Reports / What share of xAI is Human Data & Evaluation?

Q&A · report date 2026-07-23

What share of xAI is Human Data & Evaluation?

Human Data & Evaluation is 28.4% of xAI’s U.S. role mix: 425 of 1,496 public professional profiles, a 5.83× over-index versus the three-lab mix. Research / Models is 2.1%. The role mix sits in a 9,421-profile U.S. sample of OpenAI, Anthropic, and xAI, report date 2026-07-23. xAI’s distinctive non-Engineering concentration is evaluation and data operations, so talent acquisition should source that bench differently than it sources a research-heavy lab. The 28.4% full classification is a different range from the consistent cross-market share (U.S. 23.3%, 348 of 1,496; 15 priority international markets 34.4%, 129 of 375). Those measures are not additive, and none of them is official headcount.

Source evaluation and data-ops talent, not a research-lab clone

Human Data & Evaluation is 28.4% of xAI’s U.S. role mix (425 of 1,496). The full structure is in the parent map; the named list stays on this ask.

xAI Human Data & Evaluation share

xAI’s U.S. role mix in the full classification, plus the narrower cross-market Human Data & Evaluation measure. The two ranges are not additive. Report date 2026-07-23.
MeasureFigure
Human Data & Evaluation28.4% (425 of 1,496)
Over-index vs the three-lab mix5.83×
Research / Models2.1%
U.S. sample, three labs9,421
xAI U.S. profiles in that sample1,496
Consistent cross-market measure, U.S.23.3% (348 / 1,496)
Consistent cross-market measure, 15 priority international markets34.4% (129 / 375)

Source: OpenAI, Anthropic and xAI Talent Structure Benchmark · Metix AI · report date 2026-07-23

Scope and limits

Back to the OpenAI, Anthropic and xAI talent structure map

FAQ

Is 28.4% official xAI headcount?

No. It is 425 of 1,496 U.S. public professional profiles in the full role classification, inside a 9,421-profile U.S. sample. It is not official headcount.

Is Human Data & Evaluation the same as Research / Models?

No. Research / Models is 2.1%. Human Data & Evaluation is the evaluation and data-operations concentration, at 28.4% and a 5.83× over-index.

Can 28.4% be added to the 23.3% U.S. share or the 34.4% international share?

No. The parent report says the full role classification and the consistent cross-market definition cover different ranges and are not additive. The consistent measure is 23.3% in the U.S. (348 of 1,496) and 34.4% across 15 priority international markets (129 of 375).

What should talent acquisition do with this?

Treat xAI’s distinctive non-Engineering concentration as evaluation and data operations, and source that bench differently than a research-heavy lab. The parent action is to cover Academia / Research, Auto / Hardware / Robotics, and X, focusing on people who can turn domain expertise into evaluation and data workflows.

Where is the method, and how do I cite this figure?

The method page is https://metix.ai/reports/methodology. Cite this answer with this page’s URL. The parent map is https://metix.ai/reports/mapping/openai-anthropic-xai-talent-structure. Report date 2026-07-23.

How to cite

Metix AI. (2026). What share of xAI is Human Data & Evaluation? Metix AI Talent Intelligence. https://metix.ai/reports/questions/xai-human-data-share

Ask Metix AI to map this role mix

Open a Metix AI hiring workspace on the same Human Data & Evaluation cut. Report date 2026-07-23.

Start a Mira trial

Methodology · 中文

中文