Support

Examiner AI has been designed with exam integrity at its core. The system cannot access, infer, or reproduce secure assessment content. It does not replicate or mimic exam board marking processes. All feedback is generated from our independent, proprietary logic, based solely on the question and answer provided by the user.

We explicitly prohibit the upload of live scripts, unseen papers, or any secure assessment materials. Examiners and teachers must comply with their awarding body obligations, including confidentiality requirements and restrictions on third-party processing.

Examiner AI is a formative tool designed to support learning and reduce workload. Examiner AI is not a summative marking system. It does not apply grade boundaries, does not predict grades, and does not contribute to official results. Exam integrity and regulatory compliance remain central to the design of the service

Examiner AI is a formative learning tool designed for classroom and home use. It must not be used for official marking or in any context involving secure or unseen assessment materials. Examiners contracted by exam boards are subject to strict confidentiality and data-processing agreements that prohibit uploading live scripts, unseen papers, or any secure assessment content to third-party tools. Doing so would constitute a breach of contract and may be considered malpractice.

Examiner AI is a formative learning tool and is not designed, nor should it be used for awarding grades or contributing to any formal assessment. As such, it does not fall within Ofqual’s regulatory framework for high-stakes marking or awarding.

However, we have designed Examiner AI in line with the principles outlined by Ofqual for the safe and responsible use of AI in education. In particular, Examiner AI:

• uses transparent, controlled prompts rather than opaque or self-updating models;
• is designed with a clear separation between formative support and any summative assessment processes;
• includes human oversight, with teachers retaining full control over how feedback is interpreted or used;

• avoids the use of secure or unpublished assessment materials;

• operates with a privacy-first architecture, minimising the data processed and ensuring zero-retention from AI vendors;

• provides feedback that is advisory, not determinative.

We aim to align with Ofqual’s emerging guidance on AI to support safe classroom use, but Examiner AI does not replace exam board marking, does not replicate awarding processes, and does not interface with regulated assessment systems. All summative assessment decisions remain solely with the awarding bodies.

Examiner AI is designed primarily as a formative learning tool to support classroom teaching, homework, revision, and practice questions. It may be used to provide feedback on internal assessments only where the school decides that such use is appropriate and where this aligns with the school’s own assessment policies.

However:

• Examiner AI should not be used to mark high-stakes internal exams where grades contribute to formal reporting, predicted grades, or progression decisions.
• Teachers must retain full professional judgement, reviewing and interpreting any AI-generated feedback before using it in classroom assessment.
• Schools should ensure that using Examiner AI for internal assessments does not conflict with any exam board or qualification-provider guidance, particularly where pre-release or controlled materials are involved.

Students should be made aware that the marks awarded by Examiner AI are indicative only, not definitive, and do not substitute for human marking.

Used appropriately, Examiner AI can support teacher workload and student learning, but it should not replace teacher assessment in situations where accuracy, professional judgement, or fairness are critical.

No. Examiner AI is a formative learning tool that provides feedback on individual answers. It does not give students access to secure assessment content, nor does it predict final grades, reveal exam board marking tolerances, or generate model exam answers. The feedback it provides reflects broad assessment principles, based on our independent interpretation of public documents, rather than any specific awarding-body judgement.

Because Examiner AI does not attempt to replicate exam board standardisation or apply grade boundaries, it cannot be used to optimise answers for a specific mark in a real exam. Students do not gain any insight into secure processes, and teachers retain full responsibility for interpreting feedback and determining grades in any internal context.

Examiner AI supports learning and improves confidence with exam-style questions, but it does not confer an advantage in formally assessed work beyond the normal benefits of structured feedback.

Examiner AI is designed for answers written for qualifications taken by students in Year 10 and above only. In compliance with data protection regulations, users must be at least 13 years old to create an account. For students under this age, access must be managed via a teacher or parent account to ensure appropriate consent and oversight.

The marks and feedback provided by Examiner AI are for guidance purposes only and are not officially approved or endorsed by any exam boards. Results from the service should not be considered definitive or a substitute for formal assessment by a qualified examiner.

