The Art of Conversational Authoring: How AI Interaction Mirrors the Craft of Fiction Writing

A Guest Post by Mike Kentz of AI Literacy Partners

I'm excited to share an important contribution from my friend and co-author Mike Kentz, who brings a fascinating perspective on how traditional literacy skills are finding powerful new applications in our AI-rich world. Mike's article explores the striking parallels between effective AI interaction and the craft of fiction writing—revealing how skills like character development, scene setting, and narrative construction are becoming essential tools for meaningful human-AI collaboration.

This insight builds beautifully on the work we explored together in our November 2024 book, AI in Education: A Roadmap For Teacher-Led Transformation, where we emphasized how educators can leverage their existing expertise to architect learning in the AI age. Mike continues this teacher-first approach through his consultancy, AI Literacy Partners, where he helps educators, business leaders, and parents navigate AI integration with confidence and purpose.

As we're increasingly discovering, the future of AI literacy isn't about learning entirely new skills—it's about recognizing how our deep human capacities for storytelling, communication, and creative thinking provide the foundation for thriving in an AI-driven world.

Nick Potkalitsky, Ph.D.

---

“The Art of Conversational Authoring: How AI Interaction Mirrors the Craft of Fiction Writing”

Claude.ai was used in the construction of this article for research and revision

There has been much debate surrounding the validity and relevance of the term "prompt engineering" as a skill or even an accurate portrayal of what it means to engage effectively with AI. Critics increasingly argue that what occurs in human-AI interaction cannot be accurately described as "engineering" – it is fundamentally about collaborating, conversing, and communicating in ways that leverage creative thinking.

Behind these criticisms lies an important realization: working well with AI is primarily a communication and writing skill, complemented by reading skills when analyzing AI outputs through a critical lens. For educators interested in embedding AI literacy into their practices or designing instruction that teaches students to use AI in ways that expand rather than replace creativity, this recognition opens new pedagogical possibilities.

The practice of conversing with artificial intelligence aligns remarkably with the fundamental craft elements of fiction writing. When users engage with AI systems – particularly when developing specific personas or guiding conversational tone – they unconsciously employ the same creative techniques that novelists have refined over centuries: character development, environmental crafting, scene construction, contextual framing, and narrative perspective management.

The process of "conversational authoring" with AI bears striking resemblance to what literary scholars identify as the core elements of fictional craft. Understanding this parallel offers educators both a theoretical framework for comprehending AI interaction and practical strategies for teaching these essential 21st-century skills.

This analysis examines three key parallels between AI interaction and fiction writing:

• AI Character Design through role prompting – establishing personas with specific expertise, motivations, and approaches

• Scene Setting through context prompting – creating environmental and situational frameworks that guide responses

• Narrative Construction through step-by-step problem solving – building collaborative reasoning that mirrors plot development

Each parallel reveals how effective AI interaction draws on sophisticated creative writing skills while offering concrete strategies for classroom implementation. Perhaps AI interaction represents a new form of creative writing—one where the "story" emerges through collaborative dialogue rather than individual authorship.

---

AI Character Design: Architecting Digital Personas Through Role Prompting

Character development is the process of creating fictional characters with the same depth and complexity as real-life human beings, requiring writers to construct backstories, motivations, and consistent behavioral patterns. Similarly, when users work to establish an AI persona - whether requesting a formal academic tone, a creative collaborator, or a specific professional role - they engage in parallel character construction through what academic research identifies as "role prompting" or "persona prompting."

Role prompting elicits the model by asking it to take on a particular role or persona before answering a question, establishing what researchers call a "memetic proxy" that allows the AI to embody specific characteristics and expertise. This technique mirrors how fiction writers develop what literary theorists recognize as "round characters" – complex, dimensional, and well-developed personalities.

The academic literature on character development emphasizes giving them depth, relatability, and flaws, and to provide specific details and backstories that bring those characters to life. In AI interactions, users achieve this through role prompting strategies. A wide range of AI researchers have found that providing the AI explicit instructions that go step-by-step through what you want works particularly well when combined with role establishment. When someone prompts an AI with "You are a Renaissance art historian with expertise in Venetian painting. Analyze this artwork considering the cultural context of 16th century Venice," they are employing role prompting to construct a sophisticated digital persona.

