Findings
AI Input Tokens Elevated
Updated: June 19, 2025
Description
An increase in the number of input tokens being sent to the AI model has been observed.
Excessive input size may result in higher processing costs, slower response times, and API throttling.
Potential Causes:
- Unoptimized or overly detailed user inputs.
- Inefficient preprocessing or tokenization of data.
- Abuse of AI services, such as automated bot-driven input.
Remediation
Limit input sizes where possible and filter unnecessary data.
Security Frameworks
Unbounded Consumption occurs when a Large Language Model (LLM) application allows users to conduct excessive and uncontrolled inferences, leading to risks such as denial of service (DoS), economic losses, model theft, and service degradation.
Adversaries may target machine learning systems with a flood of requests for the purpose of degrading or shutting down the service. Since many machine learning systems require significant amounts of specialized compute, they are often expensive bottlenecks that can become overloaded. Adversaries can intentionally craft inputs that require heavy amounts of useless compute from the machine learning system.
Adversaries may target different machine learning services to send useless queries or computationally expensive inputs to increase the cost of running services at the victim organization. Sponge examples are a particular type of adversarial data designed to maximize energy consumption and thus operating cost.