Anthropic files for IPO amid concerns over AI industry bubble

The path to a public debut

Anthropic is preparing to make its debut as a public company this November after filing its S-1 paperwork. The company, which is well-known for its Claude AI products, first filed its documents confidentially in June. While other competitors like SpaceX have already moved to market, investors have been closely watching Anthropic's progress for months.

The company plans to trade on the Nasdaq under the ticker ANT and expects the market debut to raise more than $100 billion in liquidity. However, the filing reveals significant gaps that the company must iron out before the launch. Specifically, the firm has not yet disclosed its total valuation or the exact number of shares it intends to sell to the public.

According to a report by The Wall Street Journal, these specific details will likely not be available for another two weeks. The company is currently gauging investor feedback before updating its paperwork. This delay suggests that the final price and scale of the offering are still being negotiated with potential backers.

A lopsided financial picture

The S-1 filing reveals a startling contrast between Anthropic's rapid revenue growth and its massive operating costs. In the first half of 2026, the company generated approximately $16.3 billion in revenue, a sharp jump from the $1.2 billion reported during the same period in 2025. Despite this growth, the company reported a net operating loss of $80.7 billion during that same timeframe.

The company noted that this massive loss largely reflects an accounting adjustment related to its various financing agreements. To justify its high price tag, Anthropic is relying on unprecedented revenue forecasts that project it will generate between $190 billion and $200 billion annually by 2028. These projections are based on growth rates that have never been achieved by any other company in history.

Investors are being asked to bet on a future that the company admits is difficult to evaluate due to a limited operating history. The filing acknowledges that the industry is still nascent and evolving. This accounting approach mirrors strategies used by SpaceX, which also hinged its valuation on projected revenue years into the future rather than current profits.

The cost of safety and compute

Anthropic has built its brand on a safety-first approach, but its filing suggests that these guardrails come with a heavy price tag. The company described its implementation of safety features as resource-intensive and noted that these costs are split between safety measures and raw computing power. However, the documentation admits that the actual returns on these specific guardrail investments remain unclear at this time.

The company's financial health is also tied to the enterprise usage of its Claude models. This creates a risk because many businesses are currently pulling back on AI usage due to high token costs. If companies continue to feel burned by the expense of running these models, Anthropic's projected revenue could face significant headwinds.

Furthermore, the company faces increasing competition from cheaper open-source models and price-cutting from rivals like OpenAI. Critics argue that hundreds of billions of dollars are being invested to potentially generate only single-digit billions in profit. This creates a precarious situation where the infrastructure costs for data centers may outpace the actual financial payoff for the broader economy.

Public trust and existential risk

The filing is notable for the amount of space it dedicates to risk factors, with nearly a third of the document discussing potential harms. Anthropic warns that its advanced models could pose catastrophic or existential risks to humanity if they are misused or operate with too much autonomy. These warnings serve as a double-edged sword, acting as both a legal disclosure and a form of marketing hype for the AI industry.

Public sentiment toward these technologies has shifted toward negativity as concerns over job displacement and safety grow. Anthropic's CEO, Dario Amodei, has been vocal about the need to slow down the pace of development to ensure proper oversight. He has even suggested that agentic models could potentially wipe out 50% of entry-level positions in the future.

Despite these pleas for caution, the company is still locked in a race to release increasingly powerful models. Once the company becomes beholden to public investors, it will be under intense pressure to produce unending quarterly gains. This creates a fundamental tension between the company's stated goal of safety and the market's demand for continuous, rapid growth.