Motivations are the driving force behind a character's actions and decisions, making them essential in character development. In AI conversations, effective users similarly establish motivational frameworks through role prompts, defining not just what they want the AI to know, but how they want it to approach problems, what priorities to consider, and what goals to pursue in the response.

Consider this example from a recent exchange with Claude, where I sought assistance in revising a fiction novel:

Most AI interaction guides focus on surface-level role assignment, urging users to name a function and immediately proceed to task specification. They emphasize the user's motivations and goals as context. However, if we understand AI interaction as genuine character development, why would we not provide the persona with intrinsic motivations as well?

While AI output quality remains subjective – what proves useful to one user may feel inadequate to another – embedding motivational frameworks consistently affects conversational tone. In this example, providing the AI with specific pedagogical values and teaching philosophy created what I experienced as a "healthier interaction" with reduced sycophancy. For creative and educational contexts, this tonal shift toward authentic collaboration rather than mere compliance represents a significant advantage.

Caption: It is difficult to share full screenshots in these cases, as the outputs often give away plot points.

The iterative nature of AI persona development also mirrors what fiction writers call character arcs. Represents the transformation or growth a character undergoes over the course of the story. Through successive interactions, users refine and develop AI personas (and outputs), adjusting tone, expertise level, and response patterns – thus guiding a character's evolution throughout the "narrative" of their extended conversation.

---

Scene Setting: Constructing Digital Contexts Through Context Prompting

Fiction writing relies heavily on what scholars term "environmental storytelling" – the art of using the environment, the experiences they have, their expectations, the various clues and subtleties of their surroundings to convey meaning and atmosphere. In AI interactions, users engage in analogous environmental construction through "context prompting" – a technique that provides relevant background information to guide the AI's understanding and response generation.

Academic research on prompt engineering emphasizes that contextual prompts are foundational to effective AI interaction. Context prompting involves giving the AI specific details within your prompt to help it understand exactly what you're looking for, leading to more specific and relevant responses. Setting is where and when the story takes place. It includes the following: The immediate surroundings of the characters such as props in a scene. When users provide contextual information to AI – describing their project requirements, academic level, professional constraints, or creative goals – they are essentially constructing the "setting" within which the AI persona will operate.

The concept of environmental storytelling extends beyond physical space to encompass a conceptual place, in this case, a make-believe area. AI users create these conceptual environments through context prompting strategies, establishing not just factual parameters but emotional and intellectual atmospheres. Research shows that prompts which include clear, explicit instructions and contextual information significantly enhance model outputs in terms of accuracy, coherence, and relevance. A prompt requesting "Write a gentle, encouraging response to help a struggling undergraduate student understand calculus concepts for their upcoming exam" constructs a very different environmental context than one asking for "Provide a rigorous mathematical proof suitable for a graduate-level analysis course." (Notice that Claude immediately builds a technical artifact rather than adopting a coaching approach - as it did in the first interaction.)

Often times, the setting can reveal something about the main character as he/she functions in that place and time period. Similarly, the contextual environments users create for AI interactions both shape and reveal the "character" they are constructing, influencing how the AI persona will manifest its expertise and approach problems. Academic studies have found that adding relevant context at the beginning or the end of a prompt improves the performance of large language models, while placing context in the middle leads to worse performance.

This mirrors effective storytelling principles: just as readers need environmental context early in a narrative to understand character actions, AI systems require upfront contextual information to generate appropriately situated responses.

You're absolutely right - I shifted to a much more casual, blog-style tone. Let me revise to match the academic register of the rest of the piece:

---

Narrative Construction: Building Scenes Through Step-by-Step Problem Solving

The craft of fiction requires careful attention to scene construction – the immediate circumstances, conflicts, and objectives that drive individual narrative moments. Plot refers to the events that happen within the story, including every major turning point that characters experience. When AI users tackle complex problems, they often engage in analogous behavior through sequential, step-by-step problem solving that mirrors how scenes unfold in narrative fiction.

This approach, sometimes termed Chain-of-Thought (CoT) prompting, involves decomposing complex requests into logical progressions. While academic research on CoT's effectiveness shows mixed results across different tasks and contexts, the underlying human behavior it represents – collaborative step-by-step reasoning – mirrors fundamental principles of scene construction in storytelling.