Examiner AI is committed to protecting user privacy. Users should not enter personal data, and the service will never request it. If personal data is inadvertently included within a submission, it is handled purely as unstructured text and is not used to identify any individual. All data is processed securely in accordance with relevant UK data protection regulations, and the service ensures appropriate safeguards are in place wherever processing occurs.

Examiner AI is currently in beta. While we strive to provide accurate, reliable and consistent marking, the AI is still being refined, and some results may vary. We encourage users to treat feedback as indicative and to use it alongside traditional methods of marking exam answers.

Examiner AI is designed to assess structured exam-style questions and may not fully capture the nuances of subjective or open-ended responses. For complex or creative questions, we recommend seeking additional feedback from a human examiner.

By using Examiner AI, users confirm that the content submitted is their own, or that appropriate permissions are in place for content authored by others. Submissions of plagiarised or copied material are against the terms of use and may affect the accuracy of the evaluation.

While Examiner AI aims to provide useful feedback to help improve exam performance, we cannot guarantee that using this service will result in specific exam outcomes. Users should continue with their broader revision and exam preparation strategies, and not substitute these with this service

Feedback provided by Examiner AI is generated by artificial intelligence and is based on adapted mark schemes and prompts. It may not reflect the full range of potential human judgement, especially in areas requiring complex reasoning or interpretation, nor does it provide feedback on all aspects of answers submitted.

No. Examiner AI does not use submitted content to train, fine-tune, or improve any third-party AI models. The service operates on a strict zero-retention basis with its AI partners. Submitted data is used only for the immediate purpose of generating feedback and is not stored by any external model provider.

Examiner AI is built on a privacy-first architecture. Users are strictly instructed not to submit personal data. However, should PII (Personally Identifiable Information) be inadvertently submitted, our processing protocols are designed to treat all inputs as anonymous. We comply with relevant data protection regulations (including UK GDPR) to ensure user data is protected.

You should only submit content that you are authorised to use. If you are submitting work on behalf of someone else, such as a student, you must ensure that you have the necessary permission and that your use of Examiner AI complies with your school or organisation’s data-protection policies. Users should avoid submitting personal data wherever possible.

Yes, submitted answers and AI-generated feedback are securely stored within Examiner AI systems so that users can revisit their evaluation history. Our AI partners do not store any content submitted through Examiner AI and do not use it for model training or retention. All data transmitted to AI partners is encrypted, processed statelessly, and used solely to generate the immediate response. Users can delete their stored submissions at any time, and all associated data is permanently removed when an account is deleted.

All data is processed and stored securely within a controlled infrastructure. Some processing may occur outside the UK/EU; where this is the case, we ensure that appropriate safeguards are in place, including UK GDPR-compliant transfer mechanisms. All third-party providers involved in processing are required to meet strict privacy, security, and zero-retention obligations.

We have established a feedback mechanism for reporting safety concerns and issues which are sent to a dedicated account (examinerai@tutor2u.net) and which are investigated and addressed promptly by our team.

Yes. We use a safety-first pre-processing layer. If a submitted answer contains profanity, hate speech, or content otherwise flagged as inappropriate or harmful, the system will block the submission and refuse to generate feedback. In such situations no data is submitted for AI processing.

Examiner AI does not sell, share, or transfer submissions to any third parties for marketing, analytics, or commercial reuse. Data is used exclusively for the purpose of generating feedback within Examiner AI or for internal quality assurance. All processing by AI model vendors is conducted under strict zero-retention, zero-training agreements.

No. Examiner AI provides formative feedback but does not make automated decisions with legal or significant effects under GDPR. Results do not determine grades, progression, or certification outcomes.

No. Examiner AI is deliberately designed to mark individual questions, not full or partial exam papers or scripts.

