Key Takeaways
- Elon Musk attempted to hire OpenAI’s founding team in 2018 to lead a new AI lab within Tesla.
- The proposed AI unit would have been led by Sam Altman, Greg Brockman, and Ilya Sutskever.
- Musk wanted to appoint Altman to the board or make OpenAI a Tesla subsidiary.
- The disclosure sheds light on a crucial issue in the high-stakes trial between Musk and OpenAI.
- OpenAI’s lawyers argue that Musk was happy to commercialize the lab, provided he remained in charge.
On May 8, 2026, a high-stakes trial between Elon Musk and OpenAI, the maker of ChatGPT, revealed that Musk attempted to hire OpenAI’s founding team in 2018 to lead a new AI lab within Tesla. The proposal, which included the appointment of Sam Altman to the board or making OpenAI a Tesla subsidiary, is a crucial issue in the ongoing trial.
Background on the Trial
The trial is centered around Musk’s claim that Altman “stole a charity” by converting OpenAI into a for-profit organization. OpenAI was initially established in 2015 as a nonprofit with a mission to ensure artificial general intelligence (AGI) benefits all of humanity. Musk was a co-founder and early funder, contributing millions in initial capital and lending his name to the project. His involvement helped attract top-tier AI talent and credibility in the tech industry.
But by 2018, tensions were growing. Musk had pushed hard for tighter integration between OpenAI and Tesla, arguing that the two organizations could accelerate progress if they shared resources and leadership. He believed AI development should serve Tesla’s autonomy goals first. Others on the OpenAI board felt the organization needed to remain independent to avoid being steered by any single corporate interest, especially one as demanding as Tesla’s.
The legal dispute now hinges on what Musk thought OpenAI would become. His lawyers argue that the original agreement included a path to commercialization — but only under his oversight. OpenAI’s current leadership disputes that interpretation. Internal emails submitted in the trial suggest Musk was open to a for-profit model as long as he retained control over strategy and key appointments. When it became clear that wouldn’t happen, he stepped down from the board in 2018 and cut off further funding.
Historical Context
The 2018 proposal didn’t come out of nowhere. It was the culmination of years of collaboration and growing friction between Musk and the OpenAI team. After launching in December 2015, OpenAI quickly gained recognition for publishing high-impact AI research, including early work on reinforcement learning and language models. Musk, already deep into Tesla’s Autopilot development, saw AI as the central battleground for the future of transportation — and beyond.
By 2017, Tesla was already collecting vast amounts of driving data from its fleet, and Musk believed OpenAI could help build a general learning system capable of mastering real-world environments. He pushed for OpenAI to shift focus toward robotics and autonomous systems. Altman and Sutskever, however, were more interested in foundational AI research — building systems that could generalize across tasks, not just drive cars.
The divergence in vision became clear during a series of meetings in early 2018. Musk proposed a formal partnership: OpenAI would become a research arm of Tesla, housed under the same corporate umbrella. Altman would join Tesla’s board. Brockman would lead engineering integration. Sutskever would oversee core AI development. In return, OpenAI would gain access to Tesla’s computational resources, real-world data, and engineering teams.
The proposal was rejected. OpenAI’s board feared losing independence. They also questioned whether Tesla’s immediate product needs would distort long-term research goals. The split wasn’t public at the time, but it marked a turning point. By mid-2018, Musk had left the board, and OpenAI began laying the groundwork for a hybrid structure: a capped-profit entity under the OpenAI umbrella, designed to attract investment while preserving the nonprofit mission.
Musk’s Proposal
Musk’s proposal to hire OpenAI’s founding team is revealed in evidence submitted in the trial. According to the evidence, Musk wanted to bring Altman, Brockman, and Sutskever to Tesla to lead the new AI lab. The plan wasn’t just about personnel — it was about control. Musk envisioned a vertically integrated AI operation, where research, product development, and deployment happened under one roof.
The proposal included two pathways: either OpenAI would become a subsidiary of Tesla, or its top leaders would join Tesla directly and launch a new AI division, effectively recreating OpenAI’s mission inside the carmaker. Musk believed this would speed up development. He argued that OpenAI’s nonprofit model was too slow, too cautious, and ill-equipped to compete with deep-pocketed tech giants like Google and Facebook.
Altman and the others disagreed. They believed that keeping AI research separate from product timelines allowed for more ambitious, long-term work. They were already exploring models that could learn across domains — a concept that would later lead to GPT-3 and beyond. They didn’t want those efforts tied to quarterly vehicle production targets or battery cost metrics.
Implications for OpenAI and Tesla
The disclosure has significant implications for both OpenAI and Tesla. If Musk’s proposal had been accepted, it could have marked a significant shift in the direction of OpenAI’s research and development. It could also have given Tesla a major advantage in the development of AI-powered technologies.
