[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86584-en":3,"doc-seo-86584-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86584,7971461741311,"Ophelia","https://ap-avatar.wpscdn.com/avatar/74000253aff267980c6?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779345379180704826",8,"Research & Report","Playful AI in Professional Email Tone and Recipient Engagement Field Experiment","Large language models (LLMs) are rapidly reshaping workplace communication, but how AI-assisted writing affects recipient behavior—and through what mechanism—remains unclear. A randomized crossover field experiment involved 121 employees across six companies sending emails in three conditions: unaided writing, GPT-5 playful rewriting, and GPT-5 professional rewriting. Among 16,880 emails, playful editing increased emotional positivity while professional editing decreased it; neither directly changed opens, replies, or response times. Instead, sender-level positivity predicted opening and replying via a significant indirect effect, indicating engagement changes through produced emotional tone rather than AI presence.","Playful AI in Professional Email: A Field Experiment on Tone and  \nRecipient Engagement  \nZiv Ben-Zion 1,2,* and Teddy Lazebnik 3,4,x,*  \n1 School of Public Health, Faculty of Social Welfare and Health Sciences, University of Haifa, Haifa, Israel  \n2 Department of Psychiatry, Yale School of Medicine, New Haven, CT, USA  \n3 Department of Information Systems, University of Haifa, Haifa, Israel  \n4 Department of Computing, Jönköping University, Jönköping, Sweden  \n*These authors contributed equally x Corresponding author: [teddy.lazebnik@ju.se](teddy.lazebnik@ju.se)  \nAbstract  \nLarge language models (LLMs) are rapidly reshaping workplace communication, yet whether AI-assisted writing changes how recipients actually behave, and through what channel, remains unknown. Here, in a randomized crossover field experiment, 121 employees across six companies sent work emails under three conditions over three weeks: unaided writing, GPT-5 rewriting in a playful tone, and GPT-5 rewriting in a professional tone. Across 16,880 emails, playful editing increased emotional positivity (B=+0.068, p\u003C0.001), and professional editing decreased it (B=-0.041, p\u003C0.001), yet neither condition directly altered open rates, reply rates, or response times. Instead, within-sender positivity strongly predicted both opening (OR=2.05) and replying (OR=3.32, p\u003C0.001), a significant indirect pathway through which AI editing shaped behavior, in the absence of any direct effect. These findings suggest that AI-assisted communication shapes workplace engagement not through its use, but through the emotional tone of the language it produces.  \nIntroduction  \nGenerative artificial intelligence (AI) has moved, in only a few years, from research curiosity to an everyday co-author of workplace communication. A majority of knowledge workers now use AI at work1 , and drafting and polishing written email messages is among its established everyday uses2. Because such tools measurably improve the quality and efficiency of written output 3,4 , organizations adopting them assume that a better-written message is a more effective one: that improving the tone or polish of an email will make recipients more likely to read it and reply. Nevertheless, this assumption has never been tested where it matters-in the volume and messiness of real organizational email, with the people who receive it.  \nWhether AI-assisted writing changes recipient behavior is, at its core, a question about mechanism. Decades of research in organizational behavior and social psychology show that the emotional texture of a message (e.g., its warmth, playfulness, positivity) strengthens rapport, reduces perceived social distance, and invites reciprocal engagement 5,6 . If Large language models (LLMs) editing affects recipients at all, it may do so not because a message is \"AIassisted\", but because it shifts these affective features of language. The distinction is not merely semantic. The AI-mediated communication framework holds that AI shapes interpersonal outcomes through the features of the messages it produces, not through its mere presence 7. Recent work has begun to trace such indirect pathways, for instance, showing that repeated human–AI interaction can amplify users' own biases without their awareness 8 and that AIgenerated language can out-persuade human language when personalized9.  \nLLMs are also a uniquely tractable tool for isolating tone from content. Due to the fact that they can rewrite a human draft while preserving its meaning, they allow the substance of a message to remain fixed while randomizing its stylistic register. This manipulation is difficult to achieve at scale with human writers, but it is a natural affordance of instruction-tuned models. The emotional sensitivity we and others have documented in LLMs' own outputs and behavior 10– 12 becomes an experimental instrument. Namely, a way to ask not only whether AI editing changes workplace engagement, but through what channel.  \nIn this","cbCaibwgvfFU3Obi","https://ap.wps.com/l/cbCaibwgvfFU3Obi","pdf",436826,3,1,18,"English","en",105,"# Abstract\n# Introduction\n# Results\n## Sample characteristics and descriptive patterns","[{\"question\":\"What were the three email writing conditions tested in the field experiment?\",\"answer\":\"Employees sent work emails under three conditions: unaided writing, GPT-5 rewriting in a playful tone, and GPT-5 rewriting in a professional tone.\"},{\"question\":\"Did playful or professional AI rewriting directly change recipient behavior such as opens and replies?\",\"answer\":\"No. Neither playful nor professional rewriting directly altered open rates, reply rates, or response times across the experiment.\"},{\"question\":\"Through what mechanism did AI editing influence engagement?\",\"answer\":\"AI editing affected engagement indirectly by changing the emotional positivity of the language, and positivity strongly predicted which emails were opened and answered.\"}]",1784212774,45,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"playful-ai-in-professional-email-tone-and-recipient-engagement-field-experiment","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/playful-ai-in-professional-email-tone-and-recipient-engagement-field-experiment/86584/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-25","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What were the three email writing conditions tested in the field experiment?","Question",{"text":75,"@type":76},"Employees sent work emails under three conditions: unaided writing, GPT-5 rewriting in a playful tone, and GPT-5 rewriting in a professional tone.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Did playful or professional AI rewriting directly change recipient behavior such as opens and replies?",{"text":80,"@type":76},"No. Neither playful nor professional rewriting directly altered open rates, reply rates, or response times across the experiment.",{"name":82,"@type":73,"acceptedAnswer":83},"Through what mechanism did AI editing influence engagement?",{"text":84,"@type":76},"AI editing affected engagement indirectly by changing the emotional positivity of the language, and positivity strongly predicted which emails were opened and answered.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]