MUSK v. ALTMAN (2026)

ORDER DENYING OPENAI DEFENDANTS' MOTION FOR SUMMARY JUDGMENT; GRANTING IN PART AND DENYING IN PART MICROSOFT'S MOTION FOR SUMMARY JUDGMENT

This action arises from a contentious dispute between Elon Musk and Samuel Altman (and their respective companies) over large, charitable donations that Musk made to OpenAI, Inc. during the company's infancy and Microsoft's later investments in OpenAI, Inc.'s for-profit ventures. Defendants Altman, Greg Brockman, and the OpenAI entities ¹ (collectively, “OpenAI defendants”) move for summary judgment on Musk's claims for breach of charitable trust, constructive fraud, promissory fraud, and unjust enrichment. (Dkt. No. 327.) Microsoft separately moves for summary judgment on Musk's claims for tortious interference with contract, aiding and abetting breach of fiduciary duty, and unjust enrichment. (Dkt. No. 329.)

The Court, having carefully considered the parties' pleadings, briefs, evidence, and arguments at the January 7, 2026 hearing, and as explained below, DENIES OpenAI defendants' motion, GRANTS Microsoft's motion as to Musk's claims for tortious interference with contract and unjust enrichment and DENIES the balance of Microsoft's motion.²

I. BACKGROUND

A. FACTUAL BACKGROUND

Unless otherwise noted, the factual record set forth below is undisputed.³

Musk, Altman, Brockman, and Ilya Sutskever founded the nonprofit OpenAI, Inc. (“OpenAI”) in December 2015 with the specific purpose to “provide funding for research, development and distribution of technology related to artificial intelligence.” (Declaration of Walter H. Hawes (“Hawes Decl.”), Dkt. No. 351-3, Ex. 7.) After months of discussion, the founders concluded that “someone other than Google” should develop artificial general intelligence (“AI” or “AGI”), and formed OpenAI as a nonprofit built on the core tenants that “the resulting technology will benefit the public,” that it would “seek to open source technology for the public benefit”, and that it would not be “organized for the private gain of any person.” (Id., Exs. 2, 3, 7.) Because the founders feared AI's unconstrained potential, they agreed that “safety should be a first-class requirement” and the technology “would be owned by the foundation and used for the good of the world.” (Id., Ex. 3.) Musk and Altman were named OpenAI's co-chairs. (Declaration of David J. Wiener (“Wiener Decl.”), Dkt. No. 328-4, Ex. 70.)

1. Musk's Charitable Donations to OpenAI—2016 to 2020

From the start, Musk was one of OpenAI's largest donors. (Hawes Decl., Exs. 5, 12.) He donated approximately $38 million to OpenAI, including $5 million per quarter from 2016 to 2017, and $12.7 million to pay for OpenAI's leased office space from September 2016 until September 2020. (Id., Ex. 11; Wiener Decl., Ex. 4.)

Musk initially contributed through YC.org—OpenAI's fiscal sponsor while it secured 501(c)(3) status—per OpenAI's instruction. (Hawes Decl., Exs. 11, 69.) YC.org assured Musk before he donated that it would “be able to grant funds to OpenAI via the sponsorship arrangement.” (Id., Exs. 17, 69.) Once OpenAI acquired 501(c)(3) status, Musk contributed funds through Donor Advised Funds (“DAFs”) that Vanguard and Fidelity administered. (Id., Ex. 11.) Musk could “advise” the DAF, but was informed that once he supplied the funds, they would be “owned” by the DAF and Musk would “no longer ha[ve] control over the assets.” (Wiener Decl., Exs. 17, 20, 23.) No party identifies an instance where a DAF or YC failed to donate Musk's charitable funds to OpenAI when directed. (See Hawes Decl., Ex. 14, Birchall Dep. Tr. at 172:10–174:24.)

2. For-Profit Representations—2017

In 2017, OpenAI's founders became increasingly concerned that building AGI would “ultimately require large amounts of capital” beyond what they could collect through charitable donations. (Hawes Decl., Exs. 38, 39, 40.) Brockman, Altman, Sutskever, and Musk discussed several corporate restructuring options. Brockman, Altman, and Sutskever appeared to favor establishing a for-profit affiliate, while Musk proposed that OpenAI “attach to Tesla as its cash cow.” (Wiener Decl., Ex. 59.) Musk insisted throughout those discussions that OpenAI remain a “philanthropic endeavor.” (Hawes Decl., Ex. 39.) On September 20, 2017, amid governance discussions, Musk sent the following email:

Guys, I've had enough. This is the final straw.

Either go do something on your own or continue with OpenAI as a nonprofit. I will no longer fund OpenAI until you have made a firm commitment to stay or I'm just being a fool who is essentially providing free funding for you to create a startup.

Discussions are over.

(Id., Ex. 41.) Sam Altman responded the next day: “i remain enthusiastic about the non-profit structure!” (Id., Ex. 40.) Greg Brockman too assured Musk, via email to Shivon Zilis (an OpenAI board member and Musk employee), that he would “like to continue with the non-profit structure.” (Id., Ex. 42.) Altman further informed Zilis that Altman was “[g]reat with keeping non-profit and continuing to support it.” (Id.) In November 2017, Brockman and Sutskever presented to Musk strategies to increase OpenAI's charitable funds. (Id., Ex. 43.) Brockman reported in contemporaneous notes: “[w]e said that we wanted to get more results in the non-profit and to fundraise there.” (Id.)

3. Brockman's Personal Files—2017

Discovery in this case likewise revealed Brockman's private thoughts about OpenAI's structure, which Brockman documented in his personal files (Musk refers to these entries as a “diary”). Brockman wrote in September 2017:

This is the only chance we have to get out from Elon. Is he the “glorious leader” that I would pick? We truly have a chance to make this happen. Financially, what will take me to $1B?

․

Accepting Elon's terms nukes two things: our ability to choose (though maybe we could overrule him) and the economics.

(Id., Exs. 46, 24, Brockman Dep. Tr. at 218:11–219:7.)

After Brockman and Altman reaffirmed to Musk their commitment to OpenAI's nonprofit structure, Brockman discussed OpenAI's future with Sutskever on November 6, 2017. Brockman wrote after the meeting that the “conclusion is we truly want the b-corp. honestly we also want to get back to work, but it's not super clear how we get there.” (Id., Exs. 43, 24, Brockman Dep. Tr. at 229:2–7.) He continued:

cannot say that we are committed to the non-profit. don't want to say that we're committed. if three months later we're doing b-corp then it was a lie.

(Id., Ex. 43.) Brockman conceded that he was “not feeling so great about all of this. the true answer is that we want [Musk] out.” (Id.) He wrote:

can't see us turning this into a for-profit without a very nasty fight. i'm just thinking about the office and we're in the office. and his story will correctly be that we weren't honest with him in the end about still wanting to do the for profit just without him.

(Id.)

After the November 6, 2017 meeting with Musk—and after Brockman and Altman informed Musk that they remained committed to OpenAI's nonprofit structure—Brockman concluded that “another realization from [this meeting] is that it'd be wrong to steal the non-profit from him. to convert to a b-corp without him. that'd be pretty morally bankrupt. and he's really not an idiot.” (Id.) Days later, Brockman further wrote under the heading “our plan” that “it would be nice to be making the billions” and explained that “we've been thinking that maybe we should flip to a for profit.” (Id., Ex. 45.)

4. Musk Departs OpenAI—2018

OpenAI pressed forward as a nonprofit. In January 2018, Altman worked “on a fundraising structure that does not rely on a public offering.” (Id., Ex. 44.) Brockman continued to represent to Musk through January 2018 that OpenAI would “try [its] best to remain a non-profit” because OpenAI's “biggest tool is the moral high ground.” (Id.)

