Attack  ·  Glossary

LLM usage-counter / billing injection

An attack that manipulates the counters that track how many tokens or queries an AI service has used, letting an attacker spend AI compute without paying or corrupting billing. It can inflate or erase usage, hiding theft or breaking accounting. AI compute is metered and expensive, so gaming these counters is profitable.
An attacker who can game the metering can run large workloads on the victim's bill, and it distorts the cost controls leadership relies on.
Track this in the live feed See how this plays out in real AI security and governance developments.
Open the feed →