Key findings
- 1
All 29 job boards were read between 17 Oct 2026, 11:26 UTC and 18 Oct 2026, 06:01 UTC.
- 2
5 of 7 quantities counted by two independent routes match exactly; the others differ by 14 (0.3%) and 12 (0.3%), for the reasons given under the figure.
- 3
Of the roles open on 17 Oct 2026, 83% state an hourly rate, 100% carry a location rule, 89% have a full description and 40% an application form that was read.
- 4
2 of 6,202 recorded apply clicks were set aside (automated, or made on other sites), and 1,053 of 22,085 page views; 69.7% of the clicks counted were made on a specific role.
- 5
For 15 of 15 platforms, the median rate in their own postings falls inside the pay range compiled from their published pages.
Figure 1
The same numbers, counted two ways
Do independent counts of the same quantity agree?
| Quantity | From the role facts | From the second route | Agreement |
|---|---|---|---|
| Open roles at the end of 17 Oct 2026 | 3,668 | 3,668 | Match |
| Open roles at the start of 20 Sep 2026 | 2,595 | 2,595 | Match |
| New listings in the window | 2,252 | 2,252 | Match |
| Removed listings in the window | 2,181 | 2,181 | Match |
| Open role-days in the window | 79,828 | 79,828 | Match |
| Apply clicks on roles listed in the window | 4,308 | 4,322 | Differs by 14 (0.3%) |
| Open roles recorded on 17 Oct 2026 against the same moment rebuilt from role history | 3,670 | 3,682 | Differs by 12 (0.3%) |
Sample. Every role listed from 20 Sep 2026 to 17 Oct 2026.
First route: the role facts behind the platform, language, field, pay and geography reports, counted in the app. Second route: the database queries behind the hiring and applicant reports, written separately, and for the last line the snapshot recorded that morning.
Apply clicks can differ slightly: the second route also counts clicks on a role after it left its feed (from a page opened earlier).
The snapshot line can differ when a role was merged as a duplicate or reclassified after the day it was recorded.
Source. AITraining.jobs Research, Data Quality and Coverage, live data.
Figure 2
When each platform's job board was last read
How fresh is the data behind these reports?
| Platform | Last read | Open roles on 17 Oct 2026 |
|---|---|---|
| Invisible | 17 Oct 2026, 11:26 UTC | |
| Alignerr | 18 Oct 2026, 06:01 UTC | |
| SuperAnnotate (SME) | 18 Oct 2026, 06:01 UTC | |
| Mindrift | 18 Oct 2026, 06:01 UTC | |
| micro1 | 18 Oct 2026, 06:01 UTC | |
| Welo Data | 18 Oct 2026, 06:01 UTC | |
| Mercor | 18 Oct 2026, 06:01 UTC | |
| AfterQuery | 18 Oct 2026, 06:01 UTC | |
| Turing | 18 Oct 2026, 06:01 UTC | |
| Handshake AI | 18 Oct 2026, 06:01 UTC | |
| Anyone AI | 18 Oct 2026, 06:01 UTC | |
| Sigma AI | 18 Oct 2026, 06:01 UTC | |
| Braintrust | 18 Oct 2026, 06:01 UTC | |
| Perle Labs | 18 Oct 2026, 06:01 UTC | |
| TransPerfect DataForce | 18 Oct 2026, 06:01 UTC | |
| Claru | 18 Oct 2026, 06:01 UTC | |
| Appen | 18 Oct 2026, 06:01 UTC | |
| Babel Audio | 18 Oct 2026, 06:01 UTC | |
| OpenTrain AI | 18 Oct 2026, 06:01 UTC | |
| RemoExperts | 18 Oct 2026, 06:01 UTC | |
| OneForma | 18 Oct 2026, 06:01 UTC | |
| Terac | 18 Oct 2026, 06:01 UTC | |
| Sovrano AI | 18 Oct 2026, 06:01 UTC | |
| Vetto | 18 Oct 2026, 06:01 UTC | |
| xAI (SpaceXAI) | 18 Oct 2026, 06:00 UTC | |
| Pasiflora AI | 18 Oct 2026, 06:01 UTC | |
| Outlier (Scale AI) | 18 Oct 2026, 06:01 UTC | |
| Remotasks | 18 Oct 2026, 06:01 UTC | |
| Pila8 | 18 Oct 2026, 06:01 UTC |
Sample. The 29 platforms with open roles at the end of 17 Oct 2026.
Each job board is read in full every day. 'Last read' is the latest time a role of that platform was confirmed in its feed.
Source. AITraining.jobs Research, Data Quality and Coverage, live data.
