Asleep at the Keyboard? Assessing the Security of GitHub Copilot's Code Contributions.

This research article seeks to exploit vulnerabilities in using Github Copilot for paired human-artificial intelligence code generation. The method employs MITRE’s top 25 Common Weakness Enumeration (CWE) list to evaluate Copilot’s performance in regards to three areas- diversity of weakness, divers...

Full description

Bibliographic Details
Published in:Communications of the ACM Vol. 68; no. 2; pp. 96 - 106
Main Authors: Pearce, Hammond, Ahmad, Baleegh, Tan, Benjamin, Dolan-Gavitt, Brendan, Karri, Ramesh
Format: Article
Published: Association for Computing Machinery Feb2025
Subjects:
Online Access:View this record in EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=182365552&site=ehost-live
header:
  @attributes:
    shortDbName: hlh
    uiTerm: 182365552
    longDbName: Humanities International Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        00010782
        ACM
      jtl: Communications of the ACM
      issn: 00010782
      maglogo: N
    pubinfo:
      dt: Feb2025
      vid: 68
      iid: 2
      pid: 68
      pub: Association for Computing Machinery
    artinfo:
      ui:
        182365552
        10.1145/3610721
      ppf: 96
      ppct: 10
      formats:
      tig:
        atl: Asleep at the Keyboard? Assessing the Security of GitHub Copilot's Code Contributions.
      aug:
        au:
          Pearce, Hammond
          Ahmad, Baleegh
          Tan, Benjamin
          Dolan-Gavitt, Brendan
          Karri, Ramesh
        affil:
          New York University, Department of ECE, New York, NY, USA
          University of Calgary, Department of ESE, Calgary, Alberta, Canada
      su:
        Human-artificial intelligence interaction
        Computer programming
        Github Inc.
        Computer security vulnerabilities
        Programming languages
      sug:
        subj:
          Human-artificial intelligence interaction
          Computer programming
          Github Inc.
          Computer security vulnerabilities
          Programming languages
      ab: This research article seeks to exploit vulnerabilities in using Github Copilot for paired human-artificial intelligence code generation. The method employs MITRE’s top 25 Common Weakness Enumeration (CWE) list to evaluate Copilot’s performance in regards to three areas- diversity of weakness, diversity of prompt, and diversity of domain. The discussion of the results includes an overview on threats to validity of the results including CWE and scenario inclusion and reproducibility of code generated.
      pubtype: Periodical
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: Y
      dt:
        @attributes:
          year: 2025
    holdings:
      @attributes:
        islocal: N