Anthropic files for massive Nasdaq IPO under ANT

For the past two years, the loudest bear case against the AI boom has been simple: frontier labs are bottomless money pits that will never make a real profit.

Anthropic pushed back hard. The AI lab officially filed its S-1 registration statement with the SEC to list on the Nasdaq under the ticker ANT, giving Wall Street its first unfiltered look at the business of frontier AI.

Jaw-dropping revenue and a massive cash moat

Anthropic generated $16.3 billion in revenue for the first six months of 2026, exploding roughly 14 times from $1.2 billion in the same period last year.

The bigger shocker is the bottom line. Anthropic logged a $974 million operating profit in the June quarter alone. Its first-half operating loss narrowed significantly to $693 million, down from a $2.1 billion loss in the first half of 2025.

While the company posted an $80.7 billion net loss due to noncash accounting charges tied to convertible debt, its liquidity is practically untouchable.

Anthropic has raised $141 billion to date and still has around $132 billion sitting in the bank. In other words, it has spent a net of just $9.4 billion to build a powerhouse that now pulls in $11.6 billion a quarter.

Big-ticket enterprise customers are clearly buying in. By late September, Anthropic had over 11,900 clients spending $100,000 or more per year, 1,900 spending over $1 million, and roughly 250 enterprise contracts north of $10 million.

Claude writes the code, founders hold the wheel

Inside the lab, the tools are already running the company. In September, Claude wrote over 90% of Anthropic's internal code.

Back in April, the retail value of compute consumed by Anthropic employees using Claude actually surpassed the firm's monthly cash payroll.

Yet the filing came with a startling admission in its risk disclosures: internal tasks run by Claude could eventually become difficult for any individual or group of humans to understand and control.

To keep that risk in check, Anthropic is going public with an unconventional governance moat. The seven co-founders formed Founder LLC to lock down 50.1% of shareholder voting power, while a Long-Term Benefit Trust retains the authority to appoint most board members.

All seven co-founders have also pledged 80% of their equity to charity, decoupling their personal financial upside from the firm's voting control.

Why it matters

Anthropic just handed the entire AI ecosystem its biggest validation yet.

Critics insisted frontier models were too expensive to ever be viable. Anthropic proved that not only is the demand real, but enterprise appetite can turn an operating profit even while maintaining massive research spend. It also sets a sky-high benchmark for OpenAI and rivals preparing their own public debuts.

For enterprise teams, this filing removes the single biggest existential worry about adopting Claude: vendor survival. With $132 billion still in the vault, Anthropic has the runway to outlast any near-term market winter.

Still, we think public investors will have a complex pill to swallow. The dual-class setup leaves public shareholders with practically zero governance power over a company building technology that poses catastrophic or existential risks.

Anthropic proved its business works. Now it has to prove that a mission-driven public benefit trust can coexist with public market quarterly expectations.