Figure 3
How completely each platform's postings are described
Which facts does each platform's job board publish for its roles?
| Platform | Hourly rate | Posting date | Location rule | Full description | Application form |
|---|---|---|---|---|---|
| All platforms | 83% | 83% | 100% | 89% | 40% |
| Invisible | 99% | 100% | 100% | 100% | 100% |
| Alignerr | 100% | 0% | 100% | 0% | 0% |
| SuperAnnotate (SME) | 100% | 100% | 100% | 100% | 0% |
| Mindrift | 100% | 100% | 100% | 100% | 100% |
| micro1 | 98% | 0% | 100% | 100% | 0% |
| Welo Data | 78% | 100% | 100% | 100% | 100% |
| Mercor | 83% | 100% | 100% | 100% | 0% |
| AfterQuery | 100% | 100% | 100% | 100% | 0% |
| Turing | 11% | 100% | 99% | 98% | 0% |
| Handshake AI | 99% | 100% | 100% | 100% | 3% |
| Anyone AI | 100% | 100% | 100% | 100% | 100% |
| Sigma AI | 0% | 100% | 100% | 100% | 100% |
| Braintrust | 73% | 100% | 100% | 100% | 0% |
| Perle Labs | 0% | 100% | 100% | 100% | 0% |
| TransPerfect DataForce | 11% | 100% | 100% | 100% | 83% |
| Claru | 100% | 100% | 100% | 70% | 0% |
| Appen | 9% | 100% | 78% | 100% | 100% |
| Babel Audio | 53% | 100% | 100% | 100% | 0% |
| OpenTrain AI | 98% | 100% | 100% | 100% | 0% |
| RemoExperts | 97% | 100% | 100% | 100% | 0% |
| OneForma | 4% | 100% | 100% | 100% | 0% |
| Terac | 68% | 100% | 100% | 100% | 100% |
| Sovrano AI | 85% | 100% | 100% | 100% | 0% |
| Vetto | 15% | 100% | 100% | 100% | 100% |
| xAI (SpaceXAI) | 100% | 100% | 100% | 100% | 100% |
| Pasiflora AISmall sample | 29% | 0% | 100% | 100% | 0% |
| Outlier (Scale AI)Small sample | 100% | 100% | 100% | 100% | 0% |
| RemotasksSmall sample | 67% | 100% | 100% | 100% | 100% |
| Pila8Small sample | 100% | 100% | 100% | 100% | 0% |
Sample. 3,668 roles open at the end of 17 Oct 2026; shares of each platform's own open roles.
Hourly rate: the posting states one. Posting date: the job board gives one. Location rule: read from the posting or from what the platform states about where it hires. Description and application form: published by the platform's applicant-tracking system and read for the role.
Source. AITraining.jobs Research, Data Quality and Coverage, live data.
Figure 4
How the click and visit records are cleaned
How much automated and outside traffic is set aside before counting?
| Record | Events | Share |
|---|---|---|
| Apply clicks by people on this site (counted) | 6,200 | 100.0% |
| Apply clicks set aside as automated | 2 | 0.0% |
| Apply clicks on other sites sharing the log (left out) | 0 | 0.0% |
| Counted clicks made on a specific role | 4,322 | 69.7% |
| Page views by people (counted) | 21,032 | 95.2% |
| Page views set aside as automated | 1,053 | 4.8% |
Sample. Events recorded from 20 Sep 2026 to 17 Oct 2026.
Automated traffic (self-identifying crawlers, headless browsers, clients with no user agent) is marked as each event is recorded and never counted.
Source. AITraining.jobs Research, Data Quality and Coverage, live data.
Figure 5
What postings advertise against what platforms publish
Do the rates in a platform's own job postings fall inside the pay range it publishes about itself?
| Platform | Median advertised in postings | Published range | Result |
|---|---|---|---|
| Alignerr | $10 to $200 | Inside the range | |
| Invisible | $6 to $65 | Inside the range | |
| Handshake AI | $17 to $125 | Inside the range | |
| OpenTrain AI | $15 to $120 | Inside the range | |
| Mercor | $12 to $250 | Inside the range | |
| xAI (SpaceXAI) | $30 to $100 | Inside the range | |
| AfterQuery | $15 to $175 | Inside the range | |
| Braintrust | $20 to $200 | Inside the range | |
| micro1 | $10 to $200 | Inside the range | |
| Turing | $15 to $150 | Inside the range | |
| Babel Audio | $18 to $150 | Inside the range | |
| Anyone AI | $15 to $150 | Inside the range | |
| Claru | $5 to $100 | Inside the range | |
| SuperAnnotate (SME) | $17 to $115 | Inside the range | |
| RemoExperts | $20 to $150 | Inside the range |
Sample. Platforms with at least 8 open roles stating an hourly rate on 17 Oct 2026, and an hourly range on their AITraining.jobs page.
