CWE-1427 — Improper Neutralization of Input Used for LLM Prompting

Description

The product uses externally-provided data to build prompts provided to large language models (LLMs), but the way these prompts are constructed causes the LLM to fail to distinguish between user-supplied inputs and developer provided system directives.

When prompts are constructed using externally controllable data, it is often possible to cause an LLM to ignore the original guidance provided by its creators (known as the "system prompt") by inserting malicious instructions in plain human language or using bypasses such as special characters or tags. Because LLMs are designed to treat all instructions as legitimate, there is often no way for the model to differentiate between what prompt language is malicious when it performs inference and returns data. Many LLM systems incorporate data from other adjacent products or external data sources like Wikipedia using API calls and retrieval augmented generation (RAG). Any external sources in use that may contain untrusted data should also be considered potentially malicious.

Consequences

Confidentiality, Integrity, Availability — Execute Unauthorized Code or Commands, Varies by Context

Confidentiality — Read Application Data

Integrity — Modify Application Data, Execute Unauthorized Code or Commands

Access Control — Read Application Data, Modify Application Data, Gain Privileges or Assume Identity

Mitigations

Phase: Architecture and Design

LLM-enabled applications should be designed to ensure proper sanitization of user-controllable input, ensuring that no intentionally misleading or dangerous characters can be included. Additionally, they should be designed in a way that ensures that user-controllable input is identified as untrusted and potentially dangerous.

Phase: Implementation

LLM prompts should be constructed in a way that effectively differentiates between user-supplied input and developer-constructed system prompting to reduce the chance of model confusion at inference-time.

Phase: Architecture and Design

LLM-enabled applications should be designed to ensure proper sanitization of user-controllable input, ensuring that no intentionally misleading or dangerous characters can be included. Additionally, they should be designed in a way that ensures that user-controllable input is identified as untrusted and potentially dangerous.

Phase: Implementation

Ensure that model training includes training examples that avoid leaking secrets and disregard malicious inputs. Train the model to recognize secrets, and label training data appropriately. Note that due to the non-deterministic nature of prompting LLMs, it is necessary to perform testing of the same test case several times in order to ensure that troublesome behavior is not possible. Additionally, testing should be performed each time a new model is used or a model's weights are updated.

Phase: Installation, Operation

During deployment/operation, use components that operate externally to the system to monitor the output and act as a moderator. These components are called different terms, such as supervisors or guardrails.

Phase: System Configuration

During system configuration, the model could be fine-tuned to better control and neutralize potentially dangerous inputs.

Detection

Dynamic Analysis with Manual Results Interpretation

Use known techniques for prompt injection and other attacks, and adjust the attacks to be more specific to the model or system.

Dynamic Analysis with Automated Results Interpretation

Use known techniques for prompt injection and other attacks, and adjust the attacks to be more specific to the model or system.

Architecture or Design Review

Review of the product design can be effective, but it works best in conjunction with dynamic analysis.