On February 20, 2018, Musk resigned from the OpenAI board but agreed to “continue to donate and advise” the organization. (Id., Ex. 47.) Musk provided just under $10 million to OpenAI over the next three years to pay for OpenAI's office space. (Id., Ex. 11.)

5. OpenAI LP—2019

After Musk resigned, evidence reveals that OpenAI began building out its for-profit ventures and developing a relationship with Microsoft. (Id., Ex. 48.) The parties dispute the extent to which OpenAI's for-profit transition was incremental and whether OpenAI sufficiently informed Musk of its plans. Regardless, on August 31, 2018, Altman sent Musk a draft of a terms sheet for OpenAI LP. (Id., Ex 61.) Musk did not respond. (Id.)

On March 11, 2019, OpenAI announced that it had formed a for-profit subsidiary, OpenAI LP, which would help OpenAI “increase [its] ability to raise capital while still serving [OpenAI's] mission.” (Id., Ex. 49.) OpenAI explained to the public that under OpenAI LP, employees and investors would receive a “capped return” and that OpenAI LP likewise would “put [OpenAI's] overall mission—ensuring the creation and adoption of safe and beneficial AGI—ahead of generating returns for investors.” (Id.) The LP's profit cap was set “north of $250 billion,” which must be met before residual returns were shared with the nonprofit. (Id., Ex. 51, OpenAI 30(b)(6) Dep. Tr. at 176:22–177:9.)

After the for-profit entity was created, OpenAI transferred its assets out of the nonprofit and to the for-profit. It moved “substantially all” of its intellectual property “that was developed at the nonprofit at that point” and all employees to the for-profit entity. (Id. at 102:5–103:1; 195:12–16.)

Those close to Musk received intel on OpenAI's plans. (Wiener Decl., Exs. 60, 63, 64, 65, 66.) On March 6, 2019, Altman sent Musk a draft announcement of OpenAI LP that explained the “capped-profit” company, which was organized in a way such that “investors are clear that they should never expect a profit.” (Hawes Decl., Ex. 64.) Musk again did not respond. (Id.) On March 11, 2019, in response to press about OpenAI LP, Musk asked Altman to “[p]lease be explicit that I have no financial interest in the for-profit arm of OpenAI.” (Wiener Decl., Ex. 68.) No evidence suggests that Altman shared with Musk the specifics of OpenAI's profit cap or that it had transferred “substantially all” assets out of the nonprofit.

Musk's last payment to OpenAI was on September 14, 2020. (Id.)

6. Microsoft-OpenAI Relationship—2016 to 2025

Microsoft showed interest in OpenAI from the outset. One day after OpenAI's founding, Microsoft's Chief Executive Officer Satya Nadella shared OpenAI's introductory blog post with Microsoft executives. (Plaintiff's Responsive Separate Statement to Microsoft's Statement of Undisputed Material Facts (“PRSS”), Dkt. No. 353-2, No. 40.) In 2016, Microsoft provided discounted computing power to OpenAI in exchange for being named as a sponsor. (See Declaration of Alexandra C. Eynon (“Eynon Decl.), Dkt. No. 353-4, Ex. 10.) A Microsoft deck from August 2016 describes OpenAI as a “non-profit artificial intelligence research company” with a goal “to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return” and lists Musk as a founder and co-chair. (Id.)

In March 2018, a few weeks after Musk resigned from the OpenAI board, Altman shared plans with Microsoft to “launch a new commercial venture aimed at delivering a vertically integrated AI training hardware and related services into the market,” which Nadella circulated internally. (Id., Ex. 1.) Microsoft's Chief Technology Officer (“CTO”) Kevin Scott responded:

I wonder if the big OpenAI donors are aware of these plans? Ideologically, I can't imagine that they funded an open effort to concentrate [machine learning] talent so that they could then go build a closed, for profit thing on its back.

(Id., Ex. 1.)

In July 2019, Microsoft agreed to invest $1 billion in OpenAI LP in exchange for a convertible limited partnership interest and rights to its profits up to $20 billion. (Id., Ex. 25.) Microsoft also executed a Joint Development and Collaboration Agreement (“JDCA”) with OpenAI, Inc. (the nonprofit) and OpenAI LP (the for-profit subsidiary), which (i) provided that Microsoft would build a supercomputer for training the OpenAI entities' models and would exclusively supply cloud computing to both entities and (ii) granted Microsoft an exclusive license to commercialize one of the OpenAI entities' models for one year, during which the model could not be open source. (Id., Ex. 28; Declaration of Russell P. Cohen (“Cohen Decl.”), Dkt. No. 330-4, Ex. 2, Nadella Dep. Tr. at 109:18–110:1.) The 2019 investment agreement and JDCA also represented that both OpenAI entities had the legal authority to enter those agreements and that the agreements would not cause the entities to be in breach of any obligation to any third party. (Microsoft's Statement of Material Facts (“MSMF”), Dkt. No. 330-2, Nos. 11, 12.) OpenAI's 2016 tax exemption application to the Internal Revenue Service (“IRS”) also does not disclose any agreements between OpenAI and Musk. (Id. No. 7.)

A year later, Musk appeared to track OpenAI's growing relationship with Microsoft. In September 2020, Musk publicly responded to reporting that Microsoft had exclusively licensed OpenAI's GPT-3 model by commenting that “[t]his does seem like the opposite of open” and that “OpenAI is essentially captured by Microsoft.” (Eynon Decl., Ex. 31.) Notes from an October 2020 meeting between Microsoft executives document a discussion of not wanting “to get caught up in something” because “we are effectively owning”—presumably—OpenAI, and Nadella's comment that Microsoft needed to “think ․ through” Musk's perspective on “closed openai.” (Id., Ex. 32.)

In March 2021, Microsoft invested an additional $2 billion in OpenAI LP with rights to redeem up to $12 billion. (Id., Ex 34.) Microsoft, OpenAI, Inc., and OpenAI LP also amended the JDCA to provide that Microsoft would build additional supercomputing capacity and expand Microsoft's rights to commercialize resulting intellectual property. (Cohen Decl., Ex. 31.) The 2021 investment agreement and amended JDCA, like the earlier agreements, represented that both OpenAI entities had the power to enter those agreements and that the agreements would not cause the entities to be in breach of any obligation to any third party. (MSMF Nos. 16, 17.)

Another year later, in February 2022, Altman shared plans with Microsoft to restructure OpenAI “to address some of the challenges we've faced with our unusual structure (a nonprofit being in control of an LLC) as we become a more commercial effort.” (Eynon Decl., Ex. 36.) One such objective was to “[r]emove [the] Nonprofit from formal control to mitigate private benefit risk.” (Id., Ex. 37.)

After ChatGPT was publicly released in November 2022, Microsoft agreed in January 2023 to further invest $10 billion in OpenAI Global, LLC—a replacement vehicle to OpenAI LP in which investors could hold an economic interest in OpenAI—with rights to redeem up to $60 billion initially, increasing by 20 percent annually starting in January 2025. (Id., Ex. 41.) The JDCA was amended a second time to provide that Microsoft would build more supercomputing capacity and grant Microsoft expanded intellectual property rights. (Cohen Decl., Ex. 35.) Like the earlier agreements, the second amended JDCA represented that OpenAI's performance would not violate any obligation to any third party. (MSMF No. 20.)

In late 2024, Microsoft invested an additional $750 million in OpenAI's for-profit subsidiary. (Cohen Decl., Ex. 24, Microsoft 30(b)(6) Dep. Tr. at 237:3–15.) After OpenAI announced its plan to convert the for-profit subsidiary into a public benefit corporation (“PBC”), Microsoft and OpenAI entered a nonbinding memorandum of understanding in September 2025 stating that Microsoft may convert its investments into an equity stake in the successor PBC. (Eynon Decl., Ex. 64.)

