AI gig work news: October 2026

Short answer: October 2026 brought one large funding round for an AI data company (Labelbox, over $100 million), a $1 billion data-buying commitment from micro1, new research from Mercor showing AI beating licensed accountants on a benchmark, and a fall in US layoff announcements, with AI still the top reason for job cuts this year. On our own board, 395 roles added in October are still open and 318 closed. Every item below links to its source; figures a company states about itself are labelled that way.
This is the first edition of a monthly roundup. Each month we collect what changed for people who do AI training work: who raised money, who cut jobs, which platform rules changed, and what the listings themselves show. For the running totals across months, see our AI gig work statistics page.
What happened in October 2026, at a glance
| Date | What | Source |
|---|---|---|
| 1 Oct | US employers announced 43,281 job cuts in September; AI was cited for 3,961 of them, and for 120,136 (about 21%) so far in 2026 | Challenger, Gray & Christmas |
| 1 Oct | Mercor publishes a study: Claude Opus 5 scored 100% on simplified accounting tasks where 12 licensed CPAs averaged about 37% | Mercor blog |
| 8 Oct | Mercor announces a research team to measure AI on real work and build expert data where models fail | Mercor blog |
| 9 Oct | micro1 commits $1 billion over 12 months to buy and license enterprise operational data, financed by Citi and Hercules Capital | micro1 |
| 14 Oct | Labelbox raises over $100 million in a Series E at a $1.25 billion pre-money valuation, led by Infinitum Partners | Labelbox |
| 15 Oct | Mercor and NVIDIA report specialising Nemotron 3.5 Super for legal, software and accounting work with small, targeted data | Mercor blog |
| 27 Oct | Mercor, Stanford, the O*NET Resource Center and NBER release TaskNet, an open-source catalogue of work tasks | Mercor blog |
Who raised money in October?
Labelbox announced on 14 October that it raised more than $100 million in a Series E, made up of equity and a credit facility, at a $1.25 billion pre-money valuation. Infinitum Partners led, with Wipro Ventures, Andreessen Horowitz, B Capital, Tau Ventures and Snowpoint Ventures taking part. Labelbox says it serves more than 400 customers, including over 90% of leading US AI labs, and that it became profitable in 2026. Those are the company's own figures. Source: Labelbox's announcement and press release.
micro1 did not announce an equity round in October, but on 9 October it said it will spend $1 billion over the next 12 months acquiring and licensing de-identified enterprise data, to build reinforcement learning environments that mirror real business operations. Citi and Hercules Capital provide the capital. The announcement says nothing about hiring experts or contractors, so we do not read it as a hiring signal. Source: micro1. In late September, Forbes reported, citing people familiar with the deal, that micro1 raised over $100 million at a $4 billion valuation; micro1 has not announced it.
Mercor announced no new round in October. Its last confirmed round is the $350 million Series C at a $10 billion valuation from October 2025 (TechCrunch). Reports of talks at a higher valuation have not been confirmed by a closed round, so we leave them out.
How many jobs were cut, and how much of it was AI?
Challenger, Gray & Christmas, which publishes the most-watched monthly count of US layoff announcements, reported on 1 October that employers announced 43,281 job cuts in September, down 18% from August and the lowest September since 2022. Technology companies announced 10,799 of them. AI was the fifth most cited reason in September, with 3,961 cuts (about 9%), but it remains the leading reason for the year: 120,136 announced cuts so far in 2026, about 21% of all cuts. Source: Challenger Report, September 2026.
These are announced plans, not completed layoffs, and they cover US employers only. We found no October report of layoffs at an AI data platform itself. If you were affected by a project ending on a platform, that rarely shows up in counts like this, because contractors are not employees.
What changed for AI trainers at Mercor and micro1?
Mercor's October posts point in one direction: more expert data in specialised professional work. Its accounting study compared 12 licensed CPAs, averaging about five and a half years of experience, with AI models on four simplified month-end close tasks. The accountants averaged about 37%; Claude Opus 5 scored 100% on all 20 attempts. Mercor's own caveat is that the tasks tested file searching and instruction following, not the whole job. Its NVIDIA post describes improving a model for legal, software and accounting work with small, targeted data, which is the kind of work domain experts on Mercor are hired to produce.
For people on the platform, the practical reading is that demand is shifting toward experts who can judge whether an answer in their field is correct, and away from generic labelling. That matches what we see on our board, below. If you are preparing for a Mercor interview, use the free Mercor practice questions and check the payout calendar.
What our job board showed in October
We track live listings from Mercor, micro1 and Turing and check each one against the platform. On 1 November we had 868 open roles. Of the roles still open, 395 were added to our board in October. During October and our 1 November check, 318 listings closed and were marked expired, so they no longer appear in search or our sitemap.
| Platform | Open roles on 1 Nov | Added in October, still open | Closed in October |
|---|---|---|---|
| micro1 | 346 | 254 | 136 |
| Mercor | 293 | 85 | 88 |
| Turing | 229 | 56 | 94 |
| Total | 868 | 395 | 318 |
Most of the new roles were model evaluation work (237 of the 395), followed by coding (57), research (49) and data annotation (47). The highest stated hourly rates on the board were specialist roles: a Mercor oncologist role listing $350 to $450 an hour, and several micro1 attorney roles listing $140 to $400 an hour. Across all live listings that state an hourly rate, the typical range is $60 to $90 an hour. The full numbers, with a CSV download, are in our AI training pay dataset.
One pattern is worth knowing if you apply: listings close fast. 318 roles closed on our board in a single month, so apply early rather than waiting for a role to still be there next week. Browse professions with the most open roles, or see the Mercor and micro1 application guides.
What to watch in November
- The Challenger report for October, due in early November, will show whether AI-attributed cuts rose again.
- Whether micro1 confirms the round Forbes reported, and whether its data-buying commitment changes the kinds of roles it posts.
- Whether Mercor's research push shows up as more benchmark-style expert roles in accounting, law and medicine.
- How many October roles close in November. We will report it in the next edition.
Common questions
- Did any AI training platform lay off contractors in October 2026? We found no reported layoff at an AI data platform in October. The largest documented case remains Scale AI in July 2025: about 200 staff and 500 contractors.
- How much did Labelbox raise? More than $100 million in a Series E, at a $1.25 billion pre-money valuation, announced 14 October 2026.
- Is micro1 hiring because of its $1 billion commitment? Its announcement is about buying enterprise data and does not mention hiring experts, so we cannot say.
- How much of 2026's US job cuts are blamed on AI? About 21%: 120,136 of 573,195 announced cuts through September, according to Challenger.
- How many AI training jobs are open? 868 on our board on 1 November 2026, across Mercor, micro1 and Turing.
Sources and method
Each item comes from the company's own announcement, a research report, or a named outlet, linked inline. Anything a company says about itself (revenue, customers, profitability) is the company's claim, not verified by us. Our board figures come from our own checks of each listing against Mercor, micro1 and Turing on 1 November 2026. You may quote this roundup with a link; the running statistics are on the AI gig work statistics page.
AI Work Expert is an independent publisher, not affiliated with any platform named here. Some apply links on our job pages are referral links, which does not change what we report.


