Evidence So Far in Musk v. Altman Trial
The Musk v. Altman trial began its jury phase on Monday in a California federal courtroom. Exhibits presented include emails, photos, and corporate records from OpenAI's founding period and even earlier. These materials offer views into the AI lab's origins and initial conflicts.
Key points from the evidence: Nvidia CEO Jensen Huang supplied OpenAI with a sought-after supercomputer. Elon Musk shaped much of OpenAI's mission statement and its starting organization. OpenAI CEO Sam Altman sought strong backing from Y Combinator at the outset. OpenAI president Greg Brockman and researcher Ilya Sutskever expressed worries about Musk's degree of influence. Musk stressed the value of a nonprofit focused on AI that helps everyone.
Musk's suit targets Altman, Brockman, and investor Microsoft. It raises issues like violating OpenAI's charitable trust, fraud, and unjust enrichment. At its core, the case questions if OpenAI strayed from its original goal of making artificial general intelligence benefit humanity. Artificial general intelligence refers to AI matching or exceeding human smarts.
This marks the newest step in Musk's extended legal challenges against OpenAI and its leaders. Musk helped start the lab with Altman and Brockman and invested early. He now runs xAI, a rival AI outfit under SpaceX.
Former OpenAI staff and insiders from both sides watch closely. A jury decision could shape OpenAI's operations and its fast-evolving tech. Both OpenAI and SpaceX plan public listings this year, drawing extra attention.
Discovery already surfaced notable messages, such as Altman-Sutskever emails, Brockman's diary notes, and texts between Meta CEO Mark Zuckerberg and Musk. The trial brings fresh disclosures.
Below is a full rundown of public exhibits, with asterisks on standout ones. More will appear as the trial proceeds.
Documents Released April 29, 2026
Exhibit No. 5: A June 2015 email thread between Altman and Musk. Altman outlines a five-part strategy for an AI lab aimed at "create the first general AI and use it for individual empowerment, ie, the distributed version of the future that seems the safest. More generally, safety should be a first-class requirement." He proposes starting with seven to 10 staff and growing, using a spare Y Combinator site in Mountain View. For oversight, he lists five: himself, Musk, Bill Gates, Pierre Omidyar, Dustin Moskovitz. "The technology would be owned by the foundation and used 'for the good of the world', and in cases where it's not obvious how that should be applied the 5 of us would decide," Altman states. Researchers get "significant financial upside … uncorrelated to what they build, which should eliminate some of the conflict," plus competitive pay and Y Combinator equity. He suggests a team leader outside the board. Altman asks if Musk will join beyond oversight, maybe visiting monthly or aiding recruitment publicly, like Peter Thiel at Y Combinator. He notes a "regulation letter" for AI rules, offering to exclude Musk publicly. Musk answers, "Agree on all."
Exhibit No. 7: October 2015 emails between Altman and Musk. Altman proposes Musk commit $100 million upfront and $30 million more over five years. Gates is not yet pledged but expected soon. Zuckerberg may skip due to Facebook AI Research. Altman suggests he and Musk form the initial Safety Board, adding up to three later as the "'second key' for releasing anything that could be dangerous." Musk replies, "Let's discuss governance. This is critical. I don't want to fund something that goes in what turns out to be the wrong direction."
Exhibit No. 12: November 2015 emails between Musk and Altman on lab plans. Musk praises a call with Brockman, calls the team strong. He pushes for an independent 501c3 nonprofit with clear focus on widely shared strong AI. Revenue above costs goes to reserves. Pay includes salary and bonuses; cash could convert to Y Combinator or SpaceX stock. Musk offers vast Tesla sensor data, far more than rivals. He suggests "Freemind" as name, opposing DeepMind's control model. Musk commits time despite Tesla and SpaceX demands, saying belief in AI risk requires action. He admits full involvement publicly. Altman proposes sharing Y Combinator space and legal help, names like "Axon" or Turing-linked. Musk likes Turing ideas but avoids test link.
