Tuesday, November 17, 2026

Anthropic’s IPO: Will Singapore investors back AI’s promise or help fuel another bubble?

SINGAPORE – Anthropic’s upcoming listing will put the market’s appetite for artificial intelligence to the test.

Do investors still see AI as a lucrative investment opportunity? Or have concerns over lofty valuations, soaring spending and uncertain returns brought them back down to earth?

The Nov 19 Nasdaq listing of the company behind the Claude chatbot comes amid significant volatility in AI-related stocks.

Anthropic is seeking a valuation of up to US$1.4 trillion (S$1.8 trillion), according to its latest initial public offering (IPO) plans. It is offering 825 million shares at between US$70 and US$78 apiece, potentially raising US$64.4 billion, including shares sold by existing shareholders, which could make it the biggest IPO in history.

A strong debut could imply that investors still see fortunes to be made from AI.

On the other hand, a lacklustre debut might suggest that investors are becoming more discerning about the prices they are prepared to pay for the stock, while a subsequent fall in Anthropic’s share price could signal either a broader loss of confidence in AI or a reassessment of the company’s valuation and prospects.

This does not have to be a bad thing, though. Given the transformational potential and benefits of AI, a correction could be an opportunity for more sensible, long-term value investors to enter the market.

Promise and peril

Founded in 2021, the San Francisco-based company develops the Claude family of AI models used by individuals and businesses for tasks ranging from writing and software development to research and data analysis.

Anthropic plans to use the proceeds from the offering and a concurrent US$26 billion private placement to fund general corporate purposes, including investments in computing infrastructure, research and development, and the expansion of its business.

Its prospectus points to the technology’s potential to accelerate scientific discoveries, help researchers tackle diseases such as cancer and improve productivity across both the public and private sectors.

In healthcare, for example, AI could help doctors identify diseases earlier and researchers develop new treatments. In education, it could provide personalised learning support, while businesses could use it to improve efficiency and reduce costs.

The strategic and national importance of AI is also evident in the lengths to which the United States has gone to restrict China’s access to advanced semiconductors.

On Nov 16, Washington announced new restrictions requiring US export licences for shipments of certain advanced AI chips to data centres in Singapore, Malaysia, Thailand and Indonesia, among other countries, as it seeks to prevent Chinese companies from accessing American computing power through overseas facilities.

The rule underscores the urgency with which the Commerce Department is scrambling to crack down on Chinese access to sensitive American technology, following reporting that the agency turned a blind eye to smuggling of advanced Nvidia AI chips to China.

But Anthropic’s prospectus also lays bare the formidable costs and risks of developing the technology.

The company further cautions that sufficiently advanced AI systems could pose catastrophic or existential risks to humanity if developed or deployed without adequate safeguards.

Such risks could arise if AI systems were weaponised, misused or allowed to operate with a high degree of autonomy, particularly if advances in the technology outpace developers’ ability to evaluate and control them.

The prospectus also highlights the possibility that increasingly autonomous systems could take harmful actions without human intervention.

Meanwhile, the company is still unprofitable, recording a net loss of US$104 billion for the first nine months of the year, compared with US$25.1 billion during same period in 2025.

Anthropic has also disclosed approximately US$900.9 billion in purchase and lease commitments, much of it related to computing capacity and infrastructure extending over many years.

These levels of expenditure raise questions about whether the company can sustain its growth, maintain its technological advantage and eventually generate sufficient profits to justify investors’ expectations.

Lessons from past bubbles

But even a strong debut could prompt questions about whether speculation is running ahead of economic reality.

In this scenario, investors would do well to remember what happened during the dotcom boom.

In the late 1990s, investors poured money into internet companies on the assumption that the new technology would transform commerce and communication.

While they were right about the internet’s potential, many were wrong about the companies they backed and the prices they paid.

The lesson was that investors could correctly anticipate a technological revolution and still lose fortunes betting on it.

The cryptocurrency boom and bust offers another warning.

Cryptocurrencies attracted investors with promises of decentralised finance, better financial infrastructure and attractive returns. But soaring crypto prices also encouraged speculation, excessive leverage and a culture of hype and excess.

The collapse of the TerraUSD stablecoin and its sister token Luna in May 2022 triggered heavy losses and exposed vulnerabilities across the sector.

The crisis deepened in November that year with the implosion of FTX, once one of the world’s largest cryptocurrency exchanges. Founder Sam Bankman-Fried was subsequently convicted of fraud and sentenced to 25 years in prison.

The crypto collapse showed that investor enthusiasm can easily obscure fundamental questions about viability, governance, transparency and risk.

While AI is already generating substantial commercial revenue and delivering practical benefits across industries, the danger is the forming of another a speculative bubble.

When expectations eventually disappoint, a sharp market correction could erode confidence in the entire sector, including companies with sound business models and genuinely promising technologies. The result could be a setback for innovation and development of AI, even if the technology’s underlying potential remains intact.

Anthropic’s investors therefore face the difficult task of putting a price on a technology that promises benefits, even as its developers acknowledge costs and risks that are difficult to predict and quantify.

What is clear is that greater scrutiny following a Democratic sweep in the US midterms would put pressure on AI companies to exercise financial discipline, rein in excessive spending and demonstrate that their products can generate sustainable profits.

Importantly, it could encourage investors to pay closer attention to safety and corporate governance, particularly when failures could have far-reaching consequences for society and shareholders.

Nvidia offers a stark reminder of why governance matters.

The chipmaker, whose market value grew more than tenfold from the end of 2022 to its peak in October 2026, became the poster child of the AI boom under co-founder Jensen Huang.

But Huang’s abrupt resignation in November exposed the risks of a company so closely identified with one individual, particularly when it had no publicly known successor ready to take over.

Nvidia has lost about US$1.8 trillion in market value since its October peak, weighed down by concerns over AI valuations compounded by uncertainty over its leadership and future direction.

Its predicament shows that even a company with strong earnings and a commanding market position can suffer when investors begin questioning its governance.

AI may well change the world for the better. But that does not mean investors should pay any price for a stake in its future.

Source : https://www.straitstimes.com/business/anthropics-ipo-will-singapore-investors-back-ais-promise-or-help-fuel-another-bubble

spot_img

Latest Articles