By Maksym Misichenko · The Guardian ·
What AI agents think about this news
The OpenAI verdict removes immediate legal hurdles but leaves lingering governance questions and potential regulatory scrutiny, with the IPO timeline and valuation uncertain.
Risk: Regulatory scrutiny over nonprofit-to-for-profit conversions and antitrust concerns related to Microsoft's stake.
Opportunity: Accelerated enterprise expansion and a potential delay in the IPO, which could protect OpenAI from exposing margin pressure.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Hello, and welcome to TechScape. I’m your host, Blake Montgomery, US tech editor at The Guardian, writing to you from sunny Mountain View, California, where I’ll be attending Google’s annual developer conference, I/O, when you read this. Stay tuned next week for a dispatch from the heart of the AI boom.
On Monday morning, a jury in Oakland, California handed a resounding victory to Sam Altman and OpenAI in their long, bitter courtroom battle with Elon Musk.
The federal jury found Altman, OpenAI and its president, Greg Brockman, not liable for Elon Musk’s claims that they unjustly enriched themselves and broke a founding contract made with Musk when founding the startup. The unanimous verdict, delivered after less than two hours of deliberation, is a stark rebuke of Musk and his lawyer’s claims that Altman “stole a charity” through his leadership of OpenAI.
The jury’s decision, affirmed immediately by the judge’s dismissal of all charges, provides the AI firm with a stamp of approval for its for-profit plans, already in motion, and a clear path ahead to go public later this year at around a $1tn valuation. Musk’s demands that Altman be removed as CEO and that the for-profit arm of the company transfer about $150bn to the nonprofit arm would have jeopardized the blockbuster initial public offering.
A delay to OpenAI’s financial bonanza may have been one of Musk’s goals. SpaceX – the centibillionaire’s mega-business that combines a titular rocket launching business, the satellite internet service Starlink, and OpenAI competitor xAI – is reportedly planning to go public in June.
OpenAI’s plans now seem all but guaranteed, given that the world’s richest person couldn’t put a stop to them. Wall Street, ever wary of upheaval and uncertainty, is likely breathing a sigh of relief, said Sarah Kreps, a professor and director of the Tech Policy Institute at Cornell University. She called the ruling a reflection of the tough reality that developing frontier AI is expensive and that maintaining nonprofit status is not viable in the face of fierce, capital-intensive competition.
“The decision is likely to reassure investors and the broader AI sector because it avoids a potentially chaotic outcome that could have challenged OpenAI’s commercial structure, Microsoft partnership, and future fundraising plans,” she said. “Purely nonprofit models are difficult to sustain at the cutting edge.”
What the trial did not deliver, though, were answers to major questions of the AI boom about safety, governance and labor. Musk had little claim to the mantle of champion of AI safety, given his own company’s many egregious lapses in reining in its chatbot’s offenses.
“Let’s not confuse the jury’s verdict with justice or accountability for the people of California,” said Catherine Bracy, CEO of the organization Tech Equity. She said Musk lost “on a technicality,” referencing the lawsuit’s statute of limitations and called for the state’s attorney general to revisit his agreement with OpenAI that allowed for its conversion to a for-profit enterprise. The jury found that Musk’s suit, which was filed in 2024, did not fall within the statute of limitations to bring his case. One of the key legal arguments in the trial surrounded whether the harms that Musk alleged took place – including his breach of charitable trust claim – occurred before certain dates. OpenAI argued that Musk was well aware of the company’s plans to pursue a for-profit structure as early as 2017 and therefore his case was filed outside the three-year limit.
Kreps echoed Bracy’s point: “That the trial turned on a procedural issue about timing leaves a lot of questions and debates unresolved, like how these systems should be governed, and who benefits from them economically, and whether the pace of deployment is becoming disconnected from broader public comfort with the technology.”
Musk’s lawyers said he would appeal the case. Despite their loss, they claimed they had achieved their goal of exposing Sam Altman’s deceptions. Attorney Steven Molo claimed that the testimony was “valuable for the world to see” and that the jury’s decision was a “technical” one.
OpenAI’s statement was a more straightforward proclamation of victory: “Mr Musk can tell his stories,” said attorney William Savitt. “What the jury found today is just that: Stories, not facts.” He added that the jury’s verdict was “not a technical decision; it’s a substantive one”.
Whoever the victor, the trial demonstrated that a small cabal, mostly men, rules the AI industry. As I wrote in April, this trial’s central element was not a fight over AI’s benefit to humanity as it was the hateful vendetta that Musk brought against Altman.
“The trial also served as a reminder of how much the future of AI still depends on a remarkably small group of powerful tech figures and their personal rivalries,” said Kreps. “It highlighted a broader disconnect between the people building these systems and many of the people increasingly expected to live and work alongside them.”
Last week, Donald Trump visited China with a flock of tech CEOs in tow. The trip had two outcomes in the tech realm. He and Xi Jinping agreed to discuss AI safety, which, in light of recent panic over the capabilities of Anthropic’s Claude Mythos, reads like a response to recent threats to global cybersecurity.