B. PROCEDURAL BACKGROUND

In August 2024, Musk filed suit. (Dkt. No. 1.) On April 4, 2025, the Court bifurcated the claims into two litigation phases. On May 22, 2025, Musk filed the operative, second amended complaint.⁴ OpenAI then asserted counterclaims against Musk. (Dkt. No. 176.) On October 9, 2025, under the Court's order requiring further election (Dkt. No. 298), Musk filed notice of his election to pursue unjust enrichment rather than breach of implied-in-fact contract. (Dkt. No. 313.)

This Order addresses Musk's remaining phase one claims: breach of charitable trust, constructive fraud, fraud, and unjust enrichment against OpenAI defendants; and tortious interference with contract, aiding and abetting breach of fiduciary duty, and unjust enrichment against Microsoft.⁵

II. LEGAL STANDARD

The standard is not in dispute. Summary judgment is appropriate when “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). In reviewing summary judgment motions, courts must view all evidence in the light most favorable to the nonmoving party and draw all justified inferences on its behalf. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). “Where the non-moving party bears the burden of proof at trial, the moving party need only prove that there is an absence of evidence to support the non-moving party's case.” In re Oracle Corp. Sec. Litig., 627 F.3d 376, 387 (9th Cir. 2010).

III. OPENAI DEFENDANTS' MOTION FOR SUMMARY JUDGMENT

OpenAI defendants move for summary judgment on all four of Musk's remaining phase one claims.

A. BREACH OF CHARITABLE TRUST

Musk alleges that OpenAI defendants breached his charitable trust by effectively converting OpenAI into a for-profit, closed source entity. A charitable trust requires “a proper manifestation by the settlor of an intention to create a trust, a trust res, and a charitable purpose.” City of Palm Springs v. Living Desert Rsrv., 70 Cal.App.4th 613, 620, 82 Cal.Rptr.2d 859 (1999) (cleaned up). OpenAI defendants move on two grounds: Musk (1) does not have standing to enforce the trust and (2) cannot prove an element of the trust, namely that he identified a specific trust purpose for his charitable donations. The Court analyzes each in turn.

1. Standing

OpenAI defendants argue that Musk lacks standing to enforce the charitable trust for two reasons. First, OpenAI defendants argue that Musk, as the donor (or settlor),⁶ does not hold a “sufficient special interest” in the trust such that he could enforce its terms. Second, OpenAI defendants argue that Musk indirectly donated to OpenAI through donor advised funds (“DAFs”) and to OpenAI's fiscal sponsor, YC, and thus severed any interest that Musk might have retained in the trust's funds.

a) Common Law Settlor and Special Interest Standing

OpenAI defendants maintain that the settlor of a trust—like Musk—does not have standing to enforce the terms of the trust created. In raising that argument, OpenAI defendants, in effect, move this Court to reconsider its prior ruling that Musk had common law “settlor standing” to advance his charitable trust claim. The Court held:

Defendants also challenge plaintiffs' standing. As Musk has not been directly affiliated with OpenAI for several years, any standing must come from an interest in OpenAI's assets. California Corporations Code Section 5142(a). The Court is aware of the distinction between Restatement (Second) of Trusts § 391 and Restatement (Third) of Trusts § 94 and cmt. g, plus the California state authorities following the Restatement (Third). Thus, for purposes of this motion, the Court finds plaintiffs' standing sufficient as a settlor given the modern trend in that direction. The motion to dismiss on this issue is DENIED.

(Dkt. No. 121 at 15, n. 11.) No reason exists to disturb that prior finding. Nonetheless, the Court elaborates.

For over sixty years, the California Supreme Court has recognized that a “person having a sufficient special interest” in a trust may sue for breach. Holt v. Coll. of Osteopathic Physicians & Surgeons, 61 Cal.2d 750, 753, 40 Cal.Rptr. 244, 394 P.2d 932 (1964). Prior to Holt, only the California Attorney General could enforce a charitable trust. Id. Holt altered the status quo by broadening standing to include those having a special interest in a trust, like trustees, in part because of the practical challenges in exclusively permitting the Attorney General to enforce a charitable trust. Id. at 755, 40 Cal.Rptr. 244, 394 P.2d 932. The court reasoned that there was “no rule or policy against supplementing the Attorney General's power.” Id. In doing so, Holt relied on and incorporated the Second Restatement of Trusts into California law. Id. Justice Traynor explained in dicta that “donors who have directed that their contributions be used for certain charitable purposes” have an “interest” in the trust. Id. at 754, 40 Cal.Rptr. 244, 394 P.2d 932. Although Holt sought to protect charities from “harassing litigation,” its primary concern was lawsuits from the general public, not those like fiduciaries “who are both few in number and charged with the duty of managing the charity's affairs.” Id. at 755, 40 Cal.Rptr. 244, 394 P.2d 932.

After Holt—and prior to the Third Restatement—California courts found that donors had standing to enforce a charitable trust. In L.B. Research & Education Foundation v. UCLA Foundation, for example, the court permitted a donor to press claims against UCLA for failing to comply with the terms of its charitable donation. 130 Cal.App.4th 171, 29 Cal.Rptr.3d 710 (2005). There, the donor endowed a chair in cardiothoracic surgery, to “support basic science research activities,” that would transfer to UCSF,⁷ or to other California medical schools, should UCLA fail to meet those conditions. Id. at 175–76, 29 Cal.Rptr.3d 710. UCLA accepted the donation but failed to hire someone qualified to fill the chair. Id. When the donor moved to enforce the trust, UCLA claimed that only the Attorney General had standing to sue to enforce a charitable trust. Id. at 180, 29 Cal.Rptr.3d 710. The court rejected that argument “on the ground that the Attorney General's power to enforce charitable trusts does not in this type of case deprive the donor of standing to enforce the terms of the trust it created.” Id. A “trustee or other person having a sufficient special interest” may bring the action. Id.

To the extent that L.B. created any ambiguity as to whether settlors had standing, the Restatement (Third) of Trusts resolved it. It provides that “[a] suit for the enforcement of a charitable trust may be maintained only by the Attorney General or other appropriate public officer or by a co-trustee or successor trustee, by a settlor, or by another person who has a special interest in the enforcement of the trust.” Restatement (Third) of Trusts § 94(2) (2012) (emphasis supplied). Comments to the Restatement provide additional clarity. Under comment (g)(3), the Restatement notes that “the settlor of a charitable trust has a special interest in the performance of the trust's charitable purpose(s).” Moreover, a “charitable trust may reserve to the settlor ․ power to control or advise the trustee ․ Express powers of these types give the power holder a special interest in enforcing the charitable trust, and therefore standing ․” Id., comment (g)(2). Standing is justified by “society's interest in honoring reasonable expectations of settlors and the donor public and in enhancing enforcement of charitable trusts.” Id., comment g.  ⁸

OpenAI defendants attempt to escape that straightforward conclusion by arguing that California courts have not adopted the Third Restatement's expansion of settlor standing. That argument collapses upon a quick review of California case law. The California Supreme Court in Holt embraced the Restatement (Second) of Trusts, by “adopt[ing] a common law approach and mak[ing] the Restatement (Second) of Trusts a part of California trust law.” Autonomous Region of Narcotics Anonymous v. Narcotics Anonymous World Servs., Inc., 77 Cal.App.5th 950, 960, 292 Cal.Rptr.3d 851 (2022).⁹ Courts throughout California rely on the “updated” version of the same provision in the Third Restatement of Trusts. Id. (“We begin by defining the doctrine of special standing to enforce charitable trusts. Holt did not attempt a general definition but instead relied upon the Restatement Second of Trusts. The Restatement Third of Trusts updated the relevant provision in 2012.”) (cleaned up); Carne v. Worthington, 246 Cal.App.4th 548, 557, 200 Cal.Rptr.3d 920 (2016) (“California trust law is essentially derived from the Restatement Second of Trusts. Over a number of years, the Restatement Second of Trusts has been superseded by the Restatement Third of Trusts. As a result, we may look to the Restatement Third of Trusts for guidance.”); Lonely Maiden Prods., LLC v. GoldenTree Asset Mgmt., LP, 201 Cal.App.4th 368, 379, 135 Cal.Rptr.3d 69 (2011) (same); Est. of Giraldin, 55 Cal.4th 1058, 1072–73, 150 Cal.Rptr.3d 205, 290 P.3d 199 (2012) (considering Third Restatement of Trusts). OpenAI defendants have not cited any case that suggests that a provision-by-provision (or, as here, a phrase-by-phrase) adoption of the Third Restatement is required, or even appropriate.

