ENLIGHTENED POST

Explore, Engage, Enlighten

The not-so-glamorous side of Anthropic’s record-breaking IPO

Anthropic’s record-breaking initial public offering (IPO), which raised $81.31 billion at a valuation of around $1.5 trillion, has highlighted the enormous investor appetite for artificial intelligence. But behind the company’s rapid growth and demand for its Claude models lie mounting financial, infrastructure and cybersecurity risks.

The wider AI industry is committing unprecedented sums to infrastructure. The European Central Bank’s November 2026 Financial Stability Review estimates that cumulative capital expenditure by US Big Tech companies could reach $5.5 trillion by 2030. It also warns that AI investment requirements could exceed 3 per cent of US GDP annually between 2025 and 2032.

The concern is that companies are increasingly relying on debt and external financing as infrastructure spending exceeds operating cash flows. The ECB has also flagged circular financing arrangements, in which chipmakers, cloud providers and AI companies invest in one another, potentially inflating expected revenues. Big Tech’s off-balance-sheet commitments exceeded $4 trillion in the second quarter of 2026, according to the review.

Anthropic’s own financial disclosures illustrate the scale of the commitments. The company has around $900.9 billion in computing and infrastructure arrangements, including approximately $451.6 billion expected to be incurred before the end of 2030. It reported revenue of $33.24 billion in the first nine months of 2026, compared with $2.55 billion in the same period last year, but recorded an operating loss of around $1.26 billion.

There are also physical constraints. AI data centres require substantial electricity, cooling systems, batteries and backup generators. A Bloomberg investigation found that fluctuating power demands were putting pressure on this equipment, causing malfunctions and faster wear in some facilities.

Cybersecurity has emerged as another concern. Anthropic disclosed that three of its AI models accessed real-world systems during cybersecurity evaluations after a testing environment was inadvertently left connected to the internet. OpenAI separately reported that its research models bypassed isolation controls during evaluations and accessed parts of its research infrastructure and other systems.

These incidents involved testing environments, not publicly available chatbots independently hacking organisations. However, they highlight the risks of deploying increasingly capable models without adequately secured systems.

For investors, Anthropic’s valuation represents a bet on more than customer demand for AI. The company must also secure computing capacity, control infrastructure costs, protect its systems and turn rapid revenue growth into sustainable profits.

Source: India Today

Leave a Comment

Your email address will not be published. Required fields are marked *

Search Here

Follow Us

Recent Posts