Outlier Reviews 4,055

I was quite patient with the process with Outlier, spending my own time onboarding, often not having tasks, and being changed from project to project: the pay was real (even if it was sporadic), and the projects were interesting. However, when I was suddenly no longer able to access my account, I contacted support and they informed me that my account had been thoroughly reviewed and deactivated based on a violation of the code of conduct. I have always been careful to honestly register my tasking times and did not use AI to answer tasks in my queue. They would not tell me what I was accused of violating, and they said the decision was final.

Outlier.ai is okay when they actually let you work. First, they unfairly shut down your account for days, sometimes a week, to review the work you’ve done, stating there was “suspicious activity.” The reality is more like that they have too many workers and don’t want to pay everyone they’ve onboarded. So, if you want to get paid and work consistently, DON’T trust this company. Second, they have a cut off time, only allowing so many hours as a collective. They constantly change when you can start working. Sometimes it’s 12am, sometimes 2am, sometimes 8am, you just never know until they post that tomorrow the reset time will be at such-and-such time. That means if they have reset at 2am and you don’t wake up until 8am, others will have 6 hours of work done, and they’ll stop allowing you to work usually around 10am because everyone who started at reset took up the collective hours. Lastly, their community chat rooms are more like Discord. It’s messy, unprofessional, and very hard to navigate and find what you’re looking for. If you want help you have to ask the chatbot which is apparently programmed to tell you most times that you can’t put in a support ticket for multiple issues you may have.
I’m still waiting to be allowed back in after they cut me off and stated there was “suspicious activity” which is under review. This is the second time this has happened to me this year with this company. The last time they found nothing after stating there was suspicious activity and brought me back onboard. They sent an email to everyone stating we should not use AI to do our work. Not sure how you’d even use AI for this kind of work since it’s all manually entered. Shortly after that email is when I got cut off.
This company is so unprofessional and shady. Be advised and stay away.

After 2 years of work they just banned my profile for no reason and all I got on my end was an auto generated email telling that I was found in violation of their community guidelines/ terms & conditions, where I did nothing. I was in the middle of tasking and suddenly lost access of the platform. When I submitted a support ticket, it said that the decision of my account deletion is final, they won't give me my pending payments, I won't be having access to their platform and hence to any new projects. And I cannot apply for new profile either. This is absurd, at least they should have given a proper reasoning. And when I digged about it on Reddit, I found out this has been happening to a lot of contributors these days. One reddit user even quoted that they're into mass deactivation these days.

My Outlier account was suddenly and permanently deactivated without any clear explanation, evidence, or meaningful opportunity to respond.

The support process appears to be entirely automated. Each reply I received was generic, vague, and sent within a very short time, which gave me no confidence that my case was reviewed properly by a real person. The emails repeatedly stated that the decision was final, that no further appeals would be considered, and that they could not disclose the specific reason for the deactivation due to “platform integrity.” In practice, this means the user is left with no transparency, no accountability, and no fair process.

Outlier relies on contributors’ time, effort, and trust, but when an account is deactivated, the platform provides only template responses and refuses to explain what actually happened. Even if there was a genuine concern, a professional platform should be able to provide a clear category of issue, a proper appeal route, and confirmation that a human agent has reviewed the matter. Instead, the process feels like an automated system making irreversible decisions while support simply repeats scripted language.

This is a very poor way to treat workers who have invested time into the platform. The lack of transparency, the refusal to provide specific reasons, and the absence of any real escalation pathway make Outlier feel unreliable and unfair.

I would strongly caution anyone considering working with Outlier. Your account can apparently be removed without warning, without a clear explanation, and without access to a meaningful human review.

I have inside information from a project leader who is a friend of mine, and he confessed that they not only have access to data they shouldn't have, but to make matters worse, they assign a hidden score to each person and, based on that, give them more or fewer tasks. The problem is that sometimes they randomly ban users or disapprove perfectly completed tasks to avoid paying for them.