In any event, Musk has standing through his special interest in enforcing the trust. Musk, as the settlor, has a “special interest” in the performance of the trust's charitable purpose. Restatement (Third) of Trusts § 94(2) cmt. g(3). Musk further retained the right to advise the charitable trust, and argues, with some factual support, that he attached several conditions to his contributions. (Wiener Decl., Exs. 17, 20, 23; Hawes Decl., Exs. 2, 5, 6.)

Here, like in L.B. and under the Third Restatement, Musk, as the donor and settlor of the trust, has standing to enforce the terms of his charitable trust both as a settlor and through his special interest in the enforcement of the trust. OpenAI defendants have cited no new case law or factual basis to suggest that the Court incorrectly decided that issue. The Court therefore denies OpenAI defendants' motion for summary judgment on that basis.¹⁰

b) Article III Standing in Light of Use of DAFs

OpenAI defendants next argue that Musk lacks Article III standing because he did not directly donate to OpenAI. Musk donated approximately $38 million to OpenAI through DAFs, or donor advised funds, and OpenAI's fiscal sponsor, YC.org, which in turn provided those funds to OpenAI. (SUF 9; Wiener Decl., Ex. 4; Hawes Decl., Ex. 11.) OpenAI defendants claim that by contributing funds through DAFs and a fiscal sponsor, Musk severed any legal interest in the at-issue contributions because he ceded ownership and control to those entities.

A quick primer on DAFs and fiscal sponsorships merits discussion. A DAF is “a charitable giving vehicle that allows donors to take a present-year income tax deduction, while distributing the funds to charity at a later time.” Pinkert v. Schwab Charitable Fund, 48 F.4th 1051, 1052 (9th Cir. 2022). Charitable funds must be provided to a sponsoring organization, like Fidelity or Vanguard, that manages the DAF and otherwise “own[s] and control[s]” the investments held in the fund. 26 U.S.C. § 4966(d)(2)(A)(ii). Although the sponsoring organization owns the assets, the donor “has, or reasonably expects to have, advisory privileges with respect to the distribution or investment of amounts held in such fund or account by reason of the donor's status as a donor.” 26 U.S.C. § 4966(d)(2)(A)(iii). The sponsoring organization, however, “is not legally obligated to comply with the donor's advice.” Pinkert, 48 F.4th at 1053. Once a donor provides those funds to a sponsoring organization—and accepts a tax deduction—the donor cannot recoup those funds. Here, Musk's agreements with Vanguard, Fidelity, and YC.org reflect those statutory requirements, and caution that once Musk contributes to the fund, “he no longer has control over the assets” which are “owned” by the DAF, are “unconditional and irrevocable,” and can be used solely for charitable purposes. (Wiener Decl., Exs. 17, 20, 23, 34.)

With that background in mind, the Court turns to OpenAI defendants' argument. OpenAI defendants primarily rely on a recent Ninth Circuit case, Pinkert, to argue that Musk “lacks Article III standing because—as the statutory framework for [DAFs] provides—he gave up title and control of his donation in exchange for an immediate tax deduction.” 48 F.4th at 1051. OpenAI defendants, however, extend Pinkert too far.

There, a donor named Philip Pinkert sued Charles Schwab for deducting excessive fees from Pinkert's DAF. Id. at 1052. Pinkert argued that despite Schwab's ownership and control of those assets, he “retained the right to direct how donated funds would be invested ․ and to determine which charitable organizations would ultimately receive the donations (and in what amount),” but Schwab's excessive fees “impaired” his ability exercise that right over all DAF funds. Id. at 1055–56 (internal quotation marks omitted). The Ninth Circuit set aside the issue of whether Pinkert retained “a property right, a contractual right, or something else” in advising the fund as to charitable donations and instead rejected Pinkert's theory because he “did not reserve the right to advise where the funds necessary to pay the fees would go.” Id. at 1056. It concluded that Schwab “did not deprive Pinkert of any advisory rights,” which were rights that Pinkert “does have.” Id. (emphasis in original).

Pinkert differs from this case in several material ways. First, Musk advances claims against the beneficiary for failing to comply with the trust's alleged terms, not Musk's DAF for an ancillary action, such as charging fees. Second, Musk advances his claim based on a right that he retains— and did exercise—in advising the sponsoring organization to donate to OpenAI. Third, the district court in Pinkert (which the Ninth Circuit affirmed) distinguished this precise scenario. It reasoned that L.B. did not control because L.B. “involved a donor who gave a restricted gift and sued to enforce the restriction” and “there is no restriction or conditional donation [in Pinkert].” 2021 WL 2476869 at *6.

OpenAI defendants stress that Musk was able to take an immediate tax deduction when he contributed funds to a DAF, and under the statutory scheme, no longer owns or controls those assets. Maybe so, but the statute addresses this scenario. OpenAI conceded at the hearing that Musk contributed unconditional and irrevocable funds to the DAFs consistent with the statutory requirements.

Here, the Court finds that Musk is not stripped of standing because he donated to OpenAI via an intermediary. That is particularly true where Musk advised, and the sponsoring organizations executed, that donation. (Hawes Decl., Exs. 11, 14, Birchall Dep. Tr. at 172:10–174:24.) Holding otherwise would significantly reduce the enforcement of a large swath of charitable trusts, contrary to the modern trend.

2. Trust Purpose

OpenAI defendants next argue that Musk cannot advance a claim for the breach of charitable trust because no evidence exists to support an element of the claim, namely the existence of a “trust purpose.”

Under the Third Restatement of Trusts § 28, charitable trust purposes include both “the advancement of knowledge or education” and “other purposes that are beneficial to the community.” Id. § 28(b), (f). “A trust purpose is charitable if its accomplishment is of such social interest or benefit to the community.” Id., cmt a.¹¹ “There is no fixed standard to determine what purposes are of such interest to the community, for the interests of the community vary with time and place.” Id. Comments to the Restatement provide examples of sufficient trust purposes. Charitable trusts include those created “for educational purposes” or “for the promotion of research,” id., cmt. h, or a trust “for the poor,” id., cmt. g. Moreover, a “trust need not designate a specific charitable purpose or a method of accomplishing such a purpose in order to be charitable” so long as the trust is created “for charitable purposes” and allows a trustee to select a charitable organization. Id. The Restatement does include some limitations on trust purpose: an “outright devise” to a charitable institution, or a gift to be used for “general purposes,” does not create a trust, while a “disposition to such an institution for a specific purpose, such as to support medical research” does create a charitable trust. Id., cmt. a.

OpenAI defendants argue under the Third Restatement that Musk provided OpenAI with an outright devise to be used for OpenAI's general purposes consistent with OpenAI's charter and mission. Musk responds that his gift had a specific charitable purpose and that he attached two fundamental terms to it: that OpenAI be open source and that it would remain a nonprofit. (Hawes Decl., Ex. 12, Musk Dep. Tr. at 285:19–286:24.) A genuine dispute of material fact exists here as to whether Musk attached those terms.