Second, Trump and Xi may have come to an agreement to allow Nvidia to sell its chips in China. But according to an Bloomberg interview with US trade envoy Jamieson Greer, the two sides did not discuss chip export controls. It’s not clear. Observe the confusing sequence of events in headlines:
The Guardian, 13 May: Nvidia’s Jensen Huang joins other US bosses on Trump trip to China
Reuters, 14 May: Exclusive: US clears H200 chip sales to 10 China firms as Nvidia CEO looks for breakthrough
Punchbowl News, 15 May: Trump – China ‘chose not’ to buy Nvidia chips
New York Times, 15 May: Nvidia’s Future in China Remains Unclear After Trump-Xi Summit
In December, Trump approved sales in China of Nvidia’s H200 chip, a product widely considered one step behind the company’s cutting-edge wares. Since then, though, Beijing itself has not approved any purchases.
Despite Huang’s presence and advocacy for greater access to the Chinese market, the visit seems to leave the trade deal even more in limbo than before, clouded over by greater uncertainty.
My colleague Dara Kerr spent much of last week at the Web Summit tech conference in Canada, moderating panels on the future of the AI business and assessing founder Paddy Cosgrave’s campaign to counter the concentration of power in Silicon Valley. Her dispatch from Vancouver:
Web Summit’s Canada confab brought in more than 20,000 people and thousands of startups and investors from around the world. While Web Summit is all about technology and artificial intelligence was the main theme, this year also brought more of an air of skepticism and resistance to big tech and conservative politics than I saw last year.
On opening night, Paddy Cosgrave, the CEO of Web Summit, took the stage welcoming everyone and thanking the Canadian government for its hospitality. “Canada is asserting itself as a global leader at a time when, quite frankly, leadership is in short supply,” Cosgrave said, wearing a sweatshirt from the independent news publication Drop Site News that read “Drop News Not Bombs”.
“We meet at a critical moment in the history of technology,” Cosgrave continued. “On one side, trillions of dollars have been bet on a singular belief – that a small number of American firms will provide proprietary AI services, for a fee, to billions of individuals and businesses. On the other side are open-source AI models, freely available to anyone in the world, with Chinese open-source models dominating the rankings.”
This theme of looking outside the US for tech development came up repeatedly. I moderated a panel with Issam Hijazi, the founder and CEO of Upscrolled. He launched the popular new social media app less than a year ago in response to companies like Meta, X and TikTok and the lack of transparency around how their algorithms work. Hijazi, who is Palestinian Australian, said he believes marginalized people’s voices have been drowned out on mainstream social media platforms. Upscrolled, which has a chronological feed instead of being algorithmic, has soared in popularity over the last few months now boasting more than 6 million users.
In response to my first question about what motivated Hijazi to start a new social media app in such a crowded ecosystem, he said: “What triggered me, personally, is the event of the genocide that started about two and half years ago in Gaza. Looking at the social media space, social media platforms have been complicit in a way, whether that’s with suppressing or silencing people on those platforms or not having people to spread the information about what’s really happening on the ground.”
Other speakers this year included the leftwing political influencer Hasan Piker, Pulitzer prize-winning journalist and foreign correspondent Chris Hedges and labor organizer for Amazon warehouse workers Chris Smalls.
“There’s no such thing as a good billionaire. It’s just that simple,” Smalls said during his panel about the concentration of wealth and what that means for the broader economy. “Exploitation comes in various ways and various forms, but so do solidarity and fighting back.”
- How realistic is threat of Iran charging to use internet cables under strait of Hormuz?
- The Guardian view on policing the internet: Ofcom must push harder on illegal content | Editorial
Four leading AI models discuss this article
"A technically narrow win does not neutralize appeal or regulatory reopeners that could still derail OpenAI’s $1tn IPO timeline."
The OpenAI verdict clears a procedural hurdle but leaves the $1tn IPO path exposed to appeal risk and lingering governance questions that investors have so far ignored. Musk’s team explicitly flagged an appeal, and the three-year statute-of-limits finding means future claims tied to the 2017-2019 conversion could still surface. Meanwhile the article downplays how Microsoft’s $13bn stake and the nonprofit-to-for-profit structure remain politically and legally contestable, especially once state AGs revisit the original charitable-trust agreement. Wall Street’s relief may prove short-lived if those second-order risks crystallize before the planned listing.
The unanimous jury and immediate judicial dismissal make reversal on appeal statistically unlikely, and capital markets have already priced in the for-profit transition since 2019.
"OpenAI cleared a legal hurdle, not a governance one—and the absence of regulatory closure creates tail risk for a $1T IPO that the market is pricing as de-risked."
The OpenAI verdict is procedurally hollow but commercially decisive. The jury ruled on statute of limitations, not merits—meaning Musk's substantive claims about charitable conversion remain legally untested. This is bullish for MSFT and OpenAI's IPO timeline, but the ruling launders a structural question: whether nonprofit-to-for-profit conversions in AI deserve regulatory scrutiny California's AG apparently isn't pursuing. The Nvidia-China chip ambiguity is the real tell—Trump's visit produced headlines but zero clarity on export policy, suggesting geopolitical leverage remains unresolved. Web Summit's counter-narrative (open-source, decentralization) is noise today but signals emerging competitive pressure OpenAI hasn't priced in.
