San José, California - Amazon said on Thursday it would spend an additional $20 billion this year to expand its artificial intelligence infrastructure, citing sharply higher memory chip prices and surging demand for computing capacity.

Chief executive Andy Jassy raised the company's forecast for capital expenditures to $220 billion from $200 billion, telling analysts in a conference call that even that level of investment would not be enough to meet customer demand for cloud computing resources.

He added the supply-demand imbalance was likely to persist into 2027.

The e-commerce and technology company has been a major beneficiary of the AI boom through its Amazon Web Services (AWS) cloud business, the world's largest provider of cloud computing infrastructure.

Daily Horoscope Today's horoscope: Free daily horoscope for Friday, July 31, 2026

AWS revenue rose 37% from a year earlier to $42.2 billion in the second quarter, topping analysts' average estimate of about $40.5 billion.

Jassy said Amazon was already seeing "remarkable" demand emerging for 2028 and that the company now believed AWS could eventually generate more than $1 trillion in annual revenue.

Amazon shares rose as much as 10% in extended trading following the results. 

The company's total quarterly revenue increased 20% from a year earlier to $200.6 billion.

Net income jumped to roughly $62.65 billion from $18.1 billion a year earlier, although the figure included $53.4 billion in pre-tax gains, including from the revaluation of the company's stake in AI start-up Anthropic.

Jassy said infrastructure spending was currently elevated because Amazon was building data centers from the ground up. Once completed, he said, the facilities were expected to last for more than 30 years, while the computing hardware inside them would need to be replaced only every five to six years.

The data center push has sparked widespread public backlash over their high electricity consumption, which can strain local grids and inflate energy bills, as well as their heavy water use, noise generation, and the relatively small number of jobs they create.