Packback: Supporting NOVA Faculty in Detecting AI Use and Promoting Academic Integrity

January 9, 2025 / Tech Notes & Training

Academic Affairs is committed to providing tools and resources that support academic integrity and foster student success. As part of this effort, we are excited to share details about Packback, a platform designed to enhance student engagement through writing assignments while promoting originality. One of its newest features, The Originality Fingerprint, is now available for NOVA faculty to review after students submit their assignments (e.g., discussion boards, papers, etc). This tool provides insights through two critical indicators:

  1. The Uniqueness indicator evaluates the percentage of a student’s content that is not matched to external or internal sources:
  • External Sources: Content available on the web.
  • Internal Sources: Content from previously submitted Packback assignments.

Understanding the Uniqueness Score:

  • Lower Uniqueness Scores:
    • Indicate a larger portion of the content matches other sources.
    • May require further review to determine if plagiarism is a concern.
    • Important Note: Currently, quoted content and in-text citations are not excluded from Packback’s scan. If a score appears low, reviewing the Originality Report can provide additional context about matched sources.
  • Higher Uniqueness Scores:
    • Indicate a smaller portion of the content matches other sources.
    • Reflect greater originality in the student’s submission.

Looking Ahead: The platform will soon include scans of paywalled journal article sources, further enhancing its detection capabilities.

  1. The AI Risk indicator assesses the likelihood that a student’s content was generated using artificial intelligence. This feature analyzes word choice and sequence to provide educators with a clearer understanding of how the content may have been developed.

Levels of AI Risk:

  • Low Risk:
    • Suggests no significant evidence of AI-generated text.
    • While not definitive, this indicates a lower likelihood of AI usage.
  • Moderate Risk:
    • Suggests some characteristics of AI-generated content.
    • Faculty may wish to review the content more closely and, if needed, discuss the student’s process for developing their work.
    • Important Note: Human-written content can occasionally be flagged as Moderate Risk.
  • High Risk:
    • Suggests a significant portion of the content may have been generated by AI.
    • Requires careful evaluation and potentially a follow-up conversation with the student.

Supporting Academic Integrity at NOVA

Similar to tools like Turnitin.com, the Originality Fingerprint is a valuable addition to the NOVA teaching toolkit for promoting academic integrity and encouraging authentic work from our students. By utilizing these indicators, NOVA faculty can:

  • Identify potential issues with plagiarism.
  • Evaluate the likelihood of AI-generated content.
  • Engage students in conversations about originality and authorship.

If you are interested in integrating this tool into your classroom or have questions about Packback, please schedule an appointment with Ri Cantlin, Packback’s Senior Curriculum Consultant, at ri.cantlin@packback.coor https://hello.packback.co/c/Ri_Cantlin#/select-time.

Submitted by:
Cynthia Pascal, AVP of ELearning, CPascal@nvcc.edu