Anthropic Files for IPO at $2 Trillion Valuation. Public Backlash Is the Risk.

Anthropic filed a confidential IPO application in June, with investors placing a valuation of around $2 trillion on the maker of the Claude AI model series. The company's annualized revenue crossed $6.5 billion last month, up from $900 million at the end of last year, and it reported $1.16 billion in revenue for Q2. A Gallup poll showing that 70% of Americans oppose AI data centers near their homes is expected to appear as a named risk factor in the prospectus, a rare example of a public-sentiment issue making it into a formal filing.

What happened

Data point	Detail
IPO filing	Confidential submission, June 2025
Projected valuation	~$2 trillion (investor estimate)
Annualized revenue (latest month)	$6.5 billion+
Annualized revenue (end of 2024)	$900 million
Q2 2025 revenue	$1.16 billion
Expected revenue growth (YoY)	More than 10x
Public opposition to local data centers	70% of Americans (Gallup, May 2025)
Americans “strongly opposed”	Nearly 50%
Americans who support local projects	~25%

Anthropic was founded in 2021 by CEO Dario Amodei and several colleagues who left OpenAI over disagreements about the direction of AI development. For most of its life the company was seen as trailing OpenAI, but Claude Code’s rise in coding productivity helped accelerate revenue sharply. Management has told early investors that it expects models to keep improving and that stronger models will let it hold premium pricing even as rivals cut costs.

Private investor meetings are underway in San Francisco. CFO Krishna Rao has been fielding questions on three themes: competitive dynamics, the threat that open-weight models (models whose underlying parameters are public and can run cheaply elsewhere) pose to margins, and how quickly the company’s data center partners can expand capacity. A public prospectus is expected within weeks, with trading to begin sometime after that.

If the $2 trillion valuation holds, the offering would dwarf SpaceX’s $85.7 billion IPO last month, which the source notes was itself a record.

Why does public opposition to data centers matter for an IPO?

Compute is Anthropic’s hard constraint. More Claude users and new AI products require more servers, more GPUs, and more electricity. Data centers must be sited somewhere, and the Gallup figures show that “somewhere” is a problem: only about one in four Americans supports a data center being built nearby.

That figure is significant enough that the company is expected to list it as a formal risk factor in the prospectus. Risk factor sections in IPO filings carry legal weight. If a risk is named there and later materializes, investors have less grounds for a claim that they were not warned. Including community opposition in that section signals that slowed build-out is a scenario Anthropic’s lawyers consider realistic, not hypothetical.

Tech companies collectively have spent hundreds of billions of dollars on data centers and server GPUs this year alone, according to the source. Bottlenecks are already showing. The challenge for Anthropic is that its revenue growth story depends on being able to serve demand, and serving demand requires infrastructure that communities are pushing back on.

For context on how other companies are navigating the chip and infrastructure race, see our earlier coverage of Broadcom’s $100 billion AI chip financing effort, which names Anthropic as one of the beneficiaries.

The open-weight pricing problem

Investors are probing whether Anthropic can hold its prices as open-weight models improve. Open-weight models are AI systems whose parameters are published, meaning any company can run them at the cost of their own hardware rather than paying a per-token fee. If a competitor’s open model approaches Claude’s quality, enterprise buyers have reason to switch.

Management’s counter-argument is that Anthropic’s models will keep advancing faster than open alternatives, justifying premium pricing. Whether investors accept that logic will shape the IPO multiple they are willing to pay.

Our take

The revenue trajectory is real. Going from $900 million annualized at year-end to $6.5 billion annualized within roughly six months is not typical SaaS growth. Claude Code appears to be doing serious commercial work, and enterprise automation use cases are compounding quickly. If you are evaluating AI integration for your own business, Anthropic’s API is already a credible choice alongside OpenAI.

That said, the data center risk is not just an IPO disclosure exercise. It is a genuine operational ceiling. A company that cannot expand compute cannot grow revenue at the rate investors are pricing in. Community opposition, permitting delays, and power grid constraints are slow-moving problems that compound quietly. The fact that Anthropic is naming this explicitly before the IPO is credit to its disclosure team, but it does not make the problem smaller.

The open-weight question is the one we find most interesting for business readers. If the gap between frontier closed models and open alternatives narrows, the per-token pricing that funds Anthropic’s growth gets harder to defend. Watch how enterprises respond to Meta’s next Llama release and whether it puts visible pressure on Anthropic’s pricing.

What to do about it

1. If you currently use Claude via API, review your contract terms for pricing change provisions before the IPO closes. Post-IPO pressure to show margins could affect API pricing.
2. Test one open-weight alternative (such as Meta’s Llama series) against your current Claude workloads now, so you have a real benchmark rather than a hypothetical if prices move.
3. If you are considering AI automation for business workflows, factor infrastructure reliability into your vendor choice. A model provider with compute constraints may have availability or latency trade-offs. See our workflow automation service for how we approach this in client projects.
4. Watch the public prospectus when it drops. The risk factors section will give you a clearer picture of what Anthropic’s own lawyers think could go wrong.

The headline number is $2 trillion, but the prospectus risk factors will tell you more about Anthropic’s real situation than any valuation estimate.

Source: Bing News · Claude AI