[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-619380-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-619380-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","generative-ai-in-education-aqas-response-to-the-department-for-educations-call-for-evidence","Generative AI in education - AQA’s response to the Department for Education’s call for evidence","","Artificial Intelligence (AI) has the potential to improve education by reducing teacher workload and increasing marking reliability, but it can also produce wrong or misleading content, known as hallucinations, requiring close human supervision. AQA researchers trialled tools such as ChatGPT, GPT-4, LLaMA, and Alpaca on publicly available science papers and found AI may automate simple marking and support teachers with lesson planning. The document outlines benefits like faster feedback and quality assurance, while highlighting limits for high-stakes exams, including accuracy, bias, and closed-system concerns.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/exam/","Exam",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/generative-ai-in-education-aqas-response-to-the-department-for-educations-call-for-evidence/619380/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/generative-ai-in-education-aqas-response-to-the-department-for-educations-call-for-evidence/619380.png","ImageObject",300,407,{"name":42,"@type":43},"Ivy","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-06","2026-10-20",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",6,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What potential benefits does AQA believe generative AI can bring to education?","Question",{"text":63,"@type":64},"AQA highlights quicker feedback, faster content creation, and improved quality assurance. It also sees potential for automating simple marking and supporting teachers and awarding organisations with digital assessments.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"Why does AQA stress close human supervision for AI tools?",{"text":68,"@type":64},"AI tools can generate hallucinations—factually incorrect or misleading information. AQA also notes that, while AI may produce explanations, their effectiveness depends on factors like criteria specification and implicit knowledge, so human review remains necessary.",{"name":70,"@type":61,"acceptedAnswer":71},"What limitations did AQA find when testing generative AI for high-stakes exams?",{"text":72,"@type":64},"AQA found ChatGPT, LLaMA, and Alpaca were insufficiently accurate for high-stakes marking. GPT-4 was more promising for creating exam questions, but it still required significant human oversight and further investigation for classroom assessment support.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},619380,1793113568,{"code":4,"msg":82,"data":83},"success",[84,88,92,95,100,104,109,114,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":93,"slug":94},70,"exam",{"id":96,"doc_module":4,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},5,"Comic",60,"comic",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":101,"show_sort_weight":102,"slug":103},"Technology",50,"technology",{"id":105,"doc_module":4,"doc_module_name":25,"category_name":106,"show_sort_weight":107,"slug":108},7,"Healthcare",40,"healthcare",{"id":110,"doc_module":4,"doc_module_name":25,"category_name":111,"show_sort_weight":112,"slug":113},8,"Research & Report",30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":96,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":33,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":123,"language":139,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":12,"update_tm":143,"read_time":144,"preview_page_count":4},549758252649,"https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819","Generative AI in education  \nAQA’s response to the Department for Education’s call for evidence  \nAlex Scharaschkin  \nDirector for Assessment Research & Innovation  \nAugust 2023  \nExecutive summary  \nArtificial Intelligence (AI) has the power to transform education for the better – by cutting teacher workload and improving marking reliability.  \nWe have been exploring potential uses ofAI to improve assessment and education – we have found that there is potential for AI tools to make working life better for teachers and complement human marking. But AI tools can generate wrong or misleading information, which are known as hallucinations, and therefore need close human supervision to be effective.  \nAQA researchers trialled AI tools including ChatGPT, GPT4, LLaMA and Alpaca on a range of publicly available science papers. We found that, with appropriate regulation and human involvement, AI could be used to automate simple marking processes. This in turn could be used to help teachers plan their lessons for students.  \nIn the long run, AI could help teachers by:  \n• helping teachers create tasks and assessments for students  \n• assisting students with their own research  \n• giving students automated feedback  \n• supporting awarding organisations to create digital assessments.  \nThat said, we need to remember that AI tools are a bit like actors in Casualty: they can learn to use the language of doctors and sound like they have medical expertise, but they cannot perform an operation. They will always need close human supervision. More detailed answers to the questions posed by the DfE consultation are provided below.  \nHave you or your institution used generative AI in an educational setting? If so, could you briefly describe the ways it was used and the specific tools used.  \nAQA is an exam board and education charity that sets and marks more than half of all GCSEs and A levels taken in the UK every year. We reinvest our income to help young people fulfil their potential, in particular through cutting edge research into assessment.  \nWe have used generative AI tools as part of our programme of research in this area, and have been exploring these issues for some time.  \nAQA has found that, in the long term, there is real potential for AI tools to make working life better for teachers and complement human marking, and that there are many potential beneficial uses for AI. However, we have also found that, at present, AI tools still need some refinement.  \nThe main potential benefits ofAI as we see them are:  \n• quicker feedback  \n• quicker content creation  \n• improving quality assurance  \nAI tools could be used to automate simple marking processes to provide quicker feedback, as well as help teachers create informal assessments in class, such as quizzes. This, in turn, would save time for both teachers and awarding organisations.  \nIn addition, currently quality assurance processes for marking high-stakes examinations involve periodically giving work to more than one examiner to make sure that they come out with similar marks. But AI tools could be used to check marks hundreds of times to improve marking reliability.  \nHowever, the current limitations of AI tools mean that they are unsuitable on their own to mark or create questions for high stakes exams.  \nWe used ChatGPT, LLaMA and Alpaca to test automated marking capabilities on a range of publicly available science papers not connected to AQA. We found that they were insufficiently accurate for high stake exams.  \nWe also used GPT 4 for a proof-of-concept exploration of how it could be used to create exam questions. This was more promising, but it could not currently be used without significant human oversight for high stakes exams. It could be used to support assessment in the classroom, but this requires further investigation.  \nAQA has also trialled using ChatGPT to create curriculum summary presentations. These have still needed a lot of work, but there is potential here.  \n","cbCaimgAbPGMkOCF","https://ap.wps.com/l/cbCaimgAbPGMkOCF","pdf",167236,"English","# Executive summary\n## Potential benefits and long-term role of AI\n## Trials and proof-of-concept findings\n## Key challenges and mitigation needs","[{\"question\":\"What potential benefits does AQA believe generative AI can bring to education?\",\"answer\":\"AQA highlights quicker feedback, faster content creation, and improved quality assurance. It also sees potential for automating simple marking and supporting teachers and awarding organisations with digital assessments.\"},{\"question\":\"Why does AQA stress close human supervision for AI tools?\",\"answer\":\"AI tools can generate hallucinations—factually incorrect or misleading information. AQA also notes that, while AI may produce explanations, their effectiveness depends on factors like criteria specification and implicit knowledge, so human review remains necessary.\"},{\"question\":\"What limitations did AQA find when testing generative AI for high-stakes exams?\",\"answer\":\"AQA found ChatGPT, LLaMA, and Alpaca were insufficiently accurate for high-stakes marking. GPT-4 was more promising for creating exam questions, but it still required significant human oversight and further investigation for classroom assessment support.\"}]","Generative AI in education - AQA’s response to the Department for Education’s call for evidence | PDF",1792462863,25]