IB EE, IA or TOK Draft Flagged as AI? What the IB's Own Policy Says

At draft stage this is your school's process, run under your school's academic integrity policy. The IB's policy says "If teachers are not convinced that the work is the student’s own, it must not be submitted to the IB", says detection tools are unlikely for a while to be reliable enough "to be the sole evidence of academic misconduct", and, if the school decides the work is not yours, leaves what happens next to the school, for example whether you can submit new work, rewrite under supervision, or lose the chance to submit. So the person you have to satisfy is your teacher, and the policy lists what they look for: that they saw the work develop over time, that you can explain it, and that it is clear which ideas are yours and which are quoted.

You are not the only one. In September 2026, IB students on Reddit posted Turnitin AI scores on an Extended Essay draft (77%) and a TOK exhibition (100%), among others. Most of those drafts had not gone to the IB yet. That is the situation this page is for.

The IB policy has a paragraph for this moment, under "Cases identified by the school": "When a school identifies issues with a final piece of work before submission or upload to the IB and before the IB submission deadline, the situation must be resolved as per the school’s academic integrity policy, provided it specifies if resubmissions are permitted."

So the outcome, including whether you get to resubmit, sits with your school's own policy. That paragraph is written for a final piece of work, and the policy has no separate procedure for drafts, so a flag on a draft comes earlier than anything the IB lays out. Appendix 6, the IB's guidance on AI tools, gives examples of what a school might decide once it concludes a student has submitted work that is not their own: whether students may "submit entirely new work, to rewrite it under supervision, or do they lose the opportunity to submit anything?" Those are outcomes for after that conclusion, not for the moment a draft gets flagged.

The IB keeps one rule for itself: "work that is not the student’s own cannot be submitted for assessment". The stakes come in the next sentence: "Teachers must keep in mind that if a student does not submit coursework, then the IB will not award a grade in that subject."

One more line is worth knowing, because it shows the position your supervisor is in. The policy's list of school maladministration, meaning wrongdoing by the school rather than the student, includes "Authenticating work when there are doubts regarding its authenticity." So a teacher who signs off work they are unsure about is doing something the policy counts against the school. Telling them the score is only a number will not get you the signature. Giving them a reason to stop doubting might.

The IB is openly sceptical of AI detectors. Appendix 6 says: "The IB aims to avoid joining the “arms race” between AI tools that claims to be able to tell the difference between AI and human authors. It is better to ensure that students can speak about their work with their teachers rather than accusing them of misconduct based on increasing small differences in the predictability of their word choices."

A page later it adds: "While tools are being developed to detect essays written by AI tools, it is likely to be a while before they are reliable enough to be the sole evidence of academic misconduct that leads to a student losing their grade for the subject concerned." That was written in March 2023, when the appendix was added. As far as we can check, the IB has not replaced it (see the sources note at the end).

Read both carefully. The IB is not saying your score is wrong, and it does not tell schools to ignore detectors. It says a score alone should not cost you a grade, and that talking to you about your work is the better test. The same appendix is strict about AI text itself: "the IB does not regard any work produced—even only in part—by such tools to be their own."

Turnitin says something similar about its own number. Its FAQ: "the percentage on the AI writing indicator should not be used as the sole basis for action or a definitive grading measure by instructors." It also says "The AI writing detection indicator and report are not visible to students. However, with the PDF download feature, instructors can download and share the AI report with students." If all you have been told is a percentage, ask for the PDF, so you can see which passages were highlighted.

Appendix 6 asks: "What does the IB expect when a teacher checks the authentication box to confirm that work is the student’s own?" It gives four answers. The policy says "teacher" throughout and never uses the word "supervisor", so for an Extended Essay, read it as whoever signs your essay off.

The first row matters most. A detector only sees the finished text. If your teacher has watched the essay change over several drafts, they have evidence the detector does not. That is why the history of the document counts for more than a debate over the percentage. Our guide on how to keep version history in Word, Google Docs and Overleaf covers where that history lives and how to get copies out. Its warning applies here too: do not retype the essay to manufacture a history. A history manufactured after the flag is worse than having none.

If part of the draft came from ChatGPT or something like it, even a paragraph you then reworded, the rules are specific. The appendix says students must be told: "If they use the text (or any other product) produced by an AI tool—be that by copying or paraphrasing that text or modifying an image—they must clearly reference the AI tool in the body of their work and add it to the bibliography." The in-text citation goes in quotation marks, in the referencing style your school already uses, and has to include "the prompt given to the AI tool and the date the AI generated the text."

Left uncited, the appendix calls it "misrepresenting content", which it describes as "a form of academic misconduct". A draft is the cheapest point at which to fix that. Tell your supervisor what you used, cite it in that format, and where a cited block of AI text is doing work your own analysis should be doing, replace it with your own. The appendix is blunt about why: AI text is "most likely not of good enough quality to be awarded high marks".

It also says what is fine. Using a search engine to find research material is "entirely reasonable", and "Students should be encouraged to ask the software research questions rather than the essay title, and then explore the sources it provides". Asking an AI tool where to look is a different thing from asking it to write.