The argument that a settlor must attach specific, additional conditions to a trust where a charitable donation already aligns with the mission of an organization is not supported by caselaw. On the contrary, the Third Restatement permits broad charitable purposes. For example, a donation to a hospital or university “to support medical research” was sufficient to create a charitable trust under the Restatement even though “medical research” is intertwined in the fabric of a research hospital or university. The same is true for Musk's alleged condition that OpenAI, a nonprofit and open-source organization, remain that way.¹²

Accordingly, the Court DENIES OpenAI defendants' motion as to Musk's breach of charitable trust claim.

B. CONSTRUCTIVE FRAUD

To succeed on a claim for constructive fraud under California law, a plaintiff must prove “(1) a fiduciary or confidential relationship; (2) an act, omission or concealment involving a breach of that duty; (3) reliance; and (4) resulting damage.” (Joint Statement Outlining Elements of Claims (“Statement of Elements”), Dkt. No. 30 at 1 (quoting Fabian v. LeMahieu, No. 19-cv-0054-YGR, 2019 WL 4918431, at *14 (N.D. Cal. Oct. 4, 2019).)

OpenAI defendants argue summary judgment is warranted on Musk's constructive fraud claim because Musk's lack of standing forecloses his charitable trust claim. Here, Musk's constructive fraud claim is premised on OpenAI's fiduciary duty to Musk under the charitable trust. Because the charitable trust claim survives, the Court DENIES OpenAI defendants' motion for summary judgment as to its constructive fraud claim.

C. FRAUD

OpenAI defendants next move for summary judgment on Musk's fraud claim on the grounds that he has not identified an actionable misrepresentation, cannot prove reliance, and the statute of limitations bars his claim. All are required to prove fraud. Small v. Fritz Cos., Inc., 30 Cal.4th 167, 173, 132 Cal.Rptr.2d 490, 65 P.3d 1255 (2003) (The elements of fraud under California law are “(a) misrepresentation (false representation, concealment, or nondisclosure); (b) knowledge of falsity (or ‘scienter’); (c) intent to defraud, i.e., to induce reliance; (d) justifiable reliance; and (e) resulting damage.”). The Court addresses each argument.

1. Actionable Misrepresentations

With respect to the issue of an actionable misrepresentation, to “constitute fraud, the misrepresentation [ ] must be a material and knowingly false representation of fact.” Huntsman v. Corp. of the President of the Church of Jesus Christ of Latter-Day Saints, 127 F.4th 784, 790 (9th Cir. 2025). California law recognizes that a “promise to do something necessarily implies the intention to perform; hence, where a promise is made without such intention, there is an implied misrepresentation of fact that may be actionable fraud.” Lazar v. Superior Ct., 12 Cal.4th 631, 638, 49 Cal.Rptr.2d 377, 909 P.2d 981 (1996).

OpenAI defendants argue that although this litigation stems from Altman and Brockman's statements to Musk in late 2017 and early 2018 about their commitment to OpenAI's nonprofit structure, those statements were not factual misrepresentations and occurred after Musk had contributed most of the at-issue funds. Musk counters that Altman and Brockman repeatedly committed promissory fraud by reassuring him in late 2017 and early 2018 that OpenAI would remain a nonprofit while they secretly planned to turn OpenAI into a for-profit company.

As outlined above, the statements in late 2017 and early 2018 included assurances that Altman “remained enthusiastic about the nonprofit structure” and again committed to that structure in conversation with Shivon Zilis. (Hawes Decl., Exs. 40, 41, 42, 43, 44.) So did Brockman. (Id.) Brockman's electronic notes could be read to suggest that Brockman intended to deceive on that same issue: he “cannot say that [he is] committed to the non-profit” because that representation would be “a lie” and Musk's story would “correctly be that we weren't honest with him in the end about still wanting to do the for profit just without him.” (Id., Ex. 43.) Those statements concern a potential representation of fact—the existence and maintenance of the non-profit structure of OpenAI. Any concerns about when those statements were made relates to Musk's damages or the limitations period, not actionability. The Court rejects OpenAI defendants' arguments on those grounds.

2. Reliance

To prove justifiable reliance, a plaintiff must show “(1) that they actually relied on the defendant's misrepresentations, and (2) that they were reasonable in doing so.” Dey v. Robinhood Markets, Inc., 780 F.Supp.3d 882, 891 (N.D. Cal. 2025). “Except in the rare case where the undisputed facts leave no room for a reasonable difference of opinion, the question of whether a plaintiff's reliance is reasonable is a question of fact.” All. Mortg. Co. v. Rothwell, 10 Cal.4th 1226, 1239, 44 Cal.Rptr.2d 352, 900 P.2d 601 (1995).

OpenAI defendants assert that the undisputed evidence proves that Musk did not believe, or rely on, any of the alleged misrepresentations. OpenAI defendants point to Musk's deposition testimony that he “started to suspect that he was being swindled” in 2017 and thereafter stopped providing his $5 million quarterly contribution to OpenAI. (Id. at Ex. 1, Musk Dep. Tr. at 76:23–77:5.) Musk disagrees, countering that the evidence shows that Altman and Brockman reassured Musk of their commitment to OpenAI's nonprofit structure after Musk voiced skepticism. He also continued to donate to OpenAI until September 14, 2020. (Hawes Decl., Ex. 11.)

Genuine disputes of material fact exist as to whether Musk relied on OpenAI defendants' representations. Although Musk testified that he “felt he was being tricked” in 2017 and that he began to suspect that Altman and Brockman's “real goal was to create a closed-source maximum-profit entity,” that testimony does not conclusively establish that Musk knew, at that moment in time, that he could not rely on their statements. (Id. at 76:23–77:5; 85:1–25; 91:17–92:7.) Contrary evidence shows Altman and Brockman's repeated efforts to reassure Musk of their commitment to the non-profit structure. Musk agreed to continue to fund OpenAI once he “understood what the forward structure was” and that “the fundamental mission of being a nonprofit open-source company continued.” (Id. at 86:6–23.) That Musk continued to donate millions of dollars to OpenAI after the alleged misrepresentations creates material factual disputes over Musk's reasonable reliance. (Hawes Decl., Ex. 11.) The Court denies the motion on those grounds.

3. Statute of Limitations

With respect to OpenAI defendants' statute of limitations argument, fraud claims are subject to a three-year limitation period. Cal. Civ. Pro. Code. § 338(d). They do not accrue “until the discovery, by the aggrieved party, of the facts constituting the fraud or mistake.” Id. “A plaintiff is on inquiry notice of its fraud claims when he learns, or at least is put on notice, that a representation [is] false.” Hamilton Materials, Inc. v. Dow Chem. Corp., 494 F.3d 1203 (9th Cir. 2007). “Because ․ notice of fraud is for the trier of fact, the party seeking summary disposition has an extremely difficult burden to show that there exists no issue of material fact regarding notice.” SEC v. Seaboard Corp., 677 F.2d 1301, 1309–10 (9th Cir. 1982). Musk responds that his claim is not time barred because he discovered the true extent of OpenAI's fraud in 2023, well within the limitations period. Musk filed this action in August 2024, so his claim would be untimely if Musk discovered, or should have discovered, the alleged fraud before August 2021.

OpenAI defendants argue that the undisputed evidence shows that Musk discovered the alleged underlying fraud at two alternative periods: in 2017, when Musk testified that he suspected that he was being “swindled” and in 2019, when emails suggest that Musk was informed of OpenAI LP—the capped, for-profit OpenAI entity. The Court has already addressed, and rejected, OpenAI defendants' argument that no genuine disputes of material fact exist as to Musk's hunch that he was being “swindled” in 2017. See, supra, at III.C.2.

As for the 2019 period, OpenAI defendants argue that the undisputed record evidence illustrates the following: Musk suggested that OpenAI pivot into a for-profit entity by attaching to Tesla in February 2018 (Wiener Decl., Ex. 59); Altman sent Musk a term sheet for OpenAI LP in August 2018 (id., Ex. 61); Altman sent Musk a draft press release of OpenAI LP, a “capped-profit” entity in March 2019 (id., Ex. 66); and Microsoft publicly announced its $1 billion investment in OpenAI LP in July 2019. (Id., Ex. 69.)