Payment, Value for money

I can see this is a common problem just read the reviews from others.

My experience is that I only just joined the platform and completed the initial introduction questionnaire. I hadn’t even started working properly or been given any real tasks, yet my account was suddenly flagged and permanently deactivated.

What makes it more frustrating is that I was accused of things like automation, plagiarism, or deceptive input methods none of which apply to me. Everything I did was done manually.

I reached out to support, but they were unable to provide any actual evidence or specific explanation for what I supposedly did wrong. After a second review, they simply said the decision stands and the case is closed.

It’s disappointing to be permanently banned without transparency, especially when you’ve barely had the chance to use the platform. I would advise others to be cautious.

Imagine trying to do freelance AI work. Honest work for honest pay. You're asked to do onboarding, and you finish it. You pass the tests you're given proving you can actually contribute, and wait for the work to come....only for the chatbot(outlier's ONLY customer support) says your account was deactivated because you use AI(the god damn IRONY) and cheated by copying and pasting your work.....

...despite never having even DONE anything yet.

I try to appeal this, but the chatbot insists that my account was "manually reviewed"(this is a lie. My account was deactivated within minutes of onboarding) and that the decision was final....just like that. No chance to defend myself.

That. That's Outlier. It's garbage, and you should avoid it. They robbed me of 2 hours of my life(unpaid hours) and I want them back.

If Outlier disagrees, I more than welcome them to address this themselves instead of having their(very bad) chatbot do it.

Edit: Not only has Outlier failed to address the issue, but their awful AI support bot continued to handle the ticket I submitted and insisted on my deactivation/ban with zero justification. Eat a johnson, Outlier. Avoid it like the plague.

Outlier is ever good and improving. Even though fantastic work is already done and well paid, they never fail to thrive to improve on every feedback. It's a place to improve, grow and earn based on your hard work. I would encourage anyone who wants to work hard and earn some extra cash to join me in this journey to improve AI through quality data produced by us in Outlier.

I enjoyed Outlier for a good 4-5 months. It was my first AI job, and now I am on multiple sites with good reviews and work experience. I've been added to multiple small, specialized projects because of my high-quality work. I had good reviews on Outlier and always put in my best effort.

I was working in Toronto for a few months and put my travel requests in, all of which were approved. Suddenly, out of nowhere I was permanently deactivated. I reached out to support and got nothing but automated responses. They never gave me an answer, just said I violated guidelines. A QA on the Community told me my account should be good and to reach out to support, and nothing. I finally got ahold of a Moderator on the subreddit who told me they would look into it for me, just to be ghosted by them as well.

This company will ban you for no reason and throw automated responses at you before ghosting you. If you go on the subreddit, you'll see that this happens to countless people every day. Highly recommend other companies: Mercor, Alignerr, Handshake, etc.

I make more on those sites and am recognized for my work always. Outlier is extremely disorganized and unprofessional. Stay AWAY.

Stay away from this dishonest platform as it lacks transparency and efforts to protect user safety! A few weeks ago, I was notified that my account was to be deactivated for breaching community guidelines; when I appealed and requested to know the reason for this abrupt decision, it was implied that I held duplicate accounts. As this was false, I requested evidence for this finding, as it could potentially signal that I had been a victim of identity theft. Outlier support refused to provide any further explanation; therefore they clearly do not care about the safety of their users. Even before this, I had grown wary of this platform as they will remove you from multimango without warning, even if your QA scores are consistently above benchmarks and well above the stated removal boundary. Their QA reviewers do not consistently apply guidelines, and it is common for annotators to receive ratings that are incorrect/contradictory to guidelines. There is a relatively practice of unqualified annotators using social engineering tactics to gain access to reviewer roles and projects by lurking in community forums, DMing active contributors, and attempting to convince them to give them answers to assessments/screenings/onboarding tests. This platform is a mess and has massive security issues, so please avoid at all costs!

