Economic Survey Flags AI Bubble Risk as Leaders Question Long Term Viability

The Economic Survey 2025–26 has raised serious concerns about a possible bubble forming in the global artificial intelligence market. The warning comes amid rapid and aggressive investments in AI infrastructure, especially in data centres, with signs that financial risks may be building beneath the surface. 

According to the Survey, more than 120 billion dollars worth of data centre spending linked to AI has been moved off the balance sheets of major technology companies. Instead of being shown directly in company accounts, these investments are being funded through special purpose vehicles backed by Wall Street investors. This structure, while helping companies scale quickly, is also creating hidden debt and financial exposure. 

Adding to these concerns, the CEO of IBM has publicly questioned the economics of Large Language Model based AI. His remarks have sparked wider debate about whether current AI business models can deliver sustainable returns in the long run. The Survey notes that such scepticism strengthens fears that the industry may be making a very large and risky bet on AI technology. 

The report also highlights the risk of a broader systemic shock. If financial stress combines with technological limitations and geopolitical tensions, the impact could spread quickly across global markets. In a worst case scenario, the Survey warns that the fallout could be more severe than the 2008 global financial crisis. 

For industries like manufacturing and automation, this serves as an important signal. While AI remains a powerful tool for productivity and innovation, the focus may shift towards more practical, value driven adoption rather than unchecked spending. The message from the Survey is clear. Growth must be balanced with financial discipline to avoid long term instability.