Anthropic founders move to lock in voting control before IPO

Updated September 25, 2026 7:24 pm

Anthropic is asking shareholders to approve a governance overhaul that would give CEO Dario Amodei and the company’s six other co-founders majority voting control as the AI company prepares for a public listing. The move matters because it would let the founders steer Anthropic after the IPO even if their economic ownership remains relatively small.

According to reporting by The Information, the proposal would hand the seven founders special shares worth 50.1% of the vote on most corporate matters, provided at least three of them continue to hold a minimum stake. The structure would not give them more cash flow rights, but it would preserve decision-making power at a moment when Anthropic is expected to enter public markets at a much larger valuation than it had earlier this year.

Anthropic’s bid is the latest reminder that in the AI race, control over a company can matter almost as much as capital. As frontier AI firms scale at breakneck speed, founders and investors are increasingly debating whether long-term mission control should outweigh the typical one-share, one-vote model used by most public companies.

What exactly is Anthropic asking shareholders to approve?

Anthropic is seeking approval for a share structure that would concentrate voting power in the hands of its co-founders. The arrangement would give Amodei and the other six original founders a combined majority of votes on most corporate issues, while still leaving them with only a modest ownership percentage.

The company has reportedly told shareholders to expect a vote in the coming days. If approved, the founders would collectively control 50.1% of the vote, a narrow but decisive majority that would allow them to influence board composition, strategy, and major corporate decisions.

Crucially, the shares would not increase the founders’ economic stake. They would not be getting a bigger claim on profits simply because they receive special voting rights. Instead, the proposal is about governance and durability of control after the company goes public.

Why the founders want this now

The founders’ logic is straightforward: a public listing can dilute influence, especially for a company that may attract large institutional investors and face pressure from shareholders focused on near-term returns. For an AI company with a stated mission around safety and long-term benefit, retaining control can be seen as a way to protect strategic independence.

Anthropic has also grown into a company whose valuation and market position could expose it to more scrutiny once public. With the firm now valued far above where it stood even months ago, its leaders appear to be trying to set the rules before investors arrive with very different priorities.

How would the voting control system work?

The structure is unusual because it is built around a group of founders rather than a single dominant entrepreneur. That makes it different from the classic founder-control model seen at companies like Meta or Snap, where one person retained extra voting rights to preserve influence.

Under the plan described in The Information’s reporting, the founders’ special voting power would remain in effect only if at least three of the seven still hold a minimum ownership stake. That condition is meant to ensure the control mechanism stays connected to meaningful involvement in the company rather than becoming a permanent lock regardless of participation.

Other governance changes would also shape Anthropic’s future. Its Long-Term Benefit Trust would still have responsibility for selecting most board members, the founders would increase their board representation from two seats to three, and employees would receive a separate stock mechanism that could help break ties on certain matters.

Why does Anthropic’s structure stand out in the AI industry?

Anthropic’s proposal is notable not because founder control is new, but because of how many people it would empower and how closely the plan is tied to a company’s mission-based identity. Super-voting structures have long been common in tech, but they are usually designed around a single charismatic founder or a small number of insiders with outsized ownership.

Meta founder Mark Zuckerberg built a system that has preserved his control for years. Snap co-founder Evan Spiegel similarly retained significant authority over the company he created. Anthropic’s proposal follows the same basic logic, but it spreads control across a founding team instead of concentrating it in one person.

That distinction matters because it reflects Anthropic’s own origin story. The company has promoted itself as an AI lab with a strong emphasis on safety, responsibility, and long-term thinking. A governance design that protects the founding group could be presented as a way to keep those principles intact even if the stock price and shareholder base change dramatically after the IPO.

What makes the group approach unusual?

Most control-preserving systems reward a single founder or CEO. Anthropic’s setup would instead create a collective power center, with several co-founders sharing the same voting rights and needing to remain sufficiently committed to maintain them.

That could reduce the risk of one person becoming the sole decision-maker, but it also raises questions about how a multi-founder bloc would operate in practice. If the founders disagree, the structure could create internal tensions. If they remain aligned, however, the arrangement could produce a durable and highly influential governing coalition.

Amodei has previously argued that extreme concentration of AI wealth could become socially destabilizing, and he has said he plans to give away the bulk of his fortune.

That public stance helps explain why the company may want control without extra economic rights. The founders would keep authority over direction and governance, but they would not be increasing the wealth they stand to extract from the firm.

How big is Anthropic becoming ahead of its IPO?

