Why would an attacker prefer GPUs over ASICs when targeting a proof-of-work network?

An attacker would prefer GPUs because they can be sold or reused after the attack, whereas ASICs cannot. GPUs are general-purpose graphics chips that can mine different algorithms, run games, train AI models, and more — so if an attack fails or succeeds, the attacker still has hardware worth selling. ASICs are locked to a single algorithm, so attacking a network with them means those chips become worthless the moment the network changes or the attack ends. For Kaspa specifically, this asymmetry means ASIC-based attacks carry a punishing financial penalty that GPU-based attacks do not.

Not financial advice. This content is for education only. Nothing here is financial advice.