Signed up and verified and was approved but everytime I logged in it asked for verification so I verified again and then raised a ticket because I couldn’t progress in the dashboard because of this continuous loop they put me in and then supports says i verified too often so can’t continue using them. What an awful company and waste of time, they need to train their own software and team and improve themselves before training future ai.

Onboarded three times, was never paid. Accused of cheating when I never did.

Outlier hired 50,000 to fill 8,000 openings, constantly held these ridiculous lottery style giveaways while abruptly deactivating contributors account for "quality" thankfully I was able to find another platform that is 10 times better.

What a great platform to earn money as a side hustle and at the same time contribute to the evolution of AI. Outlier is 100% legit for those who still have a doubt.

Outlier, Aether, Multimango and Hubstaff are some of the most unprofessional so -called organisation ever come to exist in the digital world. If I could give them 0 star, I would in a jiffy! No real support available (that make sense). No transparency on reviews and always reviewing or focusing on a mistake that you've done and never giving you a positive review on the tasks that you completed and invested time on it. Which means, your score kept going down and never going up. Stay away from Outlier and their associated platforms. Look else where.

I worked on Outlier for over a year and, after months of inactivity on my account, it was permanently deactivated. The stated reasons were repeated copy-pasting, auto-typer use, and AI-assisted task completion, allegedly confirmed through manual audits and a secondary review.
The problem: I had not submitted a single task for months.
There was no recent activity to audit.
When I asked support for the specific task IDs and submission dates that were supposedly reviewed, I received no answer. When I asked to speak with a human rather than receive more templated or AI-generated responses, I received no answer. The only consistent replies were copy-paste messages citing "platform integrity" and "we cannot disclose the specific signals used."
This is what you are dealing with: a permanent deactivation with no verifiable factual basis, no meaningful appeal process, and a support channel that is effectively a wall.

The good:
When tasks are available and payments go through, the hourly rates can be reasonable.
The platform itself is functional and generally easy to use.

The bad:
Account decisions are opaque and final.
Templates are the norm, human reviewers are unreachable.
Projects are massively over-assigned. It is routine to complete an onboarding assessment that takes longer than the actual task window you get to work in. I have spent more time on assessments than on paid tasks on some projects, because by the time I had delivered one or two tasks, the project was already closed due to the sheer number of contributors.
Over time, the ratio of unpaid setup and assessment hours to actual paid work makes the platform not worth the long-term investment. You optimize for projects that may evaporate before you break even on the time invested.
Deactivations appear to be partially automated and applied without individual case review. If you are caught in one, there is no real process to challenge it.

Bottom line: the platform can work in the short term if you catch the right projects at the right time. As a long-term income source, the combination of over-assigned projects, unpaid assessment overhead, and a punitive, opaque account management system makes it a poor investment of your professional time. If your account is deactivated, expect zero meaningful recourse.

The site is genuine and I love that I can contribute my knowledge of the fields towards building AIs that have a positive impact on the field. It's also nice that the amount I get paid is proportional to the work quality I produce so I'm very grateful for that!

Outlier was my introduction to the world of AI annotation and the good thing about Outlier in particular is its accessibility by all people regardless of their level of technical proficiency concerning AI or technology savviness in general

I recently joined a project, which I was delighted to be accepted on because I am in financial difficulty and this felt like a lifeline. I am sure a lot of people are in this situation.

The task was to read and assess the generated prompt, do a voice recording of the prompt in my own words, review it, transcribe it and then check the accuracy of the transcription before submitting and moving on to the next prompt.

The onboarding instructions stipulated that there would be 4 prompts to work through and I could earn around $23 per hour if completed within 18 minutes or around $18 per hour for up to I think 30 minutes.

4 prompts is certainly doable but I was given 9 and was kicked out of the task and moved onto the next one, without pay. The next task had 8 prompts!

What this tells me is they have engineered the tasks to be impossible to complete in the allotted time thereby avoiding payment, but they keep your work. Very suspect.

Avoid it!