Musk responds that factual disputes remain because OpenAI acted gradually, and by a matter of degrees, in secretly transforming into a for-profit entity, so he did not discover, and could not have discovered, the true extent of OpenAI's fraud until 2023. (Hawes Decl., Ex. 1, Musk Dep. Tr. at 183:22–184:12.) Musk submits that, even in OpenAI's efforts to confer with him, OpenAI was not transparent about its for-profit transformation. For example, in March 2019 when Altman announced OpenAI LP, Altman informed Musk that “we've created the capped-for profit company and raised the first round, led by Reid and Vinod. We did this in a way where all investors are clear that they should never expect a profit.” (Id., Ex. 64; emphasis supplied.) Musk claims that OpenAI's public disclosures were “bare bones” and hid that OpenAI LP's profit cap was “illusory,” along with the extent to which OpenAI has transferred its intellectual property and employees out of the nonprofit. (Id., Ex. 51, OpenAI 30(b)(6) Dep. Tr. at 102:12–17; 195:12–16.)

The Court concludes, given the competing narratives discussed above, that genuine disputes of material fact remain as to when Musk discovered or should have discovered the fraud. Accordingly, viewing the evidence in the light most favorable to Musk, the Court cannot find that OpenAI defendants are entitled to summary judgment as a matter of law on Musk's fraud claim.

The Court DENIES OpenAI defendants' motion for summary judgment as to Musk's fraud claim.

D. UNJUST ENRICHMENT

To support a claim for unjust enrichment “a plaintiff must allege ‘receipt of a benefit and unjust retention of the benefit at the expense of another.’ ” (Statement of Elements at 4 (quoting Prakashpalan v. Engstrom, Lipscomb & Lack, 223 Cal.App.4th 1105, 1132, 167 Cal.Rptr.3d 832 (2014).) “When a plaintiff alleges unjust enrichment, a court may construe the cause of action as a quasi-contract claim seeking restitution.” Astiana v. Hain Celestial Grp., Inc., 783 F.3d 753, 762 (9th Cir. 2015) (cleaned up). For quasi-contract claims, a court implies a contract where “a defendant has been unjustly conferred a benefit through mistake, fraud, coercion, or request.” Id.; see also Krantz v. Old Copper Co., Inc., 794 F.Supp.3d 724, 746 (C.D. Cal. 2025).

OpenAI defendants move for summary judgment on Musk's unjust enrichment claim for three reasons: Musk chose to sue in tort, so his quasi-contract claim cannot succeed, he failed to show that he expected to be compensated, and his claim is untimely. The Court addresses each.

First, OpenAI defendants argue that because Musk chose to sue to tort, he is not entitled to restitution under a quasi-contract theory based on the same alleged conduct. Musk's unjust enrichment claim, however, and as affirmed at the hearing, is not based on the same alleged conduct and sounds in restitutionary, contract-based principles, not fraud. Each depends on different underlying claims and facts. (Dkt. No. 313) (Musk elects unjust enrichment claim over implied contract claim). Musk's quasi-contract theory is based on allegations that Musk and OpenAI defendants had an implied agreement, and OpenAI defendants failed to comply with the agreement's terms. (SAC ¶¶ 255–260; 275–278.) Musk's fraud claim, on the other hand, is based on representations that OpenAI defendants made to Musk both before and after he donated significant sums to OpenAI. (Id. ¶¶ 302–312.) Because the theories are different, the Court will not discuss unjust enrichment as duplicative.

Second, OpenAI defendants argue that Musk's quasi-contract theory (tethered to quantum meruit and fraud theories) fails because Musk did not expect to be compensated.¹³ The Court disagrees with OpenAI's false dichotomy and its “tethering” framing. In any event, as to quantum meruit (or the request theory), OpenAI defendants cite Taylor v. Google, LLC. 2024 WL 837044 (9th Cir. Feb. 28, 2024) (To “recover in quantum meruit, a plaintiff must show the circumstances were such that the services were rendered under some understanding or expectation of both parties that compensation therefor was to be made.”) OpenAI's efforts to limit Musk's quasi-contract theory do not persuade. Quantum meruit is a type of recovery in quasi-contract that involves the provision of services. In re De Laurentiis Ent. Grp. Inc., 963 F.2d 1269 (9th Cir. 1992) (quantum meruit implies an equitable remedy where “a plaintiff who has rendered services benefiting the defendant may recover the reasonable value of those services when necessary to prevent unjust enrichment of the defendant.”); compare CACI No. 371 with CACI No. 375. Here, Musk's quasi-contract theory does not relate to the provision of Musk's services, so the “expectation of compensation” requirement does not apply.

Third, OpenAI defendants once again argue that Musk's unjust enrichment claim is time barred. Musk's quasi-contract based unjust enrichment claim is subject to a two-year limitations period. Rodriguez v. Int'l Bus. Machines Corp., 2024 WL 3908119 (N.D. Cal. Aug. 19, 2024) (“Where the unjust enrichment claim is based on a quasi-contract theory, it is governed by the two-year limitations period for ‘[a]n action upon a contract, obligation or liability not founded upon an instrument of writing[.]’ ”) (quoting Cal. Civ. Proc. Code § 339).¹⁴ A claim accrues upon “the occurrence of the last element essential to the cause of action.” Aryeh v. Canon Bus. Sols., Inc., 55 Cal.4th 1185, 1191, 151 Cal.Rptr.3d 827, 292 P.3d 871 (2013). OpenAI defendants argue that because Musk last donated to OpenAI in September 2020, under either period, his claim is barred.

Like with his fraud claim, Musk argues that factual disputes remain as to when he discovered the alleged misconduct. Aryeh, 55 Cal.4th at 1192, 151 Cal.Rptr.3d 827, 292 P.3d 871 (“[T]he discovery rule ․ postpones accrual of a cause of action until the plaintiff discovers, or has reason to discover, the cause of action.”) Musk maintains that he did not discover, or have reason to discover, his cause of action until 2023 when OpenAI publicly announced its $10 billion transaction with Microsoft. If so, the claim survives. Here, genuine disputes of material fact remain as to the limitations period.

The Court DENIES OpenAI defendants' motion for summary judgment as to Musk's unjust enrichment claim.

IV. MICROSOFT'S MOTION FOR SUMMARY JUDGMENT

Microsoft separately moves for summary judgment on Musk's remaining claims against it in phase one of the litigation—tortious interference with contract, aiding and abetting breach of fiduciary duty, and unjust enrichment.

A. TORTIOUS INTERFERENCE WITH CONTRACT

“The elements necessary to state a cause of action for intentional interference with contractual relations are (1) a valid contract between plaintiff and a third party; (2) defendant's knowledge of this contract; (3) defendant's intentional acts designed to induce a breach or disruption of the contractual relationship; (4) actual breach or disruption of the contractual relationship; and (5) resulting damage.” (Statement of Elements at 6 (quoting Mintz v. Blue Cross of Cal., 172 Cal.App.4th 1594, 1603, 92 Cal.Rptr.3d 422 (2009)) (internal quotation marks omitted).)

Microsoft argues in a footnote that the Court should enter summary judgment against Musk on this claim because it is derivative of the breach of implied-in-fact contract claim that Musk abandoned to seek recovery for unjust enrichment. (See Dkt. No. 313.) Musk conceded the same at the hearing. Accordingly, the Court GRANTS Microsoft's motion as to tortious interference with contract.

B. AIDING AND ABETTING BREACH OF FIDUCIARY DUTY

With respect to the aiding and abetting claim, Microsoft urges that no evidence exists that it had actual knowledge of any breach of duties owed to Musk nor that Microsoft intended to substantially assist in any alleged breach. Musk disagrees, identifying both circumstantial and direct evidence that Microsoft knew—or must have known—that Altman, Brockman, and OpenAI were breaching their duties to Musk.

