Anthropic’s IPO Governance Could Leave Public Investors on the Sidelines

Updated October 28, 2026 1:24 am

Anthropic’s reported IPO-era governance structure would give public shareholders little control over the company’s board and major decisions, even if the AI lab lists on public markets. TechCrunch says the setup is designed to preserve the firm’s safety-first mission, but it would also concentrate power in founders, a trust tied to the company’s public-benefit status, and employee mechanisms that can override some board moves.

What TechCrunch says Anthropic is building

According to TechCrunch, Anthropic is preparing a structure that is unusual even by Silicon Valley standards. The company, founded in 2021, already operates as a public benefit company, meaning its mission is meant to balance AI safety with profit. But the governance plan described by the outlet suggests that going public would not automatically turn Anthropic into a company controlled by ordinary stockholders.

Instead, the structure would preserve powerful blocks of control above the level typically associated with public companies. TechCrunch reports that the Long Term Benefit Trust, or LTBT, would control a majority of board seats through Class T stock. The trust would also be able to elect or remove a majority of the board’s Class T directors.

Who sits in the power structure?

TechCrunch says Anthropic’s board chair is Daniela Amodei, and CEO Dario Amodei also sits on the board. Other board members listed by the outlet include Reed Hastings, Christopher Liddell, Vas Narasimhan and Yasmin Razavi, with one vacancy remaining.

The LTBT is chaired by Neil Shah and includes Ben Bernanke, Richard Fontaine and Christopher Stubbs, according to TechCrunch, with one seat still vacant. That trust layer matters because it sits between shareholders and the board, shaping who can steer the company even after an IPO.

How the founder voting system would work

TechCrunch reports that seven founders hold special stock that would give them 50.1% voting power if they remain aligned as a majority. Through that special class, the founders could elect three board members.

But the arrangement is conditional. If the founders fail to agree as a majority, TechCrunch says their shares would convert to proportional voting alongside common stockholders. That means the founders’ control is designed to be strong, but not permanent or automatic in every scenario.

What can happen if there is a dispute?

One of the more unusual parts of the setup is the role of the board nominating committee. TechCrunch says the committee can block some LTBT board-seat moves and send the dispute to an employee vote. In practical terms, employees could effectively overrule the trust on certain board-seat decisions if they side with the board.

That gives Anthropic a layered governance model: founders at one level, the trust at another, and employees as a possible check in the middle. Few public companies build such a structure around board power, especially one tied to a frontier AI lab.

Why it matters

For investors, the issue is not just abstract corporate law. If TechCrunch’s description is accurate, public shareholders could buy into Anthropic without gaining the kind of influence they might expect at a traditional listed company. The outlet’s own framing is blunt: public shareholders would have “common shareholders with little say over most major decisions.”

That matters because markets usually price not only growth prospects, but also voting rights, governance risk and the ability to challenge management. In Anthropic’s case, the company would be asking investors to back a fast-growing AI business while accepting that the most important levers of control are likely to stay elsewhere.

For the AI sector, the structure would also become another test case in how far frontier labs can go in insulating mission-driven governance from market pressure. Anthropic’s stated emphasis on safety makes that easier to defend in principle. Yet the more control is concentrated, the more questions emerge about accountability if the business scales rapidly after an IPO.

Why AI companies keep reaching for special control structures

Background in the tech sector helps explain the move. Public benefit corporations are allowed to pursue a public mission alongside profit, and dual-class or special-voting-share structures are often used when founders want to keep strategic control after listing. In AI, those arrangements can be especially appealing because model development, safety decisions and capital spending are long-term bets that can clash with short-term market expectations.

Anthropic also enters public scrutiny with a backdrop of governance turbulence elsewhere in AI. OpenAI’s board upheaval in 2023 made corporate control a central issue across the sector, and TechCrunch says that history has sharpened attention on how AI labs organize power before they ever reach public markets.

