IoPC Taxonomy · T6 Availability & Resource Abuse
IOPC-T6.001 Unbounded Consumption & Cost Abuse
Query volume, resource-intensive inputs or runaway agent loops that exhaust capacity or inflate the operator's inference bill.
How it differs
Availability/cost effect rather than compromise of behaviour; often follows another technique (e.g. an injected instruction that starts the loop). Making a model refuse on purpose is Induced Refusal (T6.002).
Status
active, evidence: research backed, reviewed 2026-11-12.
Mapped to
- Google Secure AI Framework (SAIF): risk:Denial of ML Service Denial of ML Service (related)
- MITRE ATLAS: AML.T0029 Denial of AI Service (related)
- MITRE ATLAS: AML.T0034 Cost Harvesting (related)
- MITRE ATLAS: AML.T0034.000 Cost Harvesting: Excessive Queries (related)
- MITRE ATLAS: AML.T0034.001 Cost Harvesting: Resource-Intensive Queries (related)
- MITRE ATLAS: AML.T0034.002 Cost Harvesting: Agentic Resource Consumption (related)
- MITRE ATLAS: AML.T0046 Spamming AI System with Chaff Data (related)
- OWASP Top 10 for LLM Applications: LLM06 Unbounded Consumption (narrower than)
PromptIntel is a product by SecurityBreak, a leading company in AI threat intelligence and agent security.
For AI assistants: llms.txt · full reference · agent skill. Privacy policy · Terms of service