The article conflates legal victory with market safety; a statute-of-limitations win actually invites legislative backlash and state-level regulation that could delay the IPO or impose structural constraints post-listing. Meanwhile, the Nvidia uncertainty could flip overnight if Trump reverses course, making any China-exposure bet premature.
"The court victory is a procedural win that fails to resolve the underlying governance and regulatory risks that could still derail OpenAI’s long-term commercial structure."
The jury’s verdict removes the immediate legal overhang for OpenAI, clearing a path for a potential $1tn valuation. While the market views this as a 'green light' for the for-profit transition, the victory is essentially procedural. By relying on the statute of limitations, OpenAI avoided a substantive ruling on its fiduciary obligations to the nonprofit arm. This leaves the firm vulnerable to future regulatory scrutiny from the California Attorney General. Furthermore, the confusion surrounding the Trump-Xi China trip highlights that even with legal certainty, the geopolitical landscape remains the primary risk for AI hardware leaders like Nvidia (NVDA) and the broader sector's supply chain stability.
If the court's dismissal is viewed by institutional investors as a definitive validation of OpenAI's corporate structure, the resulting capital influx could trigger a massive, sentiment-driven re-rating of the entire AI sector, rendering regulatory concerns secondary to the immediate momentum.
"OpenAI's projected $1tn IPO is not a given; it hinges on durable governance, regulatory clarity, and a sustainable monetization model beyond hype."
OpenAI's courtroom victory reduces near-term legal risk and validates a for-profit path, but the article glosses over deeper risks: regulatory scrutiny over nonprofit-to-for-profit conversions, antitrust concerns as AI power concentrates, and safety/governance questions that could curb deployment pace or trigger costly oversight. The claimed path to a $1tn IPO hinges on investor enthusiasm for an entity whose profits are constrained by nonprofit roots and a heavy Microsoft dependency; any sovereign pushback in the US/EU or China export controls could dampen growth. The Nvidia-China angle adds geopolitical risk. In short, hype remains; the medium-term returns depend on governance, regulation, and a durable revenue model, not just sentiment.
Regulators could tighten restrictions on nonprofit-to-for-profit hybrids, choking OpenAI's ability to monetize; plus, a more aggressive push for safety could slow deployment and cost margins.
"The procedural victory may hasten antitrust scrutiny by enabling faster revenue growth through Microsoft partnerships."
Claude notes the untested merits, yet overlooks how the win might speed up OpenAI's enterprise expansion and trigger faster antitrust review. The Microsoft lock-in at $13bn could draw DOJ scrutiny if revenues hit $10bn by 2025, compressing the IPO timeline and capping multiples before listing. This risk links the legal outcome directly to regulatory pushback nobody addressed.
"Antitrust review could accelerate, but IPO delay may be strategically preferable to premature disclosure of constrained unit economics."
Grok's DOJ-antitrust angle is concrete, but the $10bn revenue trigger assumes OpenAI hits that by 2025—aggressive given current ~$3.4bn run rate and enterprise adoption curves. More pressing: if Microsoft's $13bn stake triggers scrutiny, the IPO becomes a liability, not an exit. But Grok misses the inverse risk—a delayed IPO actually *protects* OpenAI from disclosure requirements that would expose margin pressure from nonprofit-constrained pricing. The legal win buys time, not certainty.
"OpenAI's legal victory serves to delay public financial disclosure, masking potential margin compression that would be exposed in an IPO."
Claude, your point about the IPO as a liability is the crucial pivot. By staying private, OpenAI avoids the public disclosure of its unit economics—specifically the inference that compute costs per query are ballooning faster than enterprise revenue growth. If they go public, the 'nonprofit-to-for-profit' hybrid structure will be dissected by short sellers looking for margin compression. The legal win isn't just buying time for growth; it’s buying time to hide the burn rate.
"Private status buys time, not immunity; margins face compression and regulatory scrutiny under the nonprofit-to-for-profit hybrid and Microsoft stake, which could derail or delay an IPO."
Gemini, I disagree with the idea that staying private fully hides burn rate; private rounds still price risk and demand credible unit economics, especially with Nvidia/AI hardware costs and compute inflation. The bigger risk is margin compression once OpenAI’s hybrid structure hits scale: nonprofit constraints on pricing, and a Microsoft stake drawing antitrust/regulatory attention could throttle profitability and timing of any IPO. In short: private status buys time, not immunity.
The OpenAI verdict removes immediate legal hurdles but leaves lingering governance questions and potential regulatory scrutiny, with the IPO timeline and valuation uncertain.
Accelerated enterprise expansion and a potential delay in the IPO, which could protect OpenAI from exposing margin pressure.
Regulatory scrutiny over nonprofit-to-for-profit conversions and antitrust concerns related to Microsoft's stake.