Background: how Anthropic got here

Anthropic was founded in 2021 and has positioned itself as an AI frontier lab with safety at the center of its identity. That mission is reflected in its public-benefit corporate form, which is intended to weigh societal goals alongside financial returns.

But mission language alone does not settle governance questions. The structure described by TechCrunch suggests Anthropic is trying to do something more specific: lock in a control system that keeps the company’s long-term direction insulated from ordinary shareholder swings, even after it begins trading publicly.

This is where the LTBT becomes central. In the model described by TechCrunch, the trust’s ability to control a majority of board seats through Class T stock would give it a direct path to influence strategy, executive oversight and potentially board composition. The founders’ special voting class adds another layer, letting them elect three board members if they remain aligned.

That combination is not accidental. It appears designed to preserve continuity in a company where product decisions and safety decisions can be tightly linked. But continuity can come at a cost: reduced accountability to investors, and a governance setup that may be harder for outsiders to interpret or challenge.

By the numbers

How does this compare with a normal IPO?

In a standard public listing, shareholders typically elect directors, vote on major corporate changes and exert influence through annual meetings and proxy campaigns. Anthropic’s structure, as described by TechCrunch, would sharply narrow that power.

The trust would control a majority of board seats through Class T stock. The founders could control three board members through a separate special-voting class, but only while they remain aligned. And employees could play a role in some disputed board-seat decisions if the nominating committee sends an issue to a vote.

That makes Anthropic’s governance less like a conventional public company and more like a layered constitutional system with different branches of authority. The advantage is stability. The downside is that public investors may find themselves financing a company without meaningful leverage over its direction.

What the LTBT is designed to do

The Long Term Benefit Trust appears to be Anthropic’s main institutional safeguard. TechCrunch reports that it continues until either 85% of common shareholders decide to end it or the trust steps down with board approval.

That is a high bar. In effect, it means the trust is not a temporary placeholder that disappears after the IPO. It is intended to remain active unless a very large shareholder coalition wants it gone, or the board agrees to unwind it.

For supporters, that could be proof that Anthropic is serious about preserving its mission. For skeptics, it is evidence that a public market listing would still leave real authority concentrated in a relatively small circle of insiders and trust appointees.

What happens next

The biggest open question is whether Anthropic will actually go public on the structure TechCrunch describes. The company has not, based on the fact sheet, publicly confirmed an IPO timeline here, and the governance model may still evolve before any listing.

Even if the broad outline holds, several unresolved points remain important:

• Who will fill the board vacancy and the LTBT vacancy.
• Whether the seven founders stay aligned enough to preserve their special voting power.
• How often the board nominating committee would escalate disputes to employees.
• Whether the trust remains acceptable to investors once Anthropic reaches the market.

Another question is time. TechCrunch says founder control can expire 10 years after an IPO, or earlier if fewer than three founders remain. That creates a long runway, but not a permanent one. Investors would therefore need to think not just about the company they are buying on day one, but about how power could shift over a decade.

For now, the story is less about a completed offering than about the architecture behind it. If Anthropic does list, its governance could become one of the clearest examples yet of an AI company trying to raise public money while limiting the power that usually comes with it.

common shareholders with little say over most major decisions

What to watch from here

Three signals will tell the story going forward. First, whether Anthropic keeps the LTBT structure intact in any final IPO filing. Second, whether investors accept the reduced control in exchange for exposure to one of the most closely watched AI companies. Third, whether Anthropic’s safety-first identity proves strong enough to justify a governance model that is deliberately resistant to ordinary market control.

If those pieces stay in place, Anthropic would be making a broader statement about the future of AI capitalism: that a company can tap public markets while insisting that strategic authority stays with founders, trustees and employees rather than dispersed shareholders.

That may appeal to some investors and policy watchers. But it also ensures that any future debate over Anthropic will be about more than valuation. It will be about who gets to steer a frontier AI company once the public owns part of it.