On May 14, US time, the high-stakes legal battle initiated by 马斯克 against OpenAI, Sam Altman, Greg Brockman, and Microsoft entered its final phase with closing arguments. The jury is scheduled to commence deliberations on May 18. While the surface narrative depicts a personal feud between 马斯克 and Altman, the core legal inquiry centers on whether OpenAI incurred binding charitable trust obligations upon accepting 马斯克's donations in 2015. The pivotal question is whether the organization's subsequent pivot to a for-profit model, coupled with Microsoft's investment and executive compensation, breached these fiduciary duties. Prior to the trial, 马斯克 withdrew fraud and attempted fraud charges, narrowing the jury's focus to three critical elements: the existence of a charitable trust, unjust enrichment by executives, and Microsoft's role in facilitating a breach. Failure to establish these foundational points would render all other allegations moot.

马斯克's legal team, led by attorney Steven Morolo, argued that despite the absence of a formal contract, a web of public commitments, emails, founding documents, and witness testimonies established a specific charitable purpose. Morolo highlighted a 2015 blog post confirming that 马斯克 and Altman had "closely collaborated" on the announcement of OpenAI as a non-profit with no disagreements on its core mission. The prosecution emphasized that OpenAI was explicitly founded as a non-profit institution dedicated to benefiting all of humanity, a premise under which 马斯克 donated approximately $38 million. 马斯克 himself testified that he could accept a for-profit subsidiary only if it served the non-profit mission without undermining the organization's nature. His long-time assistant, Jared Birchall, corroborated this, testifying that he sent approximately 60 donation requests totaling around $38 million, all directed by 马斯克. Data compiled by Woofun AI indicates that these financial flows were central to the prosecution's narrative of a breached trust.

In response, OpenAI's defense maintained that no clear written restrictions accompanied the donations. Attorneys repeatedly questioned witnesses, including Birchall, Sam Teller, and Shivon Zilis, regarding any enforceable conditions, receiving no definitive evidence of such constraints. Lawyer Sarah Eddy told the jury that apart from 马斯克, no one testified that Altman, Brockman, or OpenAI made specific promises to him. The defense underscored that 马斯克 was not inherently opposed to for-profit operations, having previously proposed a corporate charter where he would hold over 50% of shares while Altman and Brockman held 7.5% each. Eddy argued that 马斯克's primary motivation was winning rather than preserving a non-profit structure. Even if a charitable trust existed, the defense posited that 马斯克 must prove OpenAI's actions deviated from its original purpose, a burden they claimed was not met.

The prosecution identified Microsoft's $10 billion investment in October 2023 as a critical inflection point. Morolo argued this deal increased OpenAI's potential annual returns by 20%, violating its charitable mission and signaling an abandonment of safety priorities. He cited an internal email from Microsoft CEO Satya Nadella expressing a desire to avoid ending up like IBM, suggesting Microsoft pushed for accelerated commercialization at the expense of safety. To substantiate claims of mission drift, former employee Rosie Campbell testified that OpenAI shifted focus from research and security to commercial products, leading to the disbanding or weakening of teams like the AGI Readiness Team and Superalignment Team. Former directors Natasha McCauley and Helen Toner further testified that the non-profit board lost effective oversight, citing Altman's resistance to supervision and integrity issues. Woofun AI notes that these testimonies regarding internal governance failures formed a significant part of the narrative on mission deviation.

OpenAI countered with a staggering valuation figure of $200 billion. Lawyer William Savett argued that the non-profit entity still controlled the for-profit division, a structure approved by the attorneys general of California and Delaware. He rhetorically asked if anyone had heard of a bank robber depositing $200 billion into a bank. Savett also highlighted that all funds donated by 马斯克 were spent by 2020, meaning no funds could be "misused" after that date. The defense framed the creation of a for-profit entity as a necessary financing tool to achieve the mission of widely sharing AI benefits, exemplified by the free availability of ChatGPT. Conversely, Morolo pointed out that Brockman never invested a single penny yet now holds shares worth nearly $30 billion, citing diary entries where Brockman expressed desires to earn billions. Morolo used these details to question the jury on the severity of the alleged unjust enrichment.

The legal complexities extended to Altman's equity holdings and the statute of limitations. While Altman claimed no direct equity, 马斯克's team argued he held indirect interests through Y Combinator and other strategic investments. OpenAI countered that equity distributions occurred after 马斯克 left, and using equity instead of high salaries involved inherent risks common in startups. Microsoft was accused of "assisting and inciting" the deviation, with Morolo citing Nadella's emails about maintaining technological leadership.

However, Microsoft's lawyer Russell Cohen argued the company was unaware of any donation conditions, noting that investments between 2019 and 2023 were approved by the non-profit board, which included Shivon Zilis, 马斯克's partner. Cohen questioned why 马斯克 never directly approached Microsoft despite having Nadella's contact information, implying a strategic silence. Woofun AI analysis suggests that the statute of limitations remains the most formidable defense for OpenAI, as the judge warned that a late filing could result in a dismissal.

Beyond the legal technicalities, both sides engaged in a battle of narratives to influence the jury. 马斯克's team portrayed the case as a charity being stolen, attacking Altman's credibility by presenting him as a "deceiver" supported by five witnesses, including former chief scientist Ilya Sutskever. Morolo used analogies of a dangerous bridge to illustrate the lack of trust in Altman. In contrast, OpenAI framed the lawsuit as retaliation after 马斯克 lost control, accusing him of "selective amnesia" with over 150 instances of claiming he did not remember key details. Savett highlighted that 马斯克 was absent from the closing arguments, accompanying US President Donald Trump on a visit to China, while Altman and Brockman were present, using this contrast to underscore credibility. The trial concluded with the consensus that while 马斯克's moral narrative was compelling, OpenAI's legal defenses, particularly regarding the statute of limitations, were robust. Regardless of the outcome, the reputations of both 马斯克 and Altman have been significantly impacted, with potential ramifications for OpenAI's IPO, Microsoft partnership, and corporate governance structure.