# AI Code Generators for Security: Friend or Foe?

arXiv: https://arxiv.org/abs/2402.01219 · HTML: https://mlsanity.com/paper/2402.01219/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 12 references that are in the map.
- First posted on arXiv 2 Feb 2024.

## Builds on (most cited first, 4 of 4)

- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [From ChatGPT to ThreatGPT: Impact of Generative AI in Cybersecurity and Privacy](https://mlsanity.com/paper/2307.00691/) (2023)
- [Who Evaluates the Evaluators? On Automatic Metrics for Assessing AI-based Offensive Code Generators](https://mlsanity.com/paper/2212.06008/) (2022)
- [DualSC: Automatic Generation and Summarization of Shellcode via Transformer and Dual Learning](https://mlsanity.com/paper/2202.09785/) (2022)

## Built on by (most cited first, 2 of 2)

- [Enhancing AI-based Generation of Software Exploits with Contextual Information](https://mlsanity.com/paper/2408.02402/) (2024)
- [Willing but Unable: Separating Refusal from Capability in Code LLMs via Abliteration](https://mlsanity.com/paper/2606.05396/) (2026)

## Abstract (arXiv)

Recent advances of artificial intelligence (AI) code generators are opening new opportunities in software security research, including misuse by malicious actors. We review use cases for AI code generators for security and introduce an evaluation benchmark.