We believe that effective formative assessment requires a granular focus on specific skills (such as application, analysis or evaluation) within a single answer, rather than a broad summative judgment of a whole script. By refusing to process full papers, we ensure Examiner AI remains a support for learning, not a replacement for the holistic judgment and rigorous standardisation processes required to award a final grade.

No. We do not support bulk-upload features for mass marking. Examiner AI is designed to keep the human teacher (or student) in the loop. We believe that marking requires review and reflection on each piece of feedback. By processing answers one by one, we ensure that the teacher remains the gatekeeper of assessment, reviewing the rationale for every mark rather than relying on automated batch processing.

Marks are awarded based on how well your answer meets the demands of the question, covering relevant assessment skills such as knowledge & understanding, application, analysis and evaluation. Examiner AI handles questions that are points-based as well as those marked using levels of response.

The marks provided by tutor2u Examiner AI are designed to reflect the standards that a student might achieve in a real exam under exam conditions. Examiner AI uses bespoke marking promps based on our independent interpretation of public specifications The aim of these marking guidelines is to align as closely as possible with the marking criteria and standards of exams taken at the end of a course, helping students understand how they might perform in similar conditions.

No, the marks provided by Examiner AI are not officially approved or endorsed by exam boards. Examiner AI uses bespoke marking prompts designed solely for AI use, along with insights from tutor2u's subject experts, to generate feedback and marks. While we aim to align as closely as possible to real exams, the Examiner AI marks should be seen as indicative rather than definitive.

The exact mark scheme and prompt used is not displayed. However, tutor2u Examiner AI follows refined prompts and guidance for each qualification and question type, which are continuously updated for improved accuracy.

Examiner AI’s marking prompts are developed through a rigorous process led by our internal team of subject experts.

Rather than simply inputting an official mark scheme, our team analyses the public assessment requirements for each qualification based on Ofqual-approved public specifications. We utilise these documents strictly as they are intended: as public reference points to understand the required standard, rather than as content for reproduction.

Our team then codifies this logic into bespoke, proprietary instructions (prompts) designed specifically for an AI model. These prompts are iteratively tested and refined against a bank of internally-generated and stored standardised answers. This ensures Examiner AI delivers feedback that is consistent, accurate, and independent of official exam board documentation.

This depends on the question type. Where a question is marked by reference to levels of response, there is no breakdown of marks by assessment objective. However, for some other question types, Examiner AI provides a breakdown of your marks according to the key assessment objectives and the specific skills or knowledge being tested.

Yes, you can resubmit an answer after revising it. Each submission is treated as a new evaluation, so be sure to refresh the app before resubmitting.

Whilst Examiner AI primarily focuses on assessment skills, spelling and grammar errors may affect the evaluation if they hinder understanding. We recommend reviewing your answer for clarity before submission.

If you feel the level or mark assigned by Examiner AI doesn't reflect your response accurately, we encourage you to review the feedback provided to understand the rationale behind the assessment. As the service is still in beta, we welcome your feedback on any discrepancies. You can also resubmit your answer after making any adjustments based on the initial feedback to see if the evaluation changes.

Examiner AI uses a stateless architecture, meaning every answer submitted is processed in isolation to prevent the AI from becoming confused by previous interactions. Furthermore, we do not rely solely on the AI’s general knowledge. Instead, our proprietary marking prompts act as guardrails instructing the AI to only award marks if it finds specific evidence in the answer that matches the assessment criteria that we set. If the evidence is missing, no marks are awarded, or the no answer is marked.

No, the marks provided by Examiner AI are not more reliable than those awarded in actual exams. Actual exam marks are assigned by trained examiners following strict exam board guidelines, supported by rigorous standardisation processes. Examiner AI should be used as a tool to guide improvement rather than a substitute for official exam grading.

No. Examiner AI is powered by tutor2u’s proprietary marking logic. Examiner AI does not simply 'ingest' or reproduce copyrighted exam board mark schemes. Instead, we have codified our own senior examiner expertise into bespoke marking prompts that are based on our independent interpretation of the requirements of the qualification.