Anthropic is no longer a niche startup. The company was valued at $965 billion in May, and more recently it was reportedly marked at $1.5 trillion on the secondary market, a staggering jump that reflects the market’s appetite for leading AI platforms.

Those numbers make the coming IPO especially consequential. A public listing at a valuation anywhere near the current secondary-market level would instantly place Anthropic among the most closely watched technology companies in the world. That level of attention can intensify pressure on management to prioritize growth, margins, and stock performance.

From the founders’ point of view, that is precisely why governance matters now. Once public, Anthropic could face activist investors, quarterly earnings pressure, and repeated demands to explain its choices on model safety, capital spending, partnerships, and product rollout.

Key milestones in Anthropic’s path to public markets

1. January: Amodei publicly pledged to give away most of his wealth and warned about AI-driven inequality.
2. May: Anthropic was valued at $965 billion.
3. Recent secondary trading: The company’s valuation reportedly climbed to $1.5 trillion.
4. Coming days: Shareholders are expected to be asked to approve the voting-control plan.
5. IPO period ahead: The governance structure would shape the company’s public-market debut.

Why are investors and observers watching this so closely?

The biggest reason is that Anthropic sits at the center of one of the most valuable and strategically important sectors in global technology. Whoever controls the company will help determine how one of the world’s leading AI labs develops and deploys increasingly powerful models.

For investors, the question is whether founder-led control protects a company’s mission or limits shareholder influence. For the broader AI ecosystem, the question is whether mission-driven governance can coexist with public-market accountability.

Some market participants may welcome a stable leadership structure if they believe it reduces the odds of short-term decision-making. Others may worry that concentrated control weakens oversight, especially when the company is operating in a field with major social, economic, and regulatory implications.

Anthropic’s hybrid model — founder control supported by a trust, with some employee balancing rights — suggests it is trying to thread a narrow line between independence and accountability. The coming shareholder decision will show whether investors are willing to accept that balance.

What does this mean for the future of AI governance?

Anthropic’s proposal is part of a broader debate about who should control frontier AI companies as they become more powerful and more valuable. These firms often argue that their mission requires insulation from purely financial pressure. Critics counter that the larger and more influential the company becomes, the more it should be accountable to public shareholders and the wider public.

If Anthropic succeeds in locking in founder voting control, it could give other AI companies a template to follow. That would be especially relevant for startups that want to reassure customers, regulators, and employees that they will not be easily steered away from safety commitments once they access public capital.

At the same time, the proposal may invite scrutiny from governance experts who see it as another example of Silicon Valley’s preference for power structures that favor insiders. The fact that the founders each reportedly own only around 2% underscores that the mechanism is about influence, not equity.

Potential benefits and risks

• Benefit: Helps preserve Anthropic’s long-term research and safety priorities.
• Benefit: Shields the company from short-term market pressure after the IPO.
• Risk: Reduces ordinary shareholder power over major decisions.
• Risk: Creates a governance model dependent on continued alignment among seven people.
• Risk: Could become controversial if the company’s public performance diverges from its mission messaging.

How does Anthropic compare with other founder-controlled companies?

Anthropic is following a familiar Silicon Valley pattern, but with its own twist. Founder control has long been a hallmark of companies that want to preserve a particular culture or product vision as they grow. The common denominator is the belief that the creators of a platform should not lose authority too quickly after the company goes public.

Meta is the best-known example, with Zuckerberg’s voting power allowing him to guide strategy through years of enormous expansion. Snap has also maintained a structure that gives Evan Spiegel significant control. Anthropic’s move fits into that lineage, but the presence of multiple co-founders and an additional benefit trust makes it more layered than the usual setup.

That layered approach may reflect the distinct challenges of building advanced AI. Unlike consumer apps or social networks, frontier model companies often present hard trade-offs between speed, safety, commercial pressure, and long-range responsibility. Governance, in other words, becomes part of the product strategy.

What happens next?

The immediate next step is a shareholder vote on the proposed structure. If approved, Anthropic would head toward its IPO with a governance system designed to keep the founding team in the driver’s seat.

The longer-term implications will depend on how public investors react, how the company performs in the market, and whether the founders remain unified. If the structure works as intended, Anthropic will enter the public era with unusual insulation from external pressure. If it creates friction, the company may find itself balancing mission protection against shareholder expectations sooner than planned.

Either way, Anthropic is signaling that it does not want the IPO to be a surrender of control. It wants public capital, but on terms that preserve the original team’s influence over one of the most important AI companies in the world.