AI investment across the hyperscaler ecosystem could reach $1 trillion in 2027, up from about $700 billion this year and roughly $300 billion last year, JPMorgan Chase CEO Jamie Dimon said at the 11th annual JPMorgan India Conference on September 21. He said the spending could add about 1% to annual GDP growth while creating inflationary pressure as companies build factories, power plants and workforces. Dimon said AI expansion appeared likely to continue, but it was too early to identify the industry's winners and returns would not always be captured through conventional ROI metrics. He cited customer-experience improvements and investments required simply to remain competitive. Dimon has also warned about the security risks of Anthropic's Mythos AI model, which JPMorgan is using to stress-test internal systems. He said infrastructure projects, defense spending and persistent government deficits were pushing interest rates higher, while a market correction was possible although he was unsure AI would trigger it. Dimon said inflation could remain elevated or rise and urged the Federal Reserve to defend its 2% target. The investment boom is reshaping U.S. capital spending, with information-processing equipment reaching $752 billion in the second quarter versus $748 billion for residential fixed investment. Goldman Sachs expects U.S. business investment growth of 7.8% in 2026 and AI investment to exceed $800 billion by year-end. Consensus estimates cited by Apollo Global Management put six hyperscalers' combined spending at roughly $916 billion over the next 12 months and nearly $1.2 trillion the following year. On geopolitics, Dimon said the U.S. and China appeared to be making progress ahead of a summit between President Donald Trump and Chinese President Xi Jinping and urged engagement on trade, AI and security. He also called for India and the U.S. to complete a bilateral trade agreement, saying negotiations had not moved forward.