Anatomy of a $97.4B Hostile Bid: How Musk's OpenAI Play Exploits Nonprofit-to-For-Profit Conversion Mechanics No Other Major Tech Company Has Ever Faced
On 8 June 2026, a consortium led by Elon Musk, backed by xAI and a roster that includes Valor Equity, Baron Capital, Atreides, Vy Capital, 8VC (Joe Lonsdale), and Ari Emanuel, lodged a $97.4 billion unsolicited bid for control of OpenAI. The price tag is not the story. The story is the target: a hybrid entity in the middle of converting itself from a charitable nonprofit into a conventional capped-profit company, a process that has no real precedent in American business history, and a fiduciary structure that the bid is designed to crack open from the inside.
The mechanics matter. OpenAI is two things at once. There is OpenAI Inc., the nonprofit, on whose board Sam Altman sits. And there is OpenAI Global LLC, the for-profit subsidiary, of which Altman is CEO but on whose board he does not sit. That separation is the entire hinge of the bid. The nonprofit board owes a duty to the charitable mission, not to maximise shareholder value; the for-profit board owes conventional fiduciary duty to equity holders. Musk's lawyer, Toberoff, has been pressuring the California and Delaware attorneys general to open up the nonprofit's charitable assets to competitive bidding — and that pressure is the load-bearing wall of the whole play.
A hostile bid is normally a fight between two boards. This one isn't. It's a fight over which board has the authority to even entertain the offer, with a charitable trust's mission sitting in the middle like an unwilling third party. As one analyst put it on TBPN's 8 June broadcast: "There's no precedent for this type of transaction in American business history." That framing is not hyperbole. No S&P 500 company has ever had to unwind a charitable-control structure to monetise its own commercial subsidiary, and no bidder has ever had to argue, in effect, that the charity is being undercompensated by its own creation.
The numbers around the conversion give the bid its leverage. OpenAI pledged in its October 2024 $6.6 billion round to complete the for-profit conversion by late 2026, at a $157 billion valuation. The company is now separately raising up to $40 billion that could value the enterprise at $300 billion, with SoftBank reportedly leading $15–25 billion of that. Stargate, the infrastructure programme announced the day after the January 2025 inauguration, sits on top. Musk's $97.4 billion is not a market-clearing offer; it is, at roughly a third of the rumoured next-round valuation, a strategic price. It is calibrated to be high enough that rejecting it looks like breach of fiduciary duty by the nonprofit board, and low enough that it is cheaper than letting the conversion close on its current terms.
Musk's stated rationale, in his own words to the Wall Street Journal: "It's time for OpenAI to return to the open source, safety-focused force for good it once was. We will make sure that happens." The rhetoric is loud; the structure is quiet. The bid is not really an offer for equity in the for-profit. It is an offer that, by its terms, forces the nonprofit board to choose: convert and sell control at a discount, or hold the charitable assets in perpetual service of a mission that the nonprofit board itself has already decided to wind down. That is a coercive choice dressed as a market transaction.
The consortium composition is the other tell. xAI's involvement points toward a potential post-deal merger rather than a clean third-party acquisition; the xAI–OpenAI combination would consolidate frontier-model capacity under a single roof and resolve Musk's conflict-of-interest posture in one move. Valor, Baron, Atreides, Vy, and 8VC provide the capital depth; Emanuel provides the deal-making connective tissue. The cap table is built to survive a months-long regulatory review, not a quick flip.
The opposition is forming along predictable lines. Meta has already written to the California attorney general opposing the conversion on competition grounds. The California and Delaware AGs are the actual decision-makers here, not the FTC or DOJ — charitable-asset conversions sit with state regulators, which is why Toberoff's pressure campaign has been aimed at Sacramento and Wilmington rather than Washington. The structural irony is sharp: the bid is strongest in the venue where regulators are least equipped to evaluate frontier-AI market concentration, and weakest in the venue where that expertise actually lives.
The conversion itself is the chokepoint. If the nonprofit board refuses to convert, the for-profit subsidiary cannot be recapitalised at a $300 billion valuation, the SoftBank-led round cannot close, and Stargate commitments become harder to fund. If the board converts, the charitable assets — built with tax-exempt donations and intended to benefit the public — transfer into a structure whose equity holders include venture investors expecting a return. Either decision is a loss the nonprofit can be blamed for. The $97.4 billion offer is the price of making one of those losses look, in court, like the lesser harm.
There is a counter-read worth taking seriously. The Musk bid could be straight poison-pill defence for incumbents: forcing every potential OpenAI investor, including SoftBank, to disclose their hand and price into the possibility that a competing bid materialises. From that angle, the $97.4 billion is not meant to close. It is meant to make the existing $300 billion path more expensive to walk down by inserting the threat of a higher, more disruptive alternative. If so, the victims are not the OpenAI boards but the next-round investors being asked to underwrite a valuation that may not survive a competitive process.
The longer frame is corporate-governance plumbing. Major US tech companies are almost universally structured as Delaware C-corps with conventional fiduciary duties. OpenAI is not. That anomaly is what Musk is purchasing the option on: a nonprofit parent whose duty runs to a mission, not to investors, sitting on top of a commercial subsidiary whose investors expect exits. The board structure — Altman on the nonprofit board, off the for-profit board — was designed to manage exactly this conflict. It is now the conflict itself. The unanswered question is whether any conversion price, Musk's or otherwise, can satisfy a Delaware chancery court that the charitable assets have been fairly valued when the only comparable transactions are bids submitted by interested parties during a contested process. The answer to that question will outlast whoever ends up owning the equity.
Wire provenance
This editorial synthesis draws on the following public wire/social posts:
- https://www.youtube.com/watch?v=DkpHkUAzbkk