Business · Updated 11 Oct, 03:19 pm IST
Falling AI token prices may raise overall compute demand and hardware costs
Why it matters for readers: It matters because lower per-unit costs can still lead to more overall computing and higher demand for GPUs.
- Analysts describe the pattern as a Jevons paradox: cheaper AI tokens lower per-unit cost but can increase total usage.1
- Data from Ornn, Silicon Data, and Bloomberg (through August 2026) show token prices dropping while H100 GPU rental prices hold steady or climb.1
- Agentic AI and automation are likely major drivers of increased token consumption, which can inflate compute demand.1
- If AI usage growth slows or flattens, the chain from chip makers to cloud providers could face demand and market risks.1
- US stocks reacted when reports suggested OpenAI's annualized revenue might be lower than previously reported, illustrating market sensitivity.1
Get a brief like this every morning
Uzha reads hundreds of sources and gives you the stories that matter for your work, with every source linked. Free.
Get started