Recommended Free Tools
In a first-person account, Kehinde Owolabi says he received a zero on a beginner Python assignment after a grader questioned whether his work had received too much outside help. His follow-up changes an important part of the story: he says the instructor told him the grader—not the instructor—had flagged it, and that the instructor intended to have the assignment graded normally. The available account does not establish the final grade or whether an AI detector was used.
What the assignment asked for—and what Owolabi says he added
Owolabi wrote about the incident in a DEV Community post. He says he was 18, had been coding since 15, and was taking an introductory Python course he identified as CSE 110 at BYU-Pathway. According to his account, the assignment was a word-guessing game: the player guessed a secret word, received hints, had guesses counted, and had to enter guesses of the right length. Read Owolabi’s original account.
As an Amazon Associate I earn from qualifying purchases.
He says he added a menu, random word selection, a help screen, sound effects using winsound with a fallback, loading animations, navigation commands, and emoji. He reports that the initial feedback said the assignment appeared to have received too much outside help. That description comes from Owolabi; the assignment record, grade, and grader’s reasoning have not been independently established.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Who flagged the work, and what happened next?
In a follow-up post, Owolabi says his instructor told him that the instructor had not flagged the assignment; a grader had. He quotes the instructor as saying the code looked immediately like Owolabi’s work and that the instructor would tell graders to assess it normally. Read the follow-up account.
#1 Best Overall
The quoted exchange also raises a practical issue with the sound feature: the instructor reportedly could not get winsound working, disabled it, and ran the program without sound. The instructor reportedly said graders might dock points for the component and that Owolabi could resubmit without it if needed. This is Owolabi’s report of a private exchange, not an independently verified transcript. The follow-up does not document the final grade, so it does not establish that the zero was formally overturned or that the work ultimately received a normal grade.
Why polished or feature-rich code is not proof of AI use
Owolabi says he pointed to imperfections in his own program as evidence of how he had written it. He describes making a word list by splitting a string with .split(" "), leaving the alphabet as an impossible guessable entry, using a bare except:, mixing singular and plural variable names, duplicating entries, and relying on Windows-only winsound. These are the author’s descriptions of his code, not the result of an independent code review.
Those details do not prove human authorship. A polished program can be written by a student, and a rough program can be produced with outside assistance. Feature count, stylistic polish, or visible mistakes alone cannot reliably establish who wrote code. A more useful question is whether the student can explain the choices in the submission and show how the work developed, considered alongside the course’s rules and the evidence available to the grader.
What a separate study can—and cannot—say about suspicion
A 2024 PLOS ONE study by Scarfe and colleagues tested GPT-4-generated answers in five undergraduate psychology modules at the University of Reading. For those online exams, researchers reported that 94% of AI submissions went undetected under the study’s standard procedures. The study also reported an 83.4% chance that AI submissions would outperform a randomly selected group of the same size from real student submissions; that figure describes the researchers’ resampling analysis, not detector accuracy. The AI answers averaged around half a classification boundary higher than the real student work.
The findings are limited to that university, those psychology modules, the study’s 2023 exam setting, and its method. They do not measure how often current systems misclassify Python assignments, establish what happened in Owolabi’s class, or show how reliable any detector is for programming work. The paper says its exam submission and grading systems did not include an AI detector. It describes markers considering several suspicion cues, including work that seemed “too good to be true,” material not taught on the module, and citations that did not support claims; it also reports that some colleagues mentioned comparisons with ChatGPT or online AI detectors in a few suspected cases. Suspicion, detector output, and proof of authorship are different things.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.If a grader questions your own programming work
Owolabi’s proposed response is to stay calm, explain relevant prior experience, point to concrete code decisions and development evidence, and offer to walk through the program, answer questions, live-code, or submit a simpler version that matches the rubric. These are options he suggested, not a guarantee that an appeal will succeed.
- Read the feedback and course rules. Identify what specifically was questioned and check which tools, collaboration, and resubmissions the course permits. Follow the course’s stated appeal process and deadlines; the rules for Owolabi’s course were not established in the accounts.
- Gather authentic process evidence. If you already have them, organize drafts, version history, notes, test cases, or other records that show how you developed the work. Do not create or alter records to make them appear contemporaneous.
- Explain the code concretely. Be ready to describe how the program meets the assignment requirements, why you chose particular structures or libraries, how you tested it, and what you would change. A walkthrough can clarify understanding, but it is not by itself proof of authorship.
- Ask what would resolve the concern. A respectful request for the specific concern and the next step can help keep the conversation focused. If a feature is incompatible with the grading environment, ask whether an allowed alternative or resubmission is available rather than assuming one will be accepted.
In this account, the central correction is that “I got a zero” describes Owolabi’s reported initial result, while his follow-up says the instructor planned to have the work graded normally after learning a grader had flagged it. The reported instructor response matters, but the available posts do not establish the final outcome.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
