Who pays what.

Yes, with honest framing: for 38 Fortune 500 employers, Orbyt publishes the BLS national median wage for each occupation the company is known to hire, labeled with its occupation code and release. BLS publishes per occupation, never per employer, so these are occupation-level benchmarks rather than company-reported pay, and Orbyt does not model a company-specific figure for them.

A free CC BY 4.0 dataset covering 92 companies: each company's industry, Fortune 500 membership, the occupations Orbyt tracks for it, and the published BLS national median for each occupation from the two most recent OEWS releases. No modeled number ships in the file; Orbyt's estimated bands stay on the pages, labeled as estimates. Download it at orbytlabs.ai/company-salary-index.json.

Two kinds of figures appear on these pages, and each is labeled. Estimated bands and premiums are Orbyt-modeled over national medians; they are estimates, not reported offers. BLS occupation benchmarks are published government statistics for the occupation, not the employer. No figure on these pages is company-reported pay, and every page says which kind it is showing.

OpenAI has the highest average salary premium at +37% above national median, with its top-paying role (VP of AI) reaching an estimated $418,140. Meta FAIR follows closely at +35%.

On Orbyt's modeled premiums, pure AI labs like OpenAI and Anthropic sit 30-45% above national median for core AI roles, while Big Tech companies (Amazon, Microsoft, Apple) carry 15-25% premiums. These are estimates, not reported offers. The gap reflects intense competition for AI research and safety talent among smaller, well-funded labs.

OpenAI's estimated AI Engineer median salary is $186,200, reflecting a +40% premium over the national AI Engineer median of $133,000. This is base salary only and does not include equity, bonuses, or signing packages.

AI Safety Engineers and ML Engineers earn comparable base salaries at most companies. However, the supply of AI safety specialists is much smaller, which gives safety engineers stronger negotiating leverage. At companies like Anthropic where safety is a core focus, safety roles can command equal or higher premiums than general ML positions.

OpenAI has the highest average salary premium at +37% across 15 tracked AI roles. This premium reflects the intense competition for talent in the AI research and safety space, where a small number of companies compete for a limited pool of specialists.

On Orbyt's modeled premiums, Anthropic and OpenAI sit at similar levels across AI roles. OpenAI models slightly higher for research-focused positions, while Anthropic is competitive on safety and alignment roles. Both carry estimated premiums of 30-45% above national median for core AI engineering positions, among the highest modeled in the industry.