Friend or Foe – The Impact of ChatGPT on Capture the Flag Competitions
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
ChatGPT, an artificial intelligence (AI)-based chatbot, has taken the world by storm since the technology’s release to the public in November 2022. The first reactions were awe and amazement as ChatGPT presented the capability to instantly respond to various text-based questions following a conversational approach. However, it is ChatGPT’s ability to complete more advanced tasks, such as supplying source code to programming-related questions or generating complete articles focusing on a specific topic, which has caused eyebrows to be raised. The capabilities offered by ChatGPT, fuelled by popularity and easy accessibility, have introduced several new challenges for the academic sector. One such challenge is the concept of AI-assisted cheating, where students utilise chatbots, such as ChatGPT, to answer specific questions or complete assignments. Although various research studies have explored the impact of ChatGPT on university education, few studies have discussed the influence of ChatGPT on Capture the Flag (CTF) competitions. CTF competitions offer a popular platform to promote cybersecurity education, allowing students to gain hands-on experience solving cybersecurity challenges in a fun but controlled environment. The typical style of CTF challenges usually follows a question-answer format, which offers students the ideal opportunity to enlist the assistance of ChatGPT. This paper investigates the ability of ChatGPT to assist and aid students in solving CTF challenges. The exploratory study involves past CTF challenges across various categories and the questioning of ChatGPT in an attempt to solve the challenges. The outcome of the study reveals that although ChatGPT can assist students with challenges during CTF competitions, the assistance that can be offered is minimal. Instead of producing answers to CTF challenges, ChatGPT can merely offer insight or guidance regarding the questions asked.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it