[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-578790-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-578790-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","from-words-to-pixels-artificial-intelligence-struggles-with-world-englishes","From words to pixels: Artificial intelligence struggles with world Englishes","","This study examines how DALL-E 3 interprets English descriptions written by Indonesian university students. Sixteen descriptive texts were submitted to the AI tool and the resulting images were compared to original photos, revealing frequent mismatches. Misinterpretations were traced to grammatical and vocabulary patterns associated with Indonesian English alongside broader stylistic choices using vague, emotional, or abstract language. High concrete detail often reduced negative effects of non-standard grammar, highlighting limits in current tools’ ability to fairly process linguistic and stylistic diversity. The work proposes “digital intelligibility” and recommends inclusive training data and open classroom discussion on AI bias.",{"@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/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/from-words-to-pixels-artificial-intelligence-struggles-with-world-englishes/578790/",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/from-words-to-pixels-artificial-intelligence-struggles-with-world-englishes/578790.png","ImageObject",300,407,{"name":42,"@type":43},"Aria","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-15","2026-10-15",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 does the study investigate about DALL-E 3?","Question",{"text":63,"@type":64},"It investigates how DALL-E 3 interprets English descriptions written by Indonesian university students and whether the generated visuals match the original references.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What were the main causes of misinterpretation in the generated images?",{"text":68,"@type":64},"Misinterpretations were linked to grammatical and vocabulary patterns reflecting Indonesian English as well as stylistic choices like vague, emotional, or abstract language.",{"name":70,"@type":61,"acceptedAnswer":71},"How does the study suggest improving inclusive AI use in education?",{"text":72,"@type":64},"It adapts “intelligibility” into “digital intelligibility,” recommends improving training data, and encourages classroom space for open discussions about AI bias toward language diversity and stylistic choices.","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},578790,1792166366,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,105,110,114,119,122,126],{"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":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},"Technology",50,"technology",{"id":106,"doc_module":4,"doc_module_name":25,"category_name":107,"show_sort_weight":108,"slug":109},7,"Healthcare",40,"healthcare",{"id":111,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":112,"slug":113},8,30,"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":97,"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":111,"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":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":145,"preview_page_count":4},2336464648322,"https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488","The JALT CALL Journal  \nISSN 1832-4215, 21(3), 103083 (2025)  \n[https://doi.org/10.29140/jaltcall.v21n3.103083](https://doi.org/10.29140/jaltcall.v21n3.103083)[ ](https://doi.org/10.29140/jaltcall.v21n3.103083)©2025 Flora Debora Floris  \nFrom words to pixels: Artificial intelligence struggles with world Englishes  \nFlora Debora Floris  \n[debora@petra.ac.id](debora@petra.ac.id)  \nPetra Christian University, Indonesia  \nThis study examines how DALL-E 3 interprets English descriptions written by Indonesian university students. Sixteen descriptive texts were submitted to the artificial intelligence (AI) tool, and the resulting images were compared to original photos. Most outputs showed clear mismatches. The analysis found that misinterpretations originated from two main sources: grammatical and vocabulary patterns reflecting Indonesian English and broader stylistic choices, such as the use of vague, emotional, or abstract language. The study also found that a high level of concrete detail could often mitigate the negative effects of non-standard grammar. The findings suggest that current AI tools are not yet equipped to fairly process the full range of human linguistic variation, from local English features to the stylistic patterns of human-centric writing. To support more inclusive use of AI in education, this study adapts the established concept of intelligibility into the idea of “digital intelligibility,” and recommends improving training data and creating classroom space for open discussions about AI bias toward language diversity and stylistic choices.  \nKeywords: World Englishes, language variation, AI bias, text-to-image generator, intelligibility  \nIntroduction  \nThe rapid development of artificial intelligence (AI) has changed many aspects of people’s daily life. One important development is the rise of generative AI systems that can turn written text into images. A well-known example is OpenAI’s DALL-E 3, which creates detailed images based on text prompts. While the system appears to be highly advanced, its output depends heavily on how closely the user’s input matches the language patterns found in its training data.  \nAs these tools become more widely used in education, concerns have emerged about their ability to understand the many different English varieties used around the world. From a sociolinguistic standpoint, these varieties are part of what scholars call World Englishes or localized forms that reflect  \nThis work is licensed under a Creative Commons Attribution  \n4.0 International License.  \n1  \nthe cultural, social, and linguistic backgrounds of non-native speakers (Kachru et al., 2009) . One such emerging variety is Indonesian English, which shows distinct patterns in vocabulary, grammar, and pronunciation shaped by local linguistic patterns. However, most AI models are trained mainly on “standard”English, such as American English (Payne et al., 2024), which may limit their ability to process localized inputs effectively.  \nSome existing studies (e.g. Bailey et al., 2025; Jackson et al., 2024) have confirmed that AI systems often struggle with non-standard English varieties. However, despite these findings, little attention has been given to how such AI systems respond to non-standard Englishes in real classroom contexts, especially in multilingual settings where students regularly use localized forms of English.  \nMany studies in English Language Teaching (ELT) focus on the practical pedagogical use of AI in classrooms, for example, as tools for writing support, conversational practice, and language learning enhancement. Some recent studies, as listed in Floris (2025), have examined both the potential benefits and limitations of implementing such tools in ELT contexts. However, these studies do not address how well AI tools understand the kinds of English students actually use.  \nTo address this gap, the present study explores how DALL-E 3 interprets English descriptions written by Indonesian universit","cbCainGDVH8d9whK","https://ap.wps.com/l/cbCainGDVH8d9whK","pdf",604309,28,"English","# Introduction\n## Literature review\n### World Englishes","[{\"question\":\"What does the study investigate about DALL-E 3?\",\"answer\":\"It investigates how DALL-E 3 interprets English descriptions written by Indonesian university students and whether the generated visuals match the original references.\"},{\"question\":\"What were the main causes of misinterpretation in the generated images?\",\"answer\":\"Misinterpretations were linked to grammatical and vocabulary patterns reflecting Indonesian English as well as stylistic choices like vague, emotional, or abstract language.\"},{\"question\":\"How does the study suggest improving inclusive AI use in education?\",\"answer\":\"It adapts “intelligibility” into “digital intelligibility,” recommends improving training data, and encourages classroom space for open discussions about AI bias toward language diversity and stylistic choices.\"}]","From words to pixels: Artificial intelligence struggles with world Englishes | PDF",1792034412,71]