Skip to content

Three AI-deterrence-modified autogradable Qiskit assignment packages (HW1–HW3) with layered personalization and grading

Design · All designs
Evidence · 2 claims (1 for, 1 against) · 1 study (1 design), q1 · 0 of 1 report an effect size · 2 claims rest on one study

Description

Three complete take-home homework packages for an undergraduate quantum software course: HW1 (seeded basis-state circuits with bit flips and customized measurement mappings), HW2 (QFT followed by inverse-transform recovery with circuit metrics and an optional hardware extension), and HW3 (seeded Deutsch–Jozsa with customized oracle masks). Each used a deterministic student identifier so an instructor-side reference could regenerate the configuration, and layers included exact JSON submissions, hidden deterministic references, circuit metrics, reflections, and optional IBM Quantum execution.

Design Implications

Context

Requirements

  • An instructor-side reference that regenerates each student's seeded configuration and expected output so personalization works without a separate answer key per student

Constraints

  • In the tested designs, these deterrence layers did not prevent direct ChatGPT completion

Target Learners

  • undergraduate quantum software development students

Learning Goals

  • count-key ordering in Qiskit
  • QFT and inverse-QFT recovery
  • Deutsch–Jozsa oracle construction and classification

Claims

Examples

-

Key Sources

  • Alexei Kaltchenko and Gurnivaj Tiwana. (2026). ChatGPT Solves All Tested Qiskit Homework Assignments. arXiv. https://arxiv.org/abs/2608.19707
Page metadata
Field Value
type design
id ai-deterrence-qiskit-assignment-packages
title Three AI-deterrence-modified autogradable Qiskit assignment packages (HW1–HW3) with layered personalization and grading
description Three complete take-home homework packages for an undergraduate quantum software course: HW1 (seeded basis-state circuits with bit flips and customized measurement mappings), HW2 (QFT followed by inverse-transform rec...
status review
generated by process:settle-candidates on 2026-10-10
trust unverified — no one has confirmed this page's content
sources 1 — alexei-kaltchenko-and-gurnivaj-tiwana-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.