Comparison

AI Data Buyers Compared: micro1 vs Polyshares vs Mercor vs HUMXN vs Protege

Sukhrobjon (Rob) Ismoilov
Sukhrobjon (Rob) Ismoilov
Founder & Principal, Main Street Wealth

Principal buyers, connector platforms, aggregators and advisors all call themselves the way to sell company data to AI. Who takes which side of the trade, what each publishes on price and payment, and which fits which seller.

Last Updated November 16, 2026
8 min read
10 cited sources
Share:
Quick facts
Published price listsmicro1 and Polyshares[1][2]
Connector-based collectionMercor[3]
Revenue share per useProtege[4]
Trades-specificHUMXN Research Labs[5]

Section 1

Who is on which side of the trade

TypeWho pays whomExamplesWhat it means for you
Principal buyerBuyer pays you; resells or licenses onwardmicro1, Polyshares, HUMXNOne counterparty; their paper; price set by them
Connector platformBuyer pays you on delivery to a labMercorFast, but runs through your vendors’ APIs
AggregatorRevenue share each time data is usedProtegeNo upfront price; depends on usage reporting
Sell-side advisorYou pay the advisor on successMain Street WealthRepresents you across buyers; costs a fee
Classification from each company’s own description of its model.

Polyshares draws the line itself: “We are not a marketplace, and we are not an agent… We take the other side of the trade.”[2] That is accurate for every principal buyer — and it is the reason the buyer’s first draft is written for the buyer.

Section 2

Side-by-side terms

micro1PolysharesMercorHUMXNProtege
Published price$100K–$500K / $500K–$3M / $3M+$100K–$2M typicalNone$20/hr footage; records priced individuallyRevenue share
Who qualifies30+ FTEs, several years of recordsAny company with real historyListed target industriesTrades, farming, manufacturingMedia, health, audio, spatial
PaymentWithin 30 days of acceptanceNet 30–60 after approval & anonymizationOn delivery to buyer, ~2–4 weeksNot statedPer use
CollectionNot specifiedReviewed before pricingConnectors to 100+ toolsSubmission & reviewLicensing agreements
Seller feeNone statedNoneNoneNot statedNot stated
From each buyer’s own pages and the cited reporting, read 16 November 2026. Terms change; confirm on the buyer’s page and in the draft.

Sources: micro1[1], Polyshares[2], Mercor[3], HUMXN[5], Protege[4].

Section 3

Which fits which seller

  • 30+ employees, several systems, years of history: principal buyers with published bands — get both micro1 and Polyshares to quote.
  • Heavy on Slack, Google Workspace or field-service software: be careful with connector routes; Slack, Google, ServiceTitan and Procore all restrict AI training on API data.[6][7][8][9] Offer an administrator export.
  • Trades footage and job photos: HUMXN is the only buyer with a published footage rate — mind the consent issues.
  • Large media or de-identified health datasets: Protege’s aggregator model is built for that.

Two quotes beat one. Business Insider reported at least one seller fielding offers from two buyers[10]; that is the position to be in.

Section 4

Our position

Frequently asked questions

Which companies buy business data for AI?
The buyers a private company is most likely to meet include micro1, Polyshares, Mercor, HUMXN Research Labs and Protege. micro1 and Polyshares are principal buyers with published bands; Mercor connects to your tools and pays on delivery; HUMXN focuses on the trades; Protege aggregates data and pays a revenue share.
What is the difference between an AI data broker and a buyer?
A principal buyer takes the other side of the trade and pays you. A broker or advisor represents a party for a fee. Several firms describe themselves as buyers precisely to say they charge no fee — which also means they are not working for you.
Which AI data buyer pays the most?
Nobody publishes outcomes. micro1’s published bands reach $3M+, Polyshares’ typical range is $100K–$2M, and Mercor publishes no price. The only meaningful comparison is written quotes on the same scoped record categories.

References

Sourced from primary industry reports, government data, SEC filings, and peer-reviewed valuation research. Retrieved at time of publication.

  1. 1.micro1 — Monetize your company’s operational data — Company Data Partnerships program page — A buyer’s own published payout bands, eligibility criteria, payment terms and data-handling commitments; marketing claims, not audited outcomes (read November 2026)
  2. 2.Polyshares — Sell your company’s operational data — program page — A buyer’s own published deal size, terms, payment timing and anonymization statements; marketing claims, read 16 November 2026 (read November 2026)
  3. 3.Mercor — Mercor Data — Monetize your data — A buyer’s own description of its connector-based extraction, compensation basis and payment timing; read 16 November 2026 (read November 2026)
  4. 4.Protege — Protege Raises $30 Million Led by a16z to Unlock Access to Data for AI Development — Announcement of 8 January 2026; total funding of $65 million and revenue-share payouts to data partners (2026)
  5. 5.HUMXN Research Labs — HUMXN Research Labs Wants to Buy $50 Million Worth of Skilled Trades and Manufacturing Data for AI & Robotics Training — Company release of 13 October 2026, also distributed by Newsfile (2026)
  6. 6.Slack Technologies, LLC — Slack API Terms of Service — Effective 10 October 2025; applications offered outside the developer’s organization may not train an LLM on API data or bulk export messages (2025)
  7. 7.Google — Google Workspace user data and developer policy — Last updated 3 September 2026; the Limited Use rules bar using user data to train generalized AI/ML models (2026)
  8. 8.ServiceTitan, Inc. — API Terms of Use — Last updated 15 April 2026; restrictions (31)–(34) cover AI training, AI inputs, synthetic data and transfer for AI purposes (2026)
  9. 9.Procore Technologies, Inc. — API Terms and Conditions — Effective 30 September 2025; bars training or benchmarking ML/AI on API Data and bulk export without an additional agreement (2025)
  10. 10.Business Insider — Micro1 is making a $1 billion bet on AI’s latest data craze — Seller interviews, the range micro1 says it pays, and a Gartner estimate of AI data spending, 26 October 2026 (2026)

Want a straight read on what your records are worth?

Send us the systems, years and rough record counts. We tell you what is licensable, get competing written quotes, and you pay nothing if buyers pass.

Request the evaluation form
Calendar

This article is provided for general educational purposes and is not legal, tax, investment, or accounting advice. Multiples, fee ranges, and timelines are illustrative ranges compiled from third-party sources current as of November 16, 2026 and will vary by transaction and market conditions. Consult qualified counsel before acting on any information contained here.