AI’s electronic waste footprint far larger than previously believed

← Back to the feed

AI’s electronic waste footprint far larger than previously believed

The Verge · 52 minutes ago

A new report from the nonprofit Basel Action Network (BAN) warns that the electronic waste generated by the artificial intelligence boom has been significantly underestimated, potentially producing enough discarded equipment by 2050 to fill 23 million shipping containers. This matters because the world already struggles to safely process e-waste, with less than a quarter of the 68.3 million tonnes generated annually being formally recycled, and much of the rest ending up in informal, hazardous recycling operations abroad that expose workers, including children, to toxic materials such as lead and chromium.

BAN's estimate is far higher than previous studies because it accounts for the full infrastructure supporting data centres, not just servers and processors, including cooling systems, power supplies, backup power and networking equipment, as well as a broader "AI Waste Contagion" category covering obsolete telecoms gear and personal devices. The group projects global e-waste will triple to up to 211 million tonnes a year by 2050, with AI responsible for 15 to 20 percent of that, translating to between 395 and 617 million tonnes of AI-related e-waste by mid-century. This dwarfs earlier, more conservative projections, such as a 2024 study's estimate of 1.2 to 5 million tonnes by 2030, largely because past research overlooked roughly 87 percent of a data centre's electro-mechanical infrastructure.

  • AI could generate up to 617 million tonnes of e-waste by 2050
  • New report counts far more infrastructure than previous estimates did
  • Most e-waste already goes unrecycled, often dumped in hazardous informal sites

AI Americas Technology World

Read the full article at the source →

Originally published by The Verge as “The AI data center e-waste problem is huge — and getting bigger”.