How secure is AI-generated Code: A Large-Scale Comparison of Large Language Models

• URL: http://arxiv.org/abs/2404.18353v2
• Date: Wed, 11 Dec 2024 13:02:30 GMT
• Title: How secure is AI-generated Code: A Large-Scale Comparison of Large Language Models
• Authors: Norbert Tihanyi, Tamas Bisztray, Mohamed Amine Ferrag, Ridhi Jain, Lucas C. Cordeiro,
• Abstract summary: This study compares state-of-the-art Large Language Models (LLMs) on their tendency to generate vulnerabilities when writing C programs.<n>At least 62.07% of the generated programs are vulnerable.
• Score: 3.4887856546295333
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This study compares state-of-the-art Large Language Models (LLMs) on their tendency to generate vulnerabilities when writing C programs using a neutral zero-shot prompt. Tihanyi et al. introduced the FormAI dataset at PROMISE'23, featuring 112,000 C programs generated by GPT-3.5-turbo, with over 51.24% identified as vulnerable. We extended that research with a large-scale study involving 9 state-of-the-art models such as OpenAI's GPT-4o-mini, Google's Gemini Pro 1.0, TII's 180 billion-parameter Falcon, Meta's 13 billion-parameter Code Llama, and several other compact models. Additionally, we introduce the FormAI-v2 dataset, which comprises 331 000 compilable C programs generated by these LLMs. Each program in the dataset is labeled based on the vulnerabilities detected in its source code through formal verification, using the Efficient SMT-based Context-Bounded Model Checker (ESBMC). This technique minimizes false positives by providing a counterexample for the specific vulnerability and reduces false negatives by thoroughly completing the verification process. Our study reveals that at least 62.07% of the generated programs are vulnerable. The differences between the models are minor, as they all show similar coding errors with slight variations. Our research highlights that while LLMs offer promising capabilities for code generation, deploying their output in a production environment requires proper risk assessment and validation.