Exhibit No. 14: December 2015 emails drafting OpenAI's mission and release. Musk aims to draw talent, notes low pay but right goal and setup. Altman stresses no financial duties allow max human impact and broad AI spread. Final text stays close to Musk's draft.
Exhibit No. 16: OpenAI incorporation papers filed December 8, 2015. It sets the group as nonprofit for charity, to "ensure that artificial general intelligence benefits all of humanity," via research, safe development, distribution, societal study, policy. Tech benefits public, no private gain.
Exhibit No. 70: April 2016 emails between Musk and Jensen Huang. Musk seeks early supercomputer for OpenAI, notes its independence from Tesla, nonprofit funded by him and others for safe AGI. Huang agrees to prioritize one.
Exhibit No. 388: Photo of Huang delivering the computer, Musk present. Wall quote links to Rickover, echoed in Altman's 2013 post.
Exhibit No. 152: August 2017 emails with Shivon Zilis, Musk's aide and later board member. Zilis recaps Brockman-Sutskever meeting: flexible on Musk's time/control mix, need over $100 million for data center, equal equity push, control fears. Key: no absolute control, distribute power after 2-3 years regardless of founders' fates. Musk replies annoyed, suggests they start own firm.
Exhibit No. 153: September 2017 email from Jared Birchall with cap table: Musk 51.20%, Altman/Sutskever/Brockman 11.01% each, employee reserves.
Documents Released April 30, 2026
Stay updated
Get the day's AI and automation news in your inbox. No spam, unsubscribe anytime.
Exhibit No. 25: November 2015 emails. Altman mentions early "Y Combinator AI" nonprofit idea with bylaws for board consent on risky releases. Musk prefers independent from YC, standard C corp plus nonprofit.
Exhibit No. 559: December 2016 Musk email to Neuralink team worries OpenAI lags DeepMind, nonprofit setup may hinder urgency.
Exhibit No. 773: June 2017 Musk email on hiring Andrej Karpathy to Tesla Vision, expects OpenAI backlash.
Exhibit No. 631: July 2017 Musk to Sutskever/Brockman on China threat, suggests course change. Brockman agrees: nonprofit to 2017 end, for-profit research/hardware 2018, government later. Musk offers Tesla Model 3s.
Exhibit No. 642: August 2017 Musk offers meeting at his new SF mansion.
Exhibit No. 662: August 2017 Birchall holds $5 million donation; Musk approves.
Exhibit No. 686: September 2017 Sutskever proposes Musk three board seats, others one each. Musk wants four appoint rights, aims for large expert board.
Exhibit No. 679: September 2017 Brockman proposes cap table; Musk rejects as too aggressive.
Exhibit No. 691: September 2017 Musk text on starting OpenAI B Corp.
Exhibit No. 158: September 2017 group email shows negotiation peak. Brockman/Sutskever to Musk: want to work together, yield equity/control, but fear his absolute AGI power. Cite CEO insistence despite dislike. To Altman: question motives, CEO push, politics. Altman affirms nonprofit enthusiasm.
Exhibit No. 157: Musk's reply: enough, fund nonprofit or go alone.
Exhibit No. 159: Zilis on Altman's lost trust in Brockman/Sutskever, hiatus plan, Karnofsky nonprofit push.
Exhibit No. 719: October 2017 Musk okays Neuralink hiring from OpenAI.
Exhibit No. 98: January 1, 2018 Sutskever thanks Musk as top competent, for hardware push.
Exhibit No. 99: Brockman echoes honor working with Musk.
Exhibit No. 749: January 2018 emails on DeepMind lead. Musk warns failure path, rejects ICO, pushes expansion or Tesla AI. Forwards Karpathy note: OpenAI cash burn can't match Google; for-profit risks; best Tesla tie for scale via self-driving revenue.
Exhibit No. 761: February 2018 Musk-Zilis texts post-board exit: stay friendly, recruit 3-4 to Tesla. Zilis notes Sutskever upset, urges slowing DeepMind's Hassabis.
Exhibit No. 233: April 2018 Musk-Zilis on OpenAI funding likely led by Reid Hoffman.