Examiner AI does not seek to present itself as an AI version of an exam board, nor to imply that AI can substitute for the deep expertise and rigour of exam board awarding processes. This approach ensures that Examiner AI operates independently of the intellectual property of exam boards, while providing high-quality feedback that supports student progress.

Examiner AI does not expose the underlying marking prompts or criteria used in the evaluation process. Feedback is generated at the level of skills and evidence rather than rule-by-rule replication. This prevents users from inferring any structured mark scheme from tutor2u’s proprietary logic, and certainly not any official documents produced by exam boards.

Examiner AI does not host, store, or reproduce a database of official exam board documents (such as PDF question papers or mark schemes).

Instead, we maintain a library of proprietary marking prompts developed by our own subject experts. These prompts codify the assessment logic required for specific question types but are independent creations, not copies of official documentation. When a user submits an answer, the system retrieves our bespoke prompt logic to determine the marking and feedback. No reference is made during the process to any exam board materials.

No. Examiner AI does not reproduce, display, or distribute any exam board documentation to users.

The feedback and rationale provided by the service are generative and unique to the specific answer submitted. While the feedback explains why a mark was awarded based on the assessment logic, it is drafted by the AI in real-time. It does not quote, copy, or reproduce the text of any official mark scheme or question paper.

No. Examiner AI should not be used to generate predicted grades for any qualification. Predicted grades require professional judgement informed by a wide range of evidence, including classwork, assessments, teacher observation, and knowledge of the student’s progress over time.

Examiner AI is a formative tool that provides feedback on individual answers. It:

• does not aggregate marks across multiple questions or papers;
• does not apply grade boundaries;
• does not model overall performance or progression;
• cannot account for the broader evidence normally used in predicting grades.

Any marks or feedback Examiner AI provides are indicative only and should not be interpreted as a reliable estimate of future exam performance. All predicted grades should continue to be determined solely by teachers and schools, using their established assessment policies and professional expertise.

No. Examiner AI is a formative learning tool, not a summative awarding tool. It assesses the specific skills demonstrated within a single answer. It does not aggregate scores across a whole paper or series of papers; nor does it apply grade boundaries. All grading and certification authority remains solely with the official exam boards. Examiner AI focuses entirely on helping students improve their answer technique and reducing teaching marking workload.

No. Examiner AI is designed as an anonymous service. The system does not request, require, or structurally store any official exam identifiers such as Candidate Numbers, Centre Numbers, or full legal names. We do not build user profiles based on exam identities. If a user inadvertently includes a candidate number or similar detail within the text of their answer, it is treated as unstructured text and is not used to identify the individual.

No. Examiner AI cannot access or infer secure exam materials. It operates solely on the information a user manually provides in a submission. There is no connection to any exam board systems, and Examiner AI has no access to non-public assessment materials. This ensures that the security of live or future assessments cannot be compromised.

No. Examiner AI does not attempt to replicate exam board marking in full, nor does it claim to represent the authoritative view of any awarding body. It provides formative feedback based on independent marking prompts. All official awarding decisions remain solely with the exam boards.

No. Examiner AI is not designed for formal reviews of marking, post-results enquiries, or appeals processes. It cannot, and is not intended to, replace the professional judgement of trained examiners working within exam board standardisation frameworks. Users seeking official re-marks or considering reviews of marking must follow the appropriate exam board procedures, and obtain appropriate professional advice where required.

We refuse to award grades to protect the integrity of the assessment. Awarding a grade requires the aggregation of marks across a full paper and the application of official grade boundaries, which are set by exam boards after the exam series and following standardisation. Examiner AI marks skills, not papers. We provide the marks and the detailed rationale to help you improve, but we leave the summative grading entirely to the official Awarding Organisations.

Examiner AI is an online service that uses generative artificial intelligence (AI) to automatically assess and mark exam-style answers. It assigns levels and marks based on relevant mark schemes and provides concise feedback and a rationale for the awarded level and mark.

