Why enterprise GPU utilization is stuck at 5% — and why the fix makes it worse

Enterprise GPU fleets are averaging only about 5% utilization as teams keep capacity reserved during a severe supply crunch.

Cast AI says this hoarding behavior is wasting expensive AI infrastructure and distorting how companies manage GPU procurement.

At the same time, top-end GPU and memory pricing is rising or no longer falling, suggesting scarcity is now reaching even reserved and frontier hardware.

The shift marks a break from cheaper cloud compute, with persistent shortage pressure likely to reshape AI budgets and buying decisions.