Related papers

• Case Study: Fine-tuning Small Language Models for Accurate and Private CWE Detection in Python Code [0.0]
  Large Language Models (LLMs) have demonstrated significant capabilities in understanding and analyzing code for security vulnerabilities. This work explores the potential of Small Language Models (SLMs) as a viable alternative for accurate, on-premise vulnerability detection.
  arXiv  Detail & Related papers  (2025-04-23T10:05:27Z)
• Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities [63.603861880022954]
  We introduce ADV-LLM, an iterative self-tuning process that crafts adversarial LLMs with enhanced jailbreak ability. Our framework significantly reduces the computational cost of generating adversarial suffixes while achieving nearly 100% ASR on various open-source LLMs. It exhibits strong attack transferability to closed-source models, achieving 99% ASR on GPT-3.5 and 49% ASR on GPT-4, despite being optimized solely on Llama3.
  arXiv  Detail & Related papers  (2024-10-24T06:36:12Z)
• Exploring RAG-based Vulnerability Augmentation with LLMs [19.45598962972431]
  VulScribeR is a novel solution that leverages carefully curated prompt templates to augment vulnerable datasets. Our approach beats two SOTA methods Vulgen and VGX, and Random Oversampling (ROS) by 27.48%, 27.93%, and 15.41% in f1-score with 5K generated vulnerable samples on average.
  arXiv  Detail & Related papers  (2024-08-07T23:22:58Z)
• Automated Software Vulnerability Static Code Analysis Using Generative Pre-Trained Transformer Models [0.8192907805418583]
  Generative Pre-Trained Transformer models have been shown to be surprisingly effective at a variety of natural language processing tasks. We evaluate the effectiveness of open source GPT models for the task of automatic identification of the presence of vulnerable code syntax.
  arXiv  Detail & Related papers  (2024-07-31T23:33:26Z)
• Exploring Automatic Cryptographic API Misuse Detection in the Era of LLMs [60.32717556756674]
  This paper introduces a systematic evaluation framework to assess Large Language Models in detecting cryptographic misuses. Our in-depth analysis of 11,940 LLM-generated reports highlights that the inherent instabilities in LLMs can lead to over half of the reports being false positives. The optimized approach achieves a remarkable detection rate of nearly 90%, surpassing traditional methods and uncovering previously unknown misuses in established benchmarks.
  arXiv  Detail & Related papers  (2024-07-23T15:31:26Z)
• Uncovering Weaknesses in Neural Code Generation [21.552898575210534]
  We assess the quality of generated code using match-based and execution-based metrics, then conduct thematic analysis to develop a taxonomy of nine types of weaknesses. In the CoNaLa dataset, inaccurate prompts are a notable problem, causing all large models to fail in 26.84% of cases. Missing pivotal semantics is a pervasive issue across benchmarks, with one or more large models omitting key semantics in 65.78% of CoNaLa tasks. All models struggle with proper API usage, a challenge amplified by vague or complex prompts.
  arXiv  Detail & Related papers  (2024-07-13T07:31:43Z)
• SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal Behaviors [64.9938658716425]
  Existing evaluations of large language models' (LLMs) ability to recognize and reject unsafe user requests face three limitations. First, existing methods often use coarse-grained of unsafe topics, and are over-representing some fine-grained topics. Second, linguistic characteristics and formatting of prompts are often overlooked, like different languages, dialects, and more -- which are only implicitly considered in many evaluations. Third, existing evaluations rely on large LLMs for evaluation, which can be expensive.
  arXiv  Detail & Related papers  (2024-06-20T17:56:07Z)
• M2CVD: Enhancing Vulnerability Semantic through Multi-Model Collaboration for Code Vulnerability Detection [52.4455893010468]
  Large Language Models (LLMs) have strong capabilities in code comprehension, but fine-tuning costs and semantic alignment issues limit their project-specific optimization. Code models such CodeBERT are easy to fine-tune, but it is often difficult to learn vulnerability semantics from complex code languages. This paper introduces the Multi-Model Collaborative Vulnerability Detection approach (M2CVD) to improve the detection accuracy of code models.
  arXiv  Detail & Related papers  (2024-06-10T00:05:49Z)
• DeepSeek-Prover: Advancing Theorem Proving in LLMs through Large-Scale Synthetic Data [65.5290035371111]
  We introduce an approach to generate extensive Lean 4 proof data derived from high-school and undergraduate-level mathematical competition problems. We fine-tune the DeepSeekMath 7B model on this synthetic dataset, which comprises 8 million formal statements with proofs. Our model successfully proved 5 out of 148 problems in the Lean 4 Formalized International Mathematical Olympiad (FIMO) benchmark, while GPT-4 failed to prove any.
  arXiv  Detail & Related papers  (2024-05-23T09:03:42Z)
• A Comprehensive Study of the Capabilities of Large Language Models for Vulnerability Detection [9.422811525274675]
  Large Language Models (LLMs) have demonstrated great potential for code generation and other software engineering tasks. Vulnerability detection is of crucial importance to maintaining the security, integrity, and trustworthiness of software systems. Recent work has applied LLMs to vulnerability detection using generic prompting techniques, but their capabilities for this task and the types of errors they make remain unclear.
  arXiv  Detail & Related papers  (2024-03-25T21:47:36Z)
• How Easy is It to Fool Your Multimodal LLMs? An Empirical Analysis on Deceptive Prompts [54.07541591018305]
  We present MAD-Bench, a benchmark that contains 1000 test samples divided into 5 categories, such as non-existent objects, count of objects, and spatial relationship. We provide a comprehensive analysis of popular MLLMs, ranging from GPT-4v, Reka, Gemini-Pro, to open-sourced models, such as LLaVA-NeXT and MiniCPM-Llama3. While GPT-4o achieves 82.82% accuracy on MAD-Bench, the accuracy of any other model in our experiments ranges from 9% to 50%.
  arXiv  Detail & Related papers  (2024-02-20T18:31:27Z)
• Understanding the Effectiveness of Large Language Models in Detecting Security Vulnerabilities [12.82645410161464]
  We evaluate the effectiveness of 16 pre-trained Large Language Models on 5,000 code samples from five diverse security datasets. Overall, LLMs show modest effectiveness in detecting vulnerabilities, obtaining an average accuracy of 62.8% and F1 score of 0.71 across datasets. We find that advanced prompting strategies that involve step-by-step analysis significantly improve performance of LLMs on real-world datasets in terms of F1 score (by upto 0.18 on average)
  arXiv  Detail & Related papers  (2023-11-16T13:17:20Z)
• VGX: Large-Scale Sample Generation for Boosting Learning-Based Software Vulnerability Analyses [30.65722096096949]
  This paper proposes VGX, a new technique aimed for large-scale generation of high-quality vulnerability datasets. VGX materializes vulnerability-injection code editing in identified contexts using patterns of such edits. For in-the-wild sample production, VGX generated 150,392 vulnerable samples, from which we randomly chose 10% to assess how much these samples help vulnerability detection, localization, and repair.
  arXiv  Detail & Related papers  (2023-10-24T01:05:00Z)
• Zero-Shot Detection of Machine-Generated Codes [83.0342513054389]
  This work proposes a training-free approach for the detection of LLMs-generated codes. We find that existing training-based or zero-shot text detectors are ineffective in detecting code. Our method exhibits robustness against revision attacks and generalizes well to Java codes.
  arXiv  Detail & Related papers  (2023-10-08T10:08:21Z)
• A LLM Assisted Exploitation of AI-Guardian [57.572998144258705]
  We evaluate the robustness of AI-Guardian, a recent defense to adversarial examples published at IEEE S&P 2023. We write none of the code to attack this model, and instead prompt GPT-4 to implement all attack algorithms following our instructions and guidance. This process was surprisingly effective and efficient, with the language model at times producing code from ambiguous instructions faster than the author of this paper could have done.
  arXiv  Detail & Related papers  (2023-07-20T17:33:25Z)
• The FormAI Dataset: Generative AI in Software Security Through the Lens of Formal Verification [3.2925005312612323]
  This paper presents the FormAI dataset, a large collection of 112, 000 AI-generated C programs with vulnerability classification. Every program is labeled with the vulnerabilities found within the source code, indicating the type, line number, and vulnerable function name. We make the source code available for the 112, 000 programs, accompanied by a separate file containing the vulnerabilities detected in each program.
  arXiv  Detail & Related papers  (2023-07-05T10:39:58Z)
• CodeLMSec Benchmark: Systematically Evaluating and Finding Security Vulnerabilities in Black-Box Code Language Models [58.27254444280376]
  Large language models (LLMs) for automatic code generation have achieved breakthroughs in several programming tasks. Training data for these models is usually collected from the Internet (e.g., from open-source repositories) and is likely to contain faults and security vulnerabilities. This unsanitized training data can cause the language models to learn these vulnerabilities and propagate them during the code generation procedure.
  arXiv  Detail & Related papers  (2023-02-08T11:54:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.