Liability for aiding and abetting breach of fiduciary duty attaches when a defendant (1) “knows [that an]other's conduct constitutes a breach of duty” and (2) “gives substantial assistance or encouragement to the other to so act.” Casey v. U.S. Bank Nat'l Ass'n, 127 Cal.App.4th 1138, 1144, 26 Cal.Rptr.3d 401 (2005) (quoting Saunders v. Superior Ct., 27 Cal.App.4th 832, 838-39, 33 Cal.Rptr.2d 438 (1994)).¹⁵ Microsoft focuses solely on the knowledge element.¹⁶

As the Casey court explained, aiding and abetting requires “actual knowledge of the specific primary wrong the defendant substantially assisted,” such that the defendant had “intentional participation with knowledge of the object to be attained.” 127 Cal.App.4th at 1145–46, 26 Cal.Rptr.3d 401 (citations omitted) (emphasis in original). Although it “is not sufficient merely to have a vague suspicion of wrongdoing, in the nature of a hunch that something fishy was going on[,] ․ [a] plaintiff may prove actual knowledge through inference or circumstantial evidence.” AngioScore, Inc. v. TriReme Med., LLC, 70 F.Supp.3d 951, 957 (N.D. Cal. 2014) (citations and internal quotation marks omitted). This includes “circumstantial evidence that the [defendant] must have known” the relevant facts. See Martinez v. Bank of Am. Nat. Tr. & Sav. Ass'n, 82 Cal.App.4th 883, 891, 98 Cal.Rptr.2d 576 (2000) (discussing actual knowledge in the context of deciding whether a defendant owes a duty of care).

Here, Musk identified considerable evidence raising a triable issue of fact that Microsoft had actual knowledge beyond vague suspicion of wrongdoing. It includes: (i) an email by Microsoft's CTO in early 2018 stating: “Ideologically, I can't imagine that [big OpenAI donors] funded an open effort to concentrate [machine learning] talent so that they could then go build a closed, for profit thing on its back”;¹⁷ (ii) notes from an October 2020 meeting suggesting that Nadella expressed the need to “think ․ through” Musk's perspective on “closed openai,” following a discussion of not wanting “to get caught up in something” because “we are effectively owning”—presumably—OpenAI;  ¹⁸ and (iii) evidence that Microsoft executives knew of OpenAI's nonprofit mission,¹⁹ Musk's role as a major donor,²⁰ and that a goal of OpenAI's commercial restructuring was to “[r]emove Nonprofit from formal control to mitigate private benefit risk.”²¹ This is sufficient to raise a factual dispute that Microsoft knew or must have known that its investments in OpenAI's commercial ventures assisted or encouraged the breach that Musk alleges.

Although there may be no deposition testimony indicating that anyone at Microsoft was told about any alleged duties to Musk ²² and OpenAI's disclosures to the IRS and written agreements with Microsoft did not identify any supposed duties,²³ that evidence only contradicts Musk's articulation. It does not vitiate the facts identified above.

Viewing the evidence in the light most favorable to Musk, the Court finds that there are genuine disputes of material fact that Microsoft knew that its investment in OpenAI's for-profit ventures contributed to Altman, Brockman, and OpenAI's alleged breach of duties to OpenAI's donors, including Musk. As such, Microsoft's argument that it lacked intent because it lacked knowledge does not persuade. Microsoft's motion is DENIED as to the aiding and abetting breach of fiduciary duty claim.

C. UNJUST ENRICHMENT

Lastly, Microsoft argues that Musk's claim for unjust enrichment fails because it impermissibly seeks the same relief as his aiding and abetting claim.²⁴ Musk argues that California law permits plaintiffs to try their stand-alone unjust enrichment theory because it has different elements than aiding and abetting.

As noted above, federal courts applying California law construe a stand-alone unjust enrichment claim “as a quasi-contract claim seeking restitution.” Astiana, 783 F.3d at 762 (citation omitted). “To allege unjust enrichment as an independent cause of action, a plaintiff must show that the defendant received and unjustly retained a benefit at the plaintiff's expense.” ESG Cap. Partners, LP v. Stratos, 828 F.3d 1023, 1038 (9th Cir. 2016).²⁵ “Unjust enrichment ․ is an appropriate claim only when a defendant obtained a benefit from a plaintiff.” Roadrunner Intermodal Servs., LLC v. T.G.S. Transp., Inc., 2021 WL 2188138, at *14 (E.D. Cal. May 28, 2021) (emphasis in original). However, “[w]here a plaintiff brings a cause of action for unjust enrichment with a claim for aiding and abetting breach of fiduciary duty or fraud, courts will dismiss the claim for unjust enrichment so that a plaintiff can pursue it as an equitable remedy rather than a separate cause of action.” Stapleton v. JPMorgan Chase Bank, N.A., 779 F.Supp.3d 1059, 1076 (N.D. Cal. 2025) (dismissing unjust enrichment claim that “merely incorporates the allegations for aiding and abetting”). Under California law, “the restitutionary remedies of unjust enrichment and disgorgement are available for aiding and abetting breach of fiduciary duty.” Am. Master Lease LLC v. Idanta Partners, Ltd., 225 Cal.App.4th 1451, 1481, 171 Cal.Rptr.3d 548 (2014), as modified (May 27, 2014).

Here, Musk's unjust enrichment claim against Microsoft cannot survive summary judgment because Musk has not identified any evidence that suggests that he has a quasi-contractual relationship with Microsoft. See Roadrunner, 2021 WL 2188138, at *14 (granting summary judgment for defendant on plaintiff's unjust enrichment claim because plaintiff “did not confer anything to [defendant] at its own expense” and failed to “articulate any theory under which [defendant] could be said to have obtained any benefit directly from” plaintiff); see also Dairy, LLC v. Milk Moovement, Inc., 2022 WL 2392622, at *8 (E.D. Cal. July 1, 2022) (dismissing counterclaim for unjust enrichment because counterclaimant did not “allege a relationship between itself and [counterdefendant] ‘upon which a quasi-contract claim could be based’ ” (quoting Roadrunner, 2021 WL 2188138, at *14)). Unlike Musk's arguments against OpenAI defendants, here Musk failed to cite any evidence or even allege that Microsoft “obtained any benefit directly” from him. See Roadrunner, 2021 WL 2188138, at *14. Nor did Musk cite any evidence or allege any facts—apart from those for aiding and abetting—to support finding that Microsoft's retention of any benefit was unjust. See Dairy, LLC, 2022 WL 2392622, at *8 (conclusory allegation that counterdefendant was “unjustly enriched” did not support a stand-alone unjust enrichment counterclaim). At the hearing, Musk conceded that his unjust enrichment theory against Microsoft is based on identical evidence and requests identical relief as his aiding and abetting claim.

Because Musk may “pursue [unjust enrichment] as an equitable remedy” for aiding and abetting “rather than [as] a separate cause of action,” the Court GRANTS Microsoft's motion as to the unjust enrichment claim.²⁶ See Stapleton, 779 F.Supp.3d at 1076.

V. CONCLUSION

For the reasons stated above, the Court rules as follows:

• OpenAI defendants' motion for summary judgment on Counts IV (unjust enrichment), VI (constructive fraud), VII (fraud), and XVIII (breach of charitable trust) is DENIED.

• Microsoft's motion for summary judgment is GRANTED as to Count IV (unjust enrichment) and V (tortious interference with contract) and DENIED as to Count XIX (aiding and abetting breach of fiduciary duty).

This Order terminates Dkt. Nos. 327, 328, 329, 330, 351, 353, 356, 358, 360, 368, 373, 388.

IT IS SO ORDERED.

