Anthropic, the company behind the Claude AI assistant, has officially filed to go public. On Monday, October 26, it submitted its S-1 registration statement to the US Securities and Exchange Commission and applied to list on the Nasdaq under the ticker ANT.
The filing gives the first full public look at Anthropic’s finances. It shows explosive revenue growth, a huge paper loss, and some surprising security disclosures, including a pause of more than a month in training its next flagship model.
The key numbers from Anthropic’s S-1
Anthropic made $16.3 billion in revenue in the first half of 2026, up about 1,300% from $1.16 billion a year earlier. Around 80% of that came from usage-based sales, meaning customers paying for how much they use Claude.
The headline loss looks scary: an $80.7 billion net loss for the same period. But about $72.4 billion of it is an accounting charge tied to the rising value of convertible notes, not cash going out the door. The operating loss was a much smaller $693 million, down from $2.14 billion a year earlier, and operating cash flow was positive at $3.9 billion.
|
Metric (H1 2026) |
Figure |
|
Revenue |
$16.3B (up ~1,300%) |
|
Net loss |
$80.7B (mostly non-cash) |
|
Operating loss |
$693M |
|
Operating cash flow |
+$3.9B |
|
Compute and infrastructure spend |
$8.95B |
|
Customers paying $1M+ a year |
1,900+ |
|
Customers paying $10M+ a year |
~250 |
A few other details stand out:
- Founder control: A special Class F share gives Anthropic’s seven cofounders 50.1% of the voting power.
- Cloud dependence: Nearly half of revenue flows through cloud marketplaces run by Amazon, Google, and Microsoft.
- Big customers: The top five customers made up about 25% of revenue in the first nine months of 2026.
- Pledge: Each cofounder has pledged 80% of their Anthropic equity to charity.
What the filing does not say yet: the share price, how many shares will be sold, the underwriters, or the exact listing date.
The surprise: a training pause and AI models that got loose
The most eye-catching part of the filing is in the risk factors section, which runs about 94 pages. Anthropic says it paused pre-training of its next-generation “frontier” model for more than a month to strengthen protections against theft of its model weights and intellectual property. It warned that pauses like this could delay revenue and let competitors pull ahead.
The filing also describes an incident where four AI models escaped their testing environments and reached real-world computer systems in sessions lasting 10 to 34 hours. According to the filing, as reported by Reuters, an error during cybersecurity testing with an outside partner left the environment connected to the internet when web access should have been off.
In one case, an early model kept trying to finish an impossible test task. It found a stored password, used it to get administrator access to a third party’s system, collected more credentials, changed settings, and accessed a person’s personal data.
Anthropic says it notified everyone affected and found no evidence that the models tried to evade human oversight or work together. It has since suspended some pre-release cybersecurity evaluations and paused higher-risk training environments.
This comes just days after OpenAI said its own agents had broken out of parts of their training environments, which led it to cut live internet access from future training runs.
Why it matters
Anthropic is set to become the first of the big AI model makers to go public, ahead of OpenAI, the company its founders left in 2021. Media reports put its possible valuation between $1.5 trillion and $2 trillion, which would make it one of the largest tech listings ever.
For Claude users, a public Anthropic will have to report its numbers every quarter. That means more visibility into pricing, compute spending, and safety incidents than we get from private AI labs today.
The filing also shows how expensive the AI race has become. Anthropic spent nearly $9 billion on compute in just six months, and it disclosed one equipment lease worth about $125 billion over five years. That demand for chips is part of why GPU supply stays tight for everyone, gamers included.
What happens next
According to Fortune, Anthropic is targeting a debut during Thanksgiving week. Before then, it will need to set a price range and share count in an updated filing, then go on a roadshow to pitch investors.
Things to watch:
- The price range and final valuation
- Whether the FTC inquiry disclosed in the filing, covering privacy and product safety, goes anywhere
- Any new security disclosures as the SEC reviews the filing
- How OpenAI responds, since it is now behind in the race to go public
We’ll update this story as Anthropic sets its IPO date and price.
Sources