Appendix 6 has a section called "Using software to improve language and grammar". It covers everything "from simple spell checkers to complex tools that rewrite sentences", and says IB assessments usually do not mark language quality, "so there is limited benefit in using such tools." What it says, point by point:

• Language acquisition subjects: "marks are awarded for sentence structure. In these subjects the use of such tools is not permitted."
• Writing in one language and translating: "students are not permitted to write essays in one language and then translate them to be submitted to the IB in another language."
• Everything else: "the use of spell checkers and bilingual dictionaries is acceptable", and "The IB allows students to use basic tools to support their spelling and grammar when this is not what is being assessed."
• Access arrangements: "The IB will always consider the use of software to support access and inclusion requirements for students." If you have approved access arrangements, they are a separate question for your coordinator.

Separately, the policy's list of student responsibilities includes to "abstain from receiving non-permitted assistance in the completion or editing of work, such as from friends, relatives, other students, private tutors, essay writing or copy-editing services, pre-written essay banks or file sharing websites".

Put together, running a flagged draft through a tool that rewrites your sentences is not one of the remedies this policy describes. In a language acquisition subject it is banned outright. In other subjects, if the tool is an AI tool, what comes back is text produced by an AI tool: the appendix does not regard work produced "even only in part" by such tools as the student's own, and the citation rule above covers it. A cited machine rewrite of your own essay also does nothing to answer your teacher's actual question, which is whether the work is yours.

1. Find out exactly what was flagged. Which tool, which passages. If it was Turnitin, ask for the PDF of the AI report, since students cannot see it in the system themselves.
2. Read your school's academic integrity policy, specifically the part on resubmission. If your school decides the work is not yours, the IB leaves the outcome to the school; its examples are new work, a rewrite under supervision, or no submission. Your IB coordinator can tell you what your school's policy allows and how much time you have before the IB deadline.
3. Export your drafts today. Save copies of earlier versions out of Google Docs or Word into a folder you do not edit, so you are not relying on a version panel staying as it is.
4. Prepare to explain it out loud. Take the three or four most heavily highlighted passages and make sure you can say, without notes, where each idea came from and which source backs it. "The student can explain their work sufficiently" is the IB's own test.
5. If you used AI anywhere, say so now and cite it the way the appendix asks: tool, prompt, date, in the text and in the bibliography.
6. If you are offered a rewrite, keep every dated draft of it, and ask whether it will be supervised. A rewrite that leaves no trail puts you back where you started.
7. If a teacher looks at your evidence and tells you to leave the draft as it is, ask them to confirm it by email, so there is a record if the question comes up again before submission.

If your school moves to a formal meeting, our guide to the academic integrity meeting after an AI accusation covers what to bring.

Does the IB run an AI detector on my EE after I submit it? The policy does not say it does. Across its 63 pages it never mentions Turnitin, and it does not describe the IB running an AI detector on submitted work. Appendix 6 says what the IB is good at catching: it is "very effective" at "spotting similar work used by students in different schools—for example, two students buying the same essay off the internet", while for work written by a parent, tutor or essay mill "the IB is less likely than the teacher to be able to identify this." The policy also says "The IB will act upon all reports of alleged student academic misconduct and school maladministration, regardless of when they reach the organization during the assessment cycle", and that "When academic misconduct is identified after the work has been submitted to the IB, the programme coordinator must inform the IB as soon as possible." In other words, a problem found after submission still reaches the IB.

Can my school refuse to send my Extended Essay to the IB? Yes. "If teachers are not convinced that the work is the student’s own, it must not be submitted to the IB." The policy also says a school "should also not submit work with a mark of zero in the relevant component", and that a final piece with plagiarised content is "awarded an “F” on the IB internal assessment mark entry system" for an IA, or, for externally assessed components "such as the extended essay, or theory of knowledge essay", "marked as non-submission". Either way, there is no grade in that subject.

Can I appeal to the IB? The IB appeals process in the policy is for "The outcome of an investigation into school maladministration or student academic misconduct", and "An appeal will only be considered if there is evidence that the IB has failed to follow its processes correctly or new evidence is present that was unknown to the IB at the time of the original decision." A decision your school makes before submission is resolved "as per the school’s academic integrity policy", so the route to challenge it would be in your school's policy; the IB appeals process in this policy covers the outcome of investigations.

Can my supervisor go through my rewritten draft line by line? Probably not in the way you would hope. Among teachers' responsibilities the policy lists "giving feedback and ensuring students are not provided with multiple rounds of editing, which would be contrary to instructions described in the relevant subject guides". Expect a conversation about the work, not a second edit of it.

• International Baccalaureate, Academic integrity policy (published October 2019, updated March 2023): "Cases identified by the school" (p. 23), teacher and student responsibilities (pp. 13, 15), the appeals paragraph (p. 18), Appendix 1 (p. 27) and Appendix 6, "Guidance on the use of artificial intelligence tools" (pp. 53–55). This is the version the IB's own AI page links to (page marked "Last updated: 19 February 2026"). ibo.org answered our requests on 2 October 2026 with a browser check, so we read the IB's file through the Internet Archive copy captured on 10 November 2025, which is byte-for-byte the same as every complete archived copy since April 2023. If your school gives you a newer edition, that one wins.
• Turnitin, AI writing detection capabilities FAQs, read 2 October 2026.