English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper presents the FormAI dataset, a large collection of 112, 000AI-generated compilable and independent C programs with vulnerabilityclassification. We introduce a dynamic zero-shot prompting techniqueconstructed to spawn diverse programs utilizing Large Language Models (LLMs).The dataset is generated by GPT-3.5-turbo and comprises programs with varyinglevels of complexity. Some programs handle complicated tasks like networkmanagement, table games, or encryption, while others deal with simpler taskslike string manipulation. Every program is labeled with the vulnerabilitiesfound within the source code, indicating the type, line number, and vulnerablefunction name. This is accomplished by employing a formal verification methodusing the Efficient SMT-based Bounded Model Checker (ESBMC), which uses modelchecking, abstract interpretation, constraint programming, and satisfiabilitymodulo theories to reason over safety/security properties in programs. Thisapproach definitively detects vulnerabilities and offers a formal model knownas a counterexample, thus eliminating the possibility of generating falsepositive reports. We have associated the identified vulnerabilities with CommonWeakness Enumeration (CWE) numbers. We make the source code available for the112, 000 programs, accompanied by a separate file containing thevulnerabilities detected in each program, making the dataset ideal for trainingLLMs and machine learning algorithms. Our study unveiled that according toESBMC, 51.24% of the programs generated by GPT-3.5 contained vulnerabilities,thereby presenting considerable risks to software safety and security.

The FormAI Dataset: Generative AI in Software Security Through the Lens of Formal Verification