Report as of 2026-07-23 PT · Source: Metix AI
The three labs do not share one prior-employer story. OpenAI’s Enterprise SaaS plus Big Tech prior employers are 51.7% combined. Anthropic’s Enterprise SaaS share is 35.3%. At xAI, a previous stop at X is the largest single prior employer at 5.1%. These are source mixes inside the three-lab US profile sample, not the OpenAI technical-staff prior-employer table, and not official headcount. Do not add 51.7%, 35.3%, and 5.1% across companies.
Why it matters
A recruiter or TA lead can brief this cut before opening a search, and a VC can use it to test whether the bench matches the story. On this page, OpenAI: Enterprise SaaS + Big Tech is 51.7%, and Anthropic: Enterprise SaaS is 35.3%. Those figures are the published visible sample, not official headcount.
Related numbers from the parent map
- 9,421Public professional profilesData as of 2026-07-23 PT
- ≈40%Engineering share at all three companiesData as of 2026-07-23 PT
- 5.83×xAI Human Data & Evaluation over-indexData as of 2026-07-23 PT
- 1.58×Anthropic GTM / Customer over-indexData as of 2026-07-23 PT
Map the prior-employer mix company by company
OpenAI Enterprise SaaS plus Big Tech is 51.7%. Anthropic Enterprise SaaS is 35.3%. xAI’s largest single prior stop, X, is 5.1%.
Prior-employer mix at the three labs
| Measure | Figure |
|---|---|
| OpenAI: Enterprise SaaS + Big Tech | 51.7% |
| Anthropic: Enterprise SaaS | 35.3% |
| xAI: previous X, largest single prior stop | 5.1% |
Source: OpenAI, Anthropic and xAI Talent Structure Benchmark · Metix AI · report date 2026-07-23
Notes & method
Scope and limits
- Public profiles may contain update lags, self-presentation bias, and inconsistent titles.
- Figures are a visible sample for relative comparison. They are not official headcount and are not used to estimate a global workforce.
- Do not add these three percentages. Each one is a within-company source mix.
Back to OpenAI, Anthropic and xAI Talent Structure Benchmark
FAQ
Can I add 51.7%, 35.3%, and 5.1%?
No. Each percentage is that company’s own source mix.
Is 5.1% most of xAI?
No. It is the largest single prior stop, not a majority.
Is this the same table as Google 702?
No. Google 702 is the OpenAI technical-staff map. This page is the three-lab source mix.
Is this official headcount?
No. It is a visible-sample source mix.
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 https://metix.ai/reports/questions/openai-anthropic-xai-sources. The parent map is https://metix.ai/reports/mapping/openai-anthropic-xai-talent-structure. Report date 2026-07-23.
Notes & method
How to cite
Metix AI. (2026). Where do OpenAI, Anthropic and xAI hire from? Metix AI Talent Intelligence. https://metix.ai/reports/questions/openai-anthropic-xai-sources
Ask Metix AI to map these source mixes
Open a Metix AI workspace on the same three-lab source cut. Report date 2026-07-23.
Start a Mira trial