• AI CapEx concerns: The massive AI buildout—estimated at up to $8 trillion in CapEx requiring $800 billion in profits just to cover interest—is heavily funded by private debt, creating potential risks of default and broader distress if a correction occurs.
• Infrastructure payback cycle: Industry leaders compared current AI spending to long-term infrastructure assets like hotels or data centers, arguing that immediate returns should not be expected during early buildouts that typically take 5 to 15 years to yield results.
• AI stack dynamics: The AI market is divided into infrastructure, foundational, and enterprise application layers, with foundational model creators facing the highest obsolescence risk while application-layer providers are positioned for durable value creation.
• Indian IT positioning: Indian IT services are poised to benefit from enterprise application modernization and addressing tech debt, though panelists emphasized the need for India to aggressively scale up deep-tech R&D investments.