Absolutely not. Examiner AI is designed to address the significant workload burden of marking, allowing teachers to focus on high-impact intervention and teaching. It helps tackle some of the heavy lifting of formative assessment, providing students with immediate feedback so they can arrive at class better prepared. We view AI as a supportive assistant to the teacher, not a replacement for their professional judgement and expertise.

Examiner AI compares a submitted answer and related question / stem with marking criteria for the selected qualification and question type. Examiner AI then assigns a grade or level, provides marks, and delivers feedback, briefly outlining the rationale behind the awarded score.

Generative AI is a type of artificial intelligence that creates new content or responses based on patterns and information it has been trained on. In the case of tutor2u Examiner AI, the system uses generative AI to interpret exam questions and supporting stems, evaluate answers, generate feedback, and assign marks based on relevant mark schemes and assessment criteria. Unlike traditional AI, which might only recognise patterns, generative AI produces new outputs, such as feedback, specifically tailored to each individual answer.

Examiner AI currently supports a small, but growing number of qualifications across different subjects and levels. You can select your qualification level and subject from a dropdown menu, and we will add further qualifications when we can.

Yes, the grading is tailored to each answer submitted. Since the AI evaluates every submission independently, marks and levels may occasionally vary across similar answers. Similarly, a level and mark might vary slightly for the same or substantially the same answer, particularly if an answer is positioned close to a different level in a mark scheme.

Examiner AI uses bespoke marking prompts based on bespoke assessment logic created by tutor2u’s subject specialists. While these prompts are extensively tested against a bank of standardised answers, accuracy can vary between responses, particularly where answers sit close to a boundary. Examiner AI is a formative tool designed to guide improvement and should always be used alongside human judgement rather than as a substitute for it.

Examiner AI offers the advantage of instant feedback and consistency in applying mark schemes, and it is available 24/7! However, it is not necessarily better than human examiners who bring experience, professional judgement, and an ability to interpret more complex or nuanced answers, particularly in subjective or creative questions.

Exam marking is a skill and human examiners are able to use their subjective judgement to support a given grade. AI is a helpful tool for students to gauge their performance and improve their answers, but it should complement, rather than replace, human feedback.

Examiner AI is designed and tested to promote fairness across diverse user groups. We test the system using a representative range of answers and continually review outputs for any patterns that may indicate bias. While we take active steps to reduce the risk of unfair outcomes, no automated system can guarantee the complete removal of bias. A feedback mechanism is available for users to report concerns, which are reviewed promptly by our team.

To submit your answer, simply input your exam question, indicate if there is a case study or additional context, write your response, and click ‘Mark answer’. The AI will process your answer and return feedback shortly.

Examiner AI currently only supports a limited range of subjects and qualifications.

Some question types for particular qualifications require answers to use a question stem, extract or case study etc. To ensure that the marking takes this into account please either upload a PDF of the extra information or paste it into the text box provided.

Once you click ‘Mark answer’, the AI will process the submitted question, stem and answer. Feedback typically takes 10-30 seconds, although more complex questions may take slightly longer.

Yes – you are able to see and review recently submitted evaluations. If you wish to remark an answer, please start afresh, as each evaluation is handled separately by Examiner AI.

Yes. We have an integrated tool that converts handwritten scripts into digital text. This conversion happens within our secure environment and adheres to the same strict zero-retention and privacy protocols as our marking engine. User confirmation is required before the converted digital text is submitted for marking.

Examiner AI is model agnostic. We use an orchestration layer that benchmarks multiple enterprise-grade AI models against a bank of standardised answers that we have developed and marked internally. Examiner AI routes different question types to the model that demonstrates the highest psychometric validity. This ensures users get the most accurate feedback available.

Yes - you need to register and sign-in to use Examiner AI.

Ensure you have a stable internet connection and try refreshing the page. If the problem persists, contact technical support for assistance.

Yes, Examiner AI is accessible from mobile devices, tablets, and desktop computers.

Credits are used to pay for evaluations using Examiner AI. Most question types are priced at 10 credits per evaluation.

You can view and download all your invoices from your billing page.