$350M/year Mercor for human personality

TL;DR: As Mechanical Turk became gamed by bots, Prolific built a provenance-first worker marketplace for academic research with longitudinal records of personality, demographics, and cultural fluency. That data is now its bet for the next wave of AI training, the companion, therapist, and education use cases where human personality matters more than credentials. Sacra estimates Prolific hit ~$350M in annualized revenue in April 2026. For more, check out our full report and dataset on Prolific as well as our interview with COO Jemma White.

Key points via Sacra AI:

As Amazon’s Mechanical Turk (launched 2005) became increasingly gamed by bots and “professional Turkers” with keyboard macros & autoclickers, Prolific (2014) launched as a provenance-first alternative for academic researchers gathering data, with ID-verified participants & transparent fees—then shifted focus to AI data labeling & training (2023) amid unprecedented demand for verified human workers from frontier labs. Unlike Mechanical Turk, which obscured worker pay inside its fee structure, customers on Prolific’s early marketplace paid contractors directly, with Prolific charging a platform fee on top (33-43% of total pay-out), but in 2024, its revenue mix shifted towards frontier lab training where participant pay, pool curation, QA, and operations are bundled into single retainer-scale engagements.
With frontier labs, AI app developers, and enterprises racing to find enough verified human workers that could be segmented by demographics, matched & deployed at scale to generate preference data, red-team models, test cultural fluency, and evaluate frontier models, Sacra estimates Prolific hit roughly $350M in annualized revenue in April 2026, including revenue share with payments to contractors. Sacra estimates Handshake ($1.1B in annualized gross revenue in April 2026, up 349% YoY from about $245M in April 2025), Mercor ($1B in annualized gross revenue, up 897% YoY, $10B valuation), Scale AI ($1.5B in annualized revenue, up 97% YoY, $29B implied valuation at the Meta acquisition) and Surge ($1.2B in revenue)—along with Turing ($300M in 2024, up 150% YoY, $2.2B valuation), micro1 ($300M annualized in April 2026, up 140% YoY, $500M valuation), and Invisible ($134M in 2024, up 123% YoY, $2B valuation).
Just as AI training shifted from anonymous crowd workers (Scale AI) to verified, PhD-level experts (Mercor/Handshake) with the advent of reasoning models, Prolific is betting on its longitudinal record of human attributes—personality, cultural fluency, political priors, resilience, age, geography—to support assistant & companion use cases that index on bedside manner including customer support, executive assistant, therapist & medical, and tutor & education. Underlying moment-to-moment foundation model demand for particular demographics or credentialed human annotators lies the long-term trend of increased AI regulation, requiring human data (vs. synthetic) and 3rd party annotation (vs in-house), advantaging platforms like Prolific that have a track record on trust, operate independently and maintain deep supply-side participant pools.

For more, check out this other research from our platform:

$350M/year Mercor for human personality

Read more from

Prolific

Prolific revenue, growth, and valuation

Jemma White, COO of Prolific, on why humans ensure AI safety

Read more from
#b2b

AirOps revenue, growth, and valuation

$108M/year ZoomInfo for PR

Muck Rack revenue, growth, and valuation

Read more from
#gig-marketplaces

$1.1B/year Indeed for data labelers

Handshake vs Mercor

Handshake revenue, growth, and valuation

Read more from
#data-labeling

micro1 revenue, growth, and valuation

Surge AI revenue, growth, and valuation

Turing revenue, growth, and valuation

Create a free account, or log in.

Free article limit reached.

Standard membership required.