FOOTNOTES

1.   OpenAI entities include: OAI Corporation, LLC, OpenAI GP, L.L.C., OpenAI Global, LLC, OpenAI Holdings, LLC, OpenAI OpCo, LLC, OpenAI Startup Fund GP I, L.L.C., OpenAI Startup Fund I, L.P., OpenAI Startup Fund Management, LLC, OpenAI Startup Fund SPV GP I, L.L.C., OpenAI Startup Fund SPV GP II, L.L.C., OpenAI Startup Fund SPV GP III, L.L.C., OpenAI Startup Fund SPV GP IV, L.L.C., OpenAI Startup Fund SPV I, L.P., OpenAI Startup Fund SPV II, L.P., OpenAI Startup Fund SPV III, L.P., OpenAI Startup Fund SPV IV, L.P., OpenAI, Inc., and OpenAI LP.

2.   The Court GRANTS the parties' omnibus sealing motion. Dkt. Nos. 368, 388. Many proposed sealing requests involve information that is not relevant to the motions before the Court. Should the parties use the same documents as trial exhibits, the Court will again consider whether sealing is appropriate.

3.   The parties may disagree about the import or framing of the facts, but the existence of those facts, which are cited as exhibits in the briefing, are not in dispute.

4.   Second Amended Complaint (“SAC”), Dkt. No. 170. Plaintiff xAI Corp. asserts other claims not part of phase one.

5.   Plaintiffs' remaining claims in phase two of the litigation are: Counts VIII (violations of 15 U.S.C. § 1), IX (violations of 15 U.S.C. § 2), X (violations of Cal. Bus. & Prof. Code § 16720), XI (violations of Cal. Bus. & Prof. Code § 17043), XII (violations of 15 U.S.C. § 14), XIII (violations of Cal. Bus. & Prof. Code § 16727), XIV (violations of 15 U.S.C. § 18 ¶¶ 1 & 2), XVI (unfair competition under Cal. Bus. & Prof. Code § 17200 et seq.), and XVII (false advertising under 15 U.S.C. § 1125(a)(1)(B)) of the SAC. Phase two also includes OpenAI's counterclaims for unfair competition and tortious interference with prospective economic advantage. Dkt. No. 228.

6.   Musk is the settlor of the trust as “[t]he person who create[d] [the] trust.” Restatement (Third) of Trusts § 3(1) (2003).

7.   OpenAI defendants argue that, in L.B., the contract “provided for a reversion of the donation if the defendant did not meet the terms and conditions.” Mot. at 13. In L.B., however, the donation transferred to a different University of California medical (selected by the President of the University of California) school each time, and did not revert to the plaintiff as OpenAI defendants suggest. Id. at 176, 29 Cal.Rptr.3d 710. Musk retains a similar advisement interest here. Further, OpenAI's argument limiting Holt to require a direct revisionary interest ignores recent trends in the case law.

8.   The Court notes here that Second Restatement provides that “[a] suit can be maintained for the enforcement of a charitable trust by the Attorney General or other public officer, or by a co-trustee, or by a person who has a special interest in the enforcement of the charitable trust, but not by persons who have no special interest or by the settlor or his heirs, personal representatives or next of kin.” Restatement (Second) of Trusts § 391 (1959) (emphasis supplied.) The Third Restatement now omits the phrase beginning with “by the settlor.” Further, as explained, California courts have long allowed settlors to sue where they retained a “special interest” in the trust. The Third Restatement now clarifies that settlors, by definition, have a special interest in enforcing the trust and explicitly permits settlors to sue.

9.   Because California Probate Code explicitly adopts the common law of trusts, California's trust law “continues to look ‘to the contemporary and evolving rules of decision’ that courts ‘develop in the exercise of their power to adapt the law to new situations and to changing conditions.’ ” Id. (citing Cal. Prob. Code § 15002).

10.   OpenAI defendants further argue that Musk does not have statutory standing to enforce the trust because, in part, section 5142(a) of the California Corporations Code does not apply to it under the internal affairs doctrine as a nonprofit incorporated in Delaware. The Court disagrees. The internal affairs doctrine does not apply where “some other state has a more significant relationship ․ to the parties and the transaction.” Lidow v. Superior Ct., 206 Cal.App.4th 351, 358–359, 141 Cal.Rptr.3d 729 (2012). “Indeed, there is no reason why corporate acts involving the making of contracts, the commission of torts and the transfer of property should not be governed by the local law of different states.” Id. (cleaned up). Here, OpenAI has amassed its charitable assets in California, has its principal place of business and primary activities in California, operated in California, and has allegedly breached its duties in California. State Farm Mut. Auto. Ins. Co. v. Superior Ct., 114 Cal.App.4th 434, 448, 8 Cal.Rptr.3d 56 (2003) (“Further, a California court can apply local law to a foreign corporation that has sufficient contacts with the state, such as conducting business or having an office here.”). The Court finds that Musk has statutory standing under section 5142(a) given his special interest and arguable quasi-contractual interest.

11.   At the hearing, counsel for Musk repeatedly referred to “comment c” to the Restatement. That comment relates to indefinite beneficiaries and is not relevant.

12.   The Court need not address Musk's novel, and unsupported, argument that comments to the Restatement (which clarifies that general purpose gifts do not create charitable trusts) apply to only public donors, not founders of the nonprofit.

13.   OpenAI defendants appeared to abandon this argument in their reply brief but again raised this argument at the January 7, 2026 hearing.

14.   The limitations period for fraud does not apply because, as Musk conceded at the hearing, Musk's unjust enrichment theory is proceeding based on a quasi-contract theory.

15.   See Statement of Elements at 6.

16.   Microsoft also argues that it lacked intent to substantially assist in the breach because it lacked knowledge of any duties, but there is no separate intent requirement for aiding and abetting liability. See Casey, 127 Cal.App.4th at 1146, 26 Cal.Rptr.3d 401 (discussing intent in the context of deciding knowledge); see also In re First All. Mortg. Co., 471 F.3d 977, 1005 (9th Cir. 2006) (aiding and abetting “is only intentional in the sense that the aider and abettor intends to take the actions that aid and abet”).

17.   PRSS No. 43.

18.   PRSS No. 48 (citing Eynon Decl., Ex. 32).

19.   PRSS Nos. 39, 40; see also Eynon Decl., Ex. 10.

20.   PRSS No. 41.

21.   See PRSS No. 47 (citing Eynon Decl., Ex. 37); Eynon Decl., Ex. 36.

22.   See, e.g., MSMF Nos. 21–24.

23.   See, e.g., MSMF Nos. 7, 11, 12, 16, 17, 20.

24.   Microsoft asserts two more arguments. One, the unjust enrichment claim fails for the same reason that the aiding and abetting claim fails, that is, because the two are premised on the same underlying conduct. Two, its retention of benefits from its investments in OpenAI's commercial ventures cannot be deemed “unjust” since Microsoft lacked knowledge of any underlying breach or wrongful act. Musk opposes on the grounds that unjust enrichment requires only constructive—not actual—knowledge and that there is a genuine dispute of fact regarding what Microsoft should have known. The Court finds Microsoft's arguments here unpersuasive in light of its finding, supra, that there is a genuine dispute of material fact as to Microsoft's actual knowledge of the alleged breach. Regardless, and as discussed infra, the Court finds summary judgment for Microsoft as to this claim on other grounds.

25.   See Statement of Elements at 4 (citing Prakashpalan, 223 Cal.App.4th at 1132, 167 Cal.Rptr.3d 832).

26.   The Court declines to address Microsoft's argument that Sonner v. Premier Nutrition precludes Musk from seeking the same relief as his aiding and abetting claim under unjust enrichment because Musk has an adequate remedy at law. See 49 F.4th 1300 (9th Cir. 2022).

YVONNE GONZALEZ ROGERS, UNITED STATES DISTRICT COURT JUDGE