Automated TEE Adaptation with LLMs: Identifying, Transforming, and Porting Sensitive Functions in Programs

TL;DR

This paper introduces AUTOTEE, an LLM-based system that automatically identifies, transforms, and ports sensitive functions into Trusted Execution Environments, simplifying secure program adaptation with high accuracy.

Contribution

AUTOTEE is the first approach leveraging LLMs to automate TEE adaptation, reducing manual effort and domain expertise required for secure program transformation.

Findings

AUTOTEE achieves a 0.94 F1 score on Java functions.

AUTOTEE achieves an 84.3% success rate on Python functions.

Constructed a benchmark dataset of 385 sensitive functions from 68 repositories.

Abstract

Trusted Execution Environments (TEEs) isolate a special space within a device memory that is not accessible to the normal world (also known as the untrusted environment), even when the device is compromised. Therefore, developers can utilize TEEs to provide robust security guarantees for their programs, protecting sensitive operations, such as encrypted data storage, fingerprint verification, and remote attestation, from software-based attacks. Despite the robust protections offered by TEEs, adapting existing programs to leverage such security guarantees is challenging, often requiring extensive domain knowledge and manual intervention, which makes TEEs less accessible to developers. This motivates us to design AUTOTEE, the first Large Language Model (LLM) enabled approach that can automatically identify, transform, and port functions containing sensitive operations into TEEs with minimal developer intervention. By manually reviewing 68 repositories, we constructed a benchmark dataset consisting of 385 sensitive functions eligible for transformation, on which AUTOTEE achieves a F1 score of 0.94 on Java and 0.87 on Python. AUTOTEE effectively transforms these sensitive functions into TEE-compatible versions, achieving success rates of 91.8% and 84.3% for Java and Python, respectively, when using GPT-4o.