What it means that Anthropic is going public in weeks with a $2 trillion target valuation — August 26, 2026
· CompaniesAutomation
Radar Flash Edition: Reuters publishes the five things to know before Anthropic's market debut. $65 billion in projected annualized revenue, $42 billion in losses in 2025, a target valuation above $2 trillion, and social backlash against AI listed as a risk in the prospectus. What changes for your company if Claude is under your automations.
Flash edition. Anthropic —the company behind Claude, the model powering a large portion of the automations being built today in Spanish SMEs— is going public in a matter of weeks. Reuters published early this morning the five things to know before the debut. This complements this morning's Radar: while that looked at who uses AI, this is about who sells it and how much it costs them to keep selling it to you.
What happened
Reuters summarizes the filing ahead of the IPO (Source, Source). Founded in 2021 by former OpenAI executives, led by Dario Amodei and his sister Daniela, with about 5,000 employees according to PitchBook. They placed a narrow bet—programmers—and it paid off: Claude Code is their most popular product and is pushing projected annualized revenue to $65 billion. The flip side: they lost nearly $42 billion in 2025 and will continue to be in the red. They raised $65 billion in May at a valuation near $1 trillion ($965 billion) and now their investors are aiming for an October debut above $2 trillion (Source). It seeks to raise more than the $86.2 billion SpaceX raised in June: it would be the largest IPO in history. With two shadows looming: the Trump administration broke its contracts in March after it refused to give unrestricted military access to its models—the litigation could last years—and the prospectus will list social backlash against AI as an explicit risk factor (Source).
Why it matters
Until today, your model provider answered to a handful of patient investors. From the debut onwards, it answers to the market every quarter. This changes three things for the buyer. One: the pressure on pricing stops moving in only one direction. Burning $42 billion to gain market share is a defensible decision in private; doing it with a listed stock is a different conversation. Two: the numbers stop being rumors. For the first time, you will be able to read margins, customer concentration, and inference costs of your provider in an audited document, instead of deducing them from leaks. And three: the litigation with the Administration and the "AI backlash" in the prospectus are not reputational noise, they are service continuity risks declared by the company itself.
For your business
Three concrete moves, today. One: look at what percentage of your automations depends on a single model. If it's 100%, you don't have an integration, you have a dependency. Two: make the prompt and orchestration layer portable—so that changing providers means modifying an environment variable and rerunning your evaluations, not rewriting the system. With a battery of your own tests on real cases, migrating takes an afternoon; without it, it takes a quarter. And three: freeze pricing by contract where you can. If you are on list pricing and consume serious volume, now is the time to negotiate an annual commitment before the stock price starts weighing in on your margins. None of this is betting against Anthropic: it's about not letting your company's roadmap be written by someone else's earnings calendar.
Frequently Asked Questions
When is Anthropic going public and at what valuation?
Reuters places the debut weeks away (August 26, 2026) and cited investors point to October with a target valuation exceeding $2 trillion. The last private benchmark is the $965 billion from the May 2026 round. These are market expectations, not final figures: the final price is set at the placement.
Should I stop using Claude in my company because of this?
No. An IPO does not degrade the product; it adds transparency and quarterly pressure. What is advisable is to stop treating any model provider as eternal infrastructure: keep your prompts, your data, and your evaluations outside the provider, and you can decide with a calculator in hand when rates change.
What is the "AI backlash" listed as a risk in the prospectus?
The social and political rejection of AI—regarding jobs, energy, copyright, or safety—translated into financial risk. For a company to write it in its S-1 means it anticipates that regulation or public opinion could slow down its business. For you, it's a reminder that your adoption plan also needs an internal response: what you automate, what you don't, and how you communicate it to your team.