The Elon Musk vs OpenAI case is releasing some interesting bits of information about the early days of OpenAI — and how the company had valued its research talent in those times.
An email from Jared Birchall, Elon Musk’s wealth manager, dated September 11, 2017, reveals a proposed capitalization table that offers a fascinating glimpse into OpenAI’s planned transition from a non-profit to a for-profit structure. “I’ve attached a more user friendly version of the cap table that Ilya and Greg are proposing,” Birchall had written. The document, submitted as evidence in the ongoing lawsuit, shows how equity would have been distributed among founders, executives, and researchers had the proposal moved forward.
The Proposed Structure
Under the plan drafted by Ilya Sutskever and Greg Brockman, Elon Musk would have held a commanding 51.2% stake with a 100 million shares. The three co-founders who structured the deal—Sam Altman, Ilya Sutskever, and Greg Brockman—would each receive identical packages: a 3.33% investment stake plus a 7.68% grant, bringing their total to 11% each.
Board members John Schulman and Wojciech Zaremba (listed as “Woj” and “John” in the document) were slated for 2% grants each, while approximately 7.47% was earmarked for employee equity grants, with an additional 4.31% reserved for a future employee pool.
How Researchers Were Valued
The cap table provides a revealing snapshot of how OpenAI valued its technical talent in 2017. Among the researchers listed under “initial thoughts for employees”, some names stand out for their relatively modest equity allocations relative to their eventual impact on the AI industry.
Dario Amodei, allocated just 0.1% equity (195,312 shares), would go on to leave OpenAI in 2020 to co-found Anthropic, where he now serves as CEO. Anthropic has raised billions in funding and created the Claude family of AI models, making it one of OpenAI’s primary competitors.
Paul Christiano, also allocated 0.1%, departed OpenAI to found the Alignment Research Center, focusing on AI safety research. His work on reinforcement learning from human feedback (RLHF) became foundational to making language models like ChatGPT more useful and aligned.
Ilya Sutskever, OpenAI’s chief scientist at the time, was positioned as a co-founder in the proposal with 7.68% total equity. He remained at OpenAI until 2024, playing a crucial role in GPT model development before departing to start Safe Superintelligence Inc., focused on building safe AI systems.
Greg Brockman, also slated for 7.68%, continues to serve as OpenAI’s president. Sam Altman, with the same 7.68% stake in the proposal, remains CEO and has become one of the most prominent figures in AI.
The Polish Contingent
A notable cluster of researchers—Rafal Jozefowicz, Jakub Pachocki, Szymon Sidor, and Marcin Andrychowicz—were each allocated 0.6% equity. Pachocki now serves as OpenAI’s Chief Scientist following Sutskever’s departure, while the others have taken various paths in the AI industry.
Other researchers who received 0.3% allocations include names like Alec Radford (co-creator of CLIP and key contributor to GPT models) and Prafulla Dhariwal (instrumental in DALL-E development), both of whom have remained significant contributors to OpenAI’s research efforts.
A Deal That Never Was?
It’s not clear that this specific cap table ever materialized. OpenAI eventually created a “capped profit” structure in 2019 with different terms, and Musk departed from the board in 2018, citing conflicts of interest with Tesla’s AI work. The company he would have controlled with a majority stake is now valued at over $500 billion in recent private transactions.
The document offers a window into a road not taken—one where Musk would have maintained control of what became one of the most valuable AI companies in the world, and where researchers like Dario Amodei might have built transformative AI technology under OpenAI’s banner rather than as competitors. As things stand, many of the people listed in the document are no longer with OpenAI — but most are still playing an active role in shaping the broader AI revolution.