall Impression
-The lack of communication, unclear project allocation, and delayed payments create a chaotic and unprofessional environment. I strongly advise potential applicants to consider these factors carefully before applying to this company. They are not what they present themselves to be and you will waste your time thinking you have a great opportunity to work remotely, but there is no substance to their offers. 

Outlier AI is perhaps best suited as a temporary band-aid for a minimal income over an inconsistent period. It offers remote work with minimal entry barriers and highly flexible work-to-time conditions.

Income is collected by tasking with projects to which you are assigned. Usually, Outlier's system force-assigns you a main project to work on. Onboarding on a project can take several hours or longer. Later portions of projects pay you a small amount for participation. Still, the pay is meager, at least a few dollars around the US federal minimum wage. Completing project onboarding offers at least double that if you make it that far.

Inconsistency is in project availability and workload. Projects will require more or less work from you for about the same payment. Some projects will go "EQ," meaning there's no more work available, and you need to wait for your queue to fill back up. Or, you can onboard onto another project in the website marketplace for something else to do before the system jerks you back into your assigned main project. 

Some projects require more onboarding steps, such as getting credentials for another website they want you to copy/paste submit work to while also submitting to Outlier's website. Management in these projects is inconsistent, leading to confusing downtime during which none of their onboarding guides help answer what you need to do next. The Outlier forum is plagued with users asking for help during project onboarding processes.

I do not think Outlier AI is a source of sustainable income under all these conditions. As I typed, it is a band-aid solution to a little extra money for a mild amount of effort and more than mild aggravating bits of time. I am using it to add something to my resume's empty "Work Experience" portion. 

I was excited when I got this job, thinking this would be the perfect work for me as a digital nomad. What I found was a lot more frustration than I have ever had working for any company. I tried to grin and bare it for a couple of months to see if they would work out the kinks but it seemed they just keep hiring more workers instead.
From my view as a tasker this has been the most unorganized and unstable job I have ever had. I would get put on one project and learn the ropes on it, start tasking, and the "rules" for the project would change almost daily. The reviewers and the taskers were never working from the same information, so we were always getting unnecessarily bad reviews that kept our scores down. By the time I would finally feel comfortable tasking, I would get raptured and thrown into another project with no communication whatsoever. 
I never really had any leadership I could speak to about anything so I was literally teaching myself everything, and I'm almost positive that the customer service "people" that respond to you are just AI bots because they always reply with the same messages when you have an issue.
The platform is unstable and was going down almost every day that I worked, so that's pretty frustrating as well. There would be "bonus missions" offered for hours worked, but every time I tried to participate in one of those, the platform would conveniently go down before I could complete my "mission" and I wouldn't get the bonus.

**Overall, this place seems very sketchy and I would not recommend. Don't even waste your time!!**

What is the best part of working at the company?
The research-oriented work itself was interesting and had potential, especially for contributors who enjoy sourcing, verification, and pattern-recognition tasks. However, the internal workflow and monitoring systems did not always seem aligned with the actual behavior required to complete research-heavy projects accurately.

What is the most stressful part about working at the company?
In my experience, there appeared to be a strong reliance on automated behavioral flagging without enough workflow-specific context or meaningful contributor feedback. This created concerns around how legitimate research activity, such as tab switching, external verification, reading source material, and inconsistent pacing depending on task complexity, may be interpreted.  If I was a company looking to outsource AI creation, operations, or development.  I would not trust them with such a important and expensive development for my company.


What is the work environment and culture like at the company?
being a "contract" employee they cannot train you.  Questions can only be asked in group chats or during zoom meetings during unpaid time.  Task the require research or switch tabs to do the task will get you terminated because they use AI patterns to flag instead of human investigation.  That fact alone shows they are not prepared to be a outsource for AI training.  Their own AI is hugely lacking human training and recognition and proper communication.

What is a typical day like for you at the company?
searching chats and sitting in zooms, unpaid, looking for answers, never getting feedback, then being confident you are doing quality work over speed.

Top Job Titles	Average Salary
Trainer 3 salaries reported	$30.00 per hour
Software engineer 7 salaries reported	$42.86 per hour
Writer 3 salaries reported	$28.33 per hour
Prompt engineer 3 salaries reported	$32.67 per hour
Freelancer 3 salaries reported	$46.67 per hour

Working at Outlier AI: Company Overview and Reviews

Outlier AI

Outlier AI Ratings

Popular jobs at Outlier AI

Salary Satisfaction

Job Seeker Tools

Employer Tools

Browse

Stay Connected