ChatGPT flags possible AI stock market crash by September 2026 as MAS joins global caution over tech stock valuations

ChatGPT has suggested a “speculative date” that a possible artificial intelligence (AI) stock market crash could occur on September 17, 2026, according to a report by The Motley Fool.

Before the chatbot gave the specific date, it referred to an MIT study showing that 95% of firms using AI failed to make a profit and quoted Goldman Sachs’ CEO, who said he “wouldn’t be surprised if in the next 12 to 24 months we see a draw-down with respect to equity markets.”

Based on this, ChatGPT said a possible correction could happen within the next one to two years, between late 2025 and 2027, depending on triggers, with a “moderate to high” chance of a “meaningful correction of 30% to 50%” during that period.

The UK publication had asked ChatGPT how serious the concerns were about an upcoming stock market crash because of AI, and when it might happen, noting “maybe we need AI to predict an AI stock market crash” amid talks of risks linked to the technology sector’s elevated stock valuations.

It is worth noting that large language models (LLMs) like ChatGPT may provide inaccurate answers from time to time, as they can “hallucinate.”

In its new research paper “Why Language Models Hallucinate,” ChatGPT maker OpenAI described AI hallucinations as “plausible but false statements generated by language models,” which can appear even in response to seemingly straightforward questions.

Notably, Singapore’s central bank just joined the conversation, saying “some equity markets are seeing relatively stretched valuations, particularly in the technology and artificial intelligence (AI) segments,” in its annual Financial Stability Review (FSR) released on Wednesday, The Edge Singapore reported.

The Monetary Authority of Singapore (MAS) added that a pullback in optimism about AI’s ability to deliver sufficient future returns may lead to sharp corrections in the broader equity market and further defaults in the private credit market.

Policymakers and Wall Street executives have grown more vocal about inflated tech stock valuations driven by enthusiasm for AI.

The Edge Singapore reported that since Tuesday, a global selloff in semiconductor shares has wiped out around US$500 billion (S$651 billion) in market value after investors were let down by earnings forecasts from Palantir Technologies and Advanced Micro Devices.

MAS said in its annual review that much of the equity market’s rise has been fuelled by a surge in AI-related investments, leaving investors heavily exposed to the information technology sector.

It added that some major tech firms have turned to private financing arrangements, including the use of special purpose vehicles, private credit structures, and novel accounting treatment, which may mask leverage and raise funding dependencies, putting pressure on AI firms to generate sufficient revenue to repay a growing base of public investors and private creditors.

“The continued divergence between equity market valuations and rising downside risks to growth raises the prospects of disorderly corrections in the event of shocks,” MAS said.

A similar concern has been observed in venture capital funding, as GIC’s investment chief Bryan Yeo warned of a “hype bubble” forming in early-stage AI ventures, with global funding for the sector reaching a record US$192.7 billion as of October 2025.

Nvidia CEO Jensen Huang, however, doesn’t believe there is an AI bubble and said the industry has reached a “quite extraordinary” turning point that justifies the costly AI investments. /TISG

Read also: WeRide raises S$401M in Hong Kong share sale, Temasek among investors