The published range is the one compiled from each platform's own pages, help centre and job posts and shown on its platform page here, with its sources.
Source. AITraining.jobs Research, Data Quality and Coverage, live data.
Figure 6
The dataset's record
How long has the data been collected, and how complete is the daily record?
| Measure | Value |
|---|---|
| Roles tracked since | 13 Jul 2026 |
| Roles ever tracked (open or closed) | 9,572 |
| Platforms with open roles | 29 |
| Daily market snapshots recorded since | 29 Aug 2026 |
| Snapshots recorded | 44 |
| Snapshots in this window (28 days) | 28 |
Sample. The whole dataset, read on 17 Oct 2026.
Closed roles are kept, never deleted, so any past day can be rebuilt; the daily snapshots record each day as it was measured.
Source. AITraining.jobs Research, Data Quality and Coverage, live data.
For labs, data vendors and researchers
Buy the data behind this report
The figures above are a summary. Behind them are records we collect ourselves, every day:
9,572 roles tracked, each with its history
Every role on 29 job boards, read daily since 13 Jul 2026: its pay and place, and the first and last day it was open.
6,200 apply clicks in 28 days
Which roles people actually apply to, by field, pay and kind of work. It happens on this site, so nobody else can count it.
39 lab and vendor hiring boards
The job boards of 25 AI labs and 14 human-data vendors, read every day: who is hiring for human data, and who is hiring to buy it.
What you can buy
The records
The rows behind any figure, with their history, not only the totals.
A regular feed
The same figures every week or month, as files.
A custom cut or benchmark
Your countries, languages or fields, or your own platform against the market.
Method and definitions
The job-board feeds, the click and visit records, the daily snapshots and the platform pages of this site. Every figure is computed by one database query over the window shown, the same query for the live report and for every edition. Each figure states its sample beneath it.
- Independent routes
- The role facts counted in the app, the database queries behind the hiring and applicant reports, and the daily snapshots recorded at the time.
- Published range
- The pay range compiled from a platform's own pages and shown, with sources, on its page on this site.
- Open role
- A role listed by a given day and still in its platform's feed that day.
- Window
- Live: the last 28 complete days, in UTC. Edition: one calendar month, in UTC.
What this data cannot tell you
- A check can only compare what both routes measure; a definition shared by both is not checked by their agreement.
How the job feeds are read and roles are classified: site methodology. How fresh, complete and consistent the data is, checked every day: data quality and coverage.
Editions
The live report covers the last 28 complete days and changes every day. Each complete month is also published as an edition, frozen two days after the month ends, so a figure you cite from an edition never moves.
- Live dataLast 28 complete days, recomputed daily
Cite and reuse
Free to quote, chart and republish under CC BY 4.0 with a link to this page. Live figures change daily: cite a monthly edition for numbers that will read the same later, or keep the access date.
Citation
AITraining.jobs (2026). Data Quality and Coverage (live data, 20 Sep 2026 to 17 Oct 2026). https://www.aitraining.team/research/data-quality-and-coverage (accessed 18 Oct 2026)
BibTeX
@misc{aitrainingjobs_data_quality_and_coverage_live,
author = {{AITraining.jobs}},
title = {Data Quality and Coverage (live data, 20 Sep 2026 to 17 Oct 2026)},
year = {2026},
howpublished = {\url{https://www.aitraining.team/research/data-quality-and-coverage}},
note = {Accessed 18 Oct 2026},
}More research
- Which AI Labs Are Hiring for Human DataWhich AI labs are hiring for the teams that buy, run and evaluate human data, and for what roles?Read the report
- Who Is Buying AI Training DataWhich AI labs and data vendors have people whose job is to source, license or acquire training data right now?Read the report
- Where Human-Data Vendors Are BuildingHow fast are the human-data vendors hiring their own staff, and for which functions?Read the report
- What AI Labs Pay for Human DataWhat do AI labs spend on human data, what do vendors bill, what do the experts receive, and what do RL environments and tasks cost?Read the report
- What AI Labs Are BuyingWhich kinds of training data are in rising demand, and which are cooling?Read the report
- AI Training Data Licensing DealsWho has licensed content and data to which AI companies, on what terms and for how much?Read the report
- The State of AI Training PayWhat does AI training work pay, by kind of work and by platform?Read the report