Anthropic Prepares Supervoting Shares for Founders Ahead of Mega-IPO

Anthropic has been preparing to grant CEO Dario Amodei and its other co-founders a class of stock carrying extra voting power, The Information reported on August 18, citing two people familiar with the matter. The move, still fluid in its specifics, is designed to insulate the leadership of the Claude maker from external shareholder pressure ahead of an initial public offering that could land as soon as September — and that is widely expected to rank among the biggest market debuts in history.

The report lands at a peculiar moment for the company. Anthropic’s annualized revenue run rate topped $65 billion at the end of July, per Reuters — roughly double what OpenAI is tracking — and Q2 revenue of $11.6 billion came with the company’s first quarter of positive adjusted operating income. Yet the people who founded the company hold remarkably little of it. Amodei himself owns only about 2% of Anthropic, according to a person close to the company cited by The Information. Supervoting shares are the classic instrument for resolving that asymmetry.

What we know about the plan

According to the reporting, the supervoting structure would give Amodei and fellow co-founders — a group that includes President Daniela Amodei — outsized voting power relative to their economic stake. Crucially, the specifics of the voting arrangements could not be learned, and the plans could still change before the S-1 lands. Anthropic did not immediately respond to a Reuters request for comment.

What is clearer is what the plan is for. Institutional ownership of Anthropic has been climbing — it is now believed to exceed 60% in economic terms across investors such as Google, Amazon, and the syndicates behind its recent Series H — and a public float would dilute founder influence further. The supervoting class is a preemptive lock on mission control:

• Insulation from activist pressure. Public-market investors can agitate for faster monetization, cheaper safety spending, or licensing deals that a safety-focused leadership team might resist.
• Continuity of the safety mission. Anthropic has consistently argued that its governance is built to survive shareholder pressure, not just describe it.
• Parity with peers. Dual-class structures are table stakes among founder-led AI and tech companies — a point the Reuters report makes explicitly.

The comparisons are instructive. At Meta, Mark Zuckerberg holds about 60% voting control through super-voting shares despite owning a minority of the economic equity. SpaceX’s dual-class structure gives Elon Musk decisive voting power. OpenAI, by contrast, spent 2025 unwinding its own nonprofit-controlled structure under political and legal pressure — a reminder that governance architecture is never static, and that founders who lack contractual control tend to lose arguments about it.

The trust stays

The most consequential detail in the report may be what is not changing: Anthropic also plans to maintain its existing body of non-shareholder trustees, who hold a special class of stock — Class T shares — empowering them to elect a majority of the company’s board of directors over time. That is the Long-Term Benefit Trust (LTBT), the independent oversight body Anthropic created in 2023 to hold it accountable to its public-benefit mission, independent of both founders and investors.

The result is a three-layer governance stack that no other frontier lab is attempting:

1. Founders with supervoting shares — operational and strategic control insulated from dilution and activist pressure.
2. The LTBT with board-elected Class T shares — mission oversight that no shareholder, including the founders, can simply vote away.
3. Public benefit corporation status — the legal shell that obliges the company to balance commercial success against public benefit, and gives the trust’s mandate teeth.

Each layer answers a different failure mode. Supervoting shares protect against hostile capital; the trust protects against founder drift; the PBC charter protects against both. Whether the stack survives contact with public markets is the open question — public investors have historically demanded governance concessions in exchange for the liquidity and scale of a flagship listing, and index funds in particular have campaigned against multi-class structures for years.

The financing signal

The governance news did not arrive alone. Hours earlier, Bloomberg reported that Anthropic’s pre-IPO revolving credit facility is set to exceed its roughly $10 billion target, with lead banks — Goldman Sachs, Morgan Stanley and JPMorgan among them — asked to commit about $1.25 billion each, and secondary participants around $1 billion. Credit lines of this size historically serve one purpose: proving to IPO underwriters that the company can fund operations and capital expenditure without dependence on sequential private rounds.

Read together, the two reports sketch the final pre-filing checklist. Revenue trajectory: confirmed, and record-setting. Profitability: achieved, at least on an adjusted operating basis. Bank financing: oversubscribed. Governance: being finalized now — and the founders intend to keep the keys.

Why it matters

For an industry still argueing about who should control frontier AI, Anthropic’s IPO will be the first real-world stress test of mission-locked governance at public-market scale. If the supervoting-plus-trust structure holds through the inevitable first activist campaign or takeover rumor, it becomes the template other safety-oriented labs will copy. If it buckles — if the trust is quietly diluted or the founders’ extra votes get “simplified” away in a later restructuring — it will confirm skeptics who argue that governance promises in AI are just preferences that expire when the money gets serious.

The September window is not confirmed. But the direction of travel is: the most consequential AI company by revenue is about to ask public markets to buy equity in a company whose board they will never control, on the strength of a promise that a trust will keep it honest. How investors price that promise is, in a real sense, how the market prices safety itself.

---

Figures in this article are drawn from The Information, Reuters, and Bloomberg reporting published August 18, 2026. Anthropic has not publicly confirmed the supervoting plan or the IPO timing.