For OpenAI, joining Tesla would have meant faster access to real-world data and hardware. Tesla’s fleet of millions of vehicles generates petabytes of sensor data — a goldmine for training autonomous driving systems. But it also would have meant subordinating research priorities to Tesla’s engineering roadmap. That might have accelerated progress in robotics and self-driving, but it could have delayed or derailed the development of large language models.
For Tesla, gaining control of OpenAI’s core team would have supercharged its AI capabilities. Instead of building an in-house AI team from scratch, Tesla could have inherited one of the strongest research groups in the world. The company might have achieved full self-driving years earlier — or pivoted into consumer AI products much sooner.
But there’s another angle: public trust. OpenAI’s independence became a selling point when it released ChatGPT in 2022. Users and developers believed the organization was focused on broad accessibility and safety, not tied to a single corporation’s profit motives. That perception helped it gain rapid adoption. Had OpenAI been seen as a Tesla subsidiary, its credibility as a neutral AI pioneer might have been undermined.
Key Dates and Details
- 2018: Musk attempts to hire OpenAI’s founding team to lead a new AI lab within Tesla.
- 2015: OpenAI is founded as a nonprofit by Musk, Altman, Brockman, Sutskever, and others.
- 2017: Tensions grow over OpenAI’s research direction and relationship with Tesla.
- 2018 (mid-year): Musk leaves OpenAI’s board after proposal is rejected.
- 2019: OpenAI announces a capped-profit model to raise outside funding.
- May 2026: The trial reveals the proposal, shedding light on the high-stakes case.
- Tesla: The carmaker has been investing heavily in AI research and development.
- OpenAI: The organization has been at the forefront of AI research, developing technologies like ChatGPT.
What This Means For You
The implications of Musk’s proposal for OpenAI and Tesla are significant. For developers and builders, the development of AI-powered technologies is a key area of focus. The outcome of the trial, and the potential changes to OpenAI’s research and development direction, could have major implications for the development of AI-powered technologies in the future.
Consider a startup founder building an AI assistant for customer service. If OpenAI had become a Tesla subsidiary, its API might have prioritized robotics and vehicle integration over language models. That could have delayed the release of GPT-3 or limited public access. The tools many startups rely on today — fine-tuning models, using embeddings, accessing real-time inference — might not exist in their current form.
For a software engineer working on autonomous systems, this history is a cautionary tale about organizational independence. Tesla has since built its own powerful AI team, especially in computer vision and motion planning. But it lacks the broad AI research capacity of OpenAI. That gap shows in areas like natural language interaction in vehicles. While Tesla cars can respond to voice commands, they don’t engage in conversational AI like some competitors. That limitation traces back to 2018 — the year Tesla didn’t get OpenAI’s brain trust.
For investors, this case highlights how early governance decisions shape tech outcomes. The choice to keep OpenAI independent allowed it to partner with Microsoft, raise billions, and move fast on product development. Had it folded into Tesla, it might have been constrained by Musk’s priorities and Tesla’s financial cycles. The $90 billion valuation OpenAI reached by 2025 might never have materialized.
Looking Forward
The trial is ongoing, and the outcome is still uncertain. However: the development of AI-powered technologies is a key area of focus for both OpenAI and Tesla. The outcome of the trial could have major implications for the development of AI-powered technologies in the future.
The court’s decision may clarify the legal boundaries of nonprofit governance, especially when high-profile founders are involved. If Musk wins, it could set a precedent that co-founders retain control rights even after stepping away. If OpenAI prevails, it will reinforce the idea that mission-driven organizations can evolve beyond their founders’ original vision.
What Happens Next
The trial will likely continue for several more weeks. Both sides are expected to call additional witnesses, including former OpenAI board members and Tesla executives. The judge has signaled that internal communications — emails, messages, meeting notes — will play a central role in determining intent.
One possible outcome is a negotiated settlement. Musk may seek a public acknowledgment of his role in OpenAI’s founding, or a financial settlement tied to past contributions. OpenAI, for its part, may want to avoid a ruling that undermines its current structure or deters future investment.
Regardless of the verdict, the revelations have already reshaped public understanding of how OpenAI came to be. The 2018 proposal was a pivot point — a moment when the future of AI could have gone in a very different direction. That it didn’t suggests that independence, not integration, was the condition that allowed generative AI to explode when it did.
The trial may end, but the debate won’t. How much control should any one person have over significant technology? Can mission-driven research survive inside a product-driven company? And what happens when the people building the future can’t agree on what it should look like? Those questions are just getting started.
Sources: Ars Technica, original report
A dimly lit conference room with a whiteboard in the center, covered in complex equations and diagrams related to AI research. A single, overhead light casts a cold glow over the room, highlighting the intensity of the conversations taking place around the table.
