Technology
Brookings study warns AI buildout requires massive power and financing growth
Slashdot reports that a Brookings Institution paper highlights economic risks and vast infrastructure requirements for artificial intelligence expansion.
The short version
- A Brookings Institution study by professor Stijn van Nieuwerburgh project that the AI buildout could average 3.63% of annual U.S. GDP.[Slashdot]
- The expansion requires adding an estimated 183 gigawatts of data-center capacity over seven years, compared to 57 gigawatts installed today.[Slashdot]
- To generate expected returns on this investment, AI revenues would need to reach $3.7 trillion annually by 2032, requiring roughly 80% annual growth from current baseline estimates.[Slashdot]
Key facts
- A Brookings Institution study notes the AI infrastructure footprint includes specialized chips, electricity, and purpose-built data centers.[Slashdot]
- The projected AI buildout is estimated to average 3.63 percent of U.S. GDP annually, exceeding historical U.S. investment booms in canals, railways, and electrification.[Slashdot]
- Columbia Business School professor Stijn van Nieuwerburgh estimates 183 gigawatts of new data-center capacity will be required over seven years, compared to 57 gigawatts currently installed.[Slashdot]
- Reaching expected returns on investment would require AI sector annual revenues to rise to $3.7 trillion by 2032, implying an 80% annual growth rate from current combined revenues.[Slashdot]
What remains uncertain
- Whether AI industry revenue can maintain the required ~80% annual growth rate to achieve $3.7 trillion in annual revenue by 2032 remains unproven.[Slashdot]
Sources
Outlet counts describe coverage, not independent confirmation. Reports may share a wire service or original source.