Fiction writers develop scenes by establishing circumstances, introducing complications, building through rising action, and progressing toward resolution. AI users employ parallel structures when confronting complex intellectual challenges. Rather than posing a simple query such as "Should our school ban phones?" (see chat) a user practicing narrative construction might structure their prompt as: "First, analyze the primary arguments supporting phone restrictions in educational settings. Subsequently, examine counterarguments and alternative perspectives. Then, evaluate empirical research on mobile device usage and learning outcomes. Finally, synthesize these elements into an evidence-based recommendation." (see chat)

This structural approach creates what constitutes a narrative arc for collaborative reasoning – establishing the scene (the policy question), introducing complexity (multiple perspectives), developing tension through conflicting evidence, and progressing toward resolution through reasoned synthesis. Students who learn to recognize when complex problems require narrative construction versus direct inquiry develop sophisticated analytical skills. Multi-faceted research questions and analytical tasks benefit from structured collaborative exploration, while straightforward factual queries may not require such elaborate frameworks.

The pedagogical value emerges not from any particular prompting technique's technical superiority, but from how problem decomposition mirrors the fundamental human activity of constructing coherent narratives. When students practice decomposing complex assignments into sequential reasoning conversations, they develop essential skills in logical sequencing, collaborative reasoning, and intellectual scene-building – engaging in the essentially narrative act of structuring complex thinking around systematic exploration of ideas.

---

Implications for Human-AI Interaction Design

This parallel between AI interaction and fiction craft suggests several important implications for how we understand and optimize human-AI communication. The unconscious application of literary techniques in AI prompting indicates that humans naturally employ narrative frameworks to structure meaningful interaction with artificial agents.

Understanding AI interaction through the lens of fiction craft may help users develop more sophisticated conversational authoring strategies. Just as creative writing can be taught through attention to character, environmental crafting, scene construction, and perspective, AI interaction skills might benefit from explicit attention to these same elements.

Academic research suggests that these conversational authoring techniques can be combined and scaled based on task complexity. Simple interactions might require only role prompting to establish character, while complex problem-solving scenarios might benefit from combining all four techniques: establishing a role, providing context, structuring the reasoning process through chain-of-thought, and demonstrating desired perspective through few-shot examples.

---

Implications for Educators: Teaching Conversational Authoring as Creative Writing

For educators grappling with AI's role in the classroom, understanding conversational authoring through the lens of fiction craft offers both pedagogical opportunities and practical solutions to persistent challenges.

In my classroom, safe and effective AI use was taught through the pedagogical methods associated with early-stage literacy and writing instruction - precisely because of the clear links between AI and writing, creative writing, and more. You can read more about the approach in the foreword of my exhibit at The WAC Repository.

---

Conclusion: The Narrative Nature of Intelligence

However, the point remains that the parallels between AI interaction and fiction craft reveal something fundamental about human intelligence and communication. We appear to be naturally narrative beings who structure meaning through character, environment, scene, and perspective – even when interacting with artificial systems. This suggests that as AI technology continues to develop, the most effective interfaces may be those that acknowledge and leverage these deep-seated narrative patterns.

The art of conversing with AI, like the craft of writing fiction, requires attention to character consistency, environmental context, scene-specific objectives, and perspective management. Recognition of these parallels may help both users and developers create more effective, engaging, and meaningful human-AI interactions. In essence, every conversation with an AI becomes a collaborative act of storytelling—one where both human and artificial agents participate in the ongoing construction of meaning, context, and relationship through conversational authoring.

---

Check out some of our favorite Substacks:

Mike Kentz’s AI EduPathways: Insights from one of our most insightful, creative, and eloquent AI educators in the business!!!

Terry Underwood’s Learning to Read, Reading to Learn: The most penetrating investigation of the intersections between compositional theory, literacy studies, and AI on the internet!!!

Suzi’s When Life Gives You AI: A cutting-edge exploration of the intersection among computer science, neuroscience, and philosophy

Alejandro Piad Morffis’s The Computerist Journal: Unmatched investigations into coding, machine learning, computational theory, and practical AI applications

Michael Woudenberg’s Polymathic Being: Polymathic wisdom brought to you every Sunday morning with your first cup of coffee

Rob Nelson’s AI Log: Incredibly deep and insightful essay about AI’s impact on higher ed, society, and culture.

Michael Spencer’s AI Supremacy: The most comprehensive and current analysis of AI news and trends, featuring numerous intriguing guest posts

Daniel Bashir’s The Gradient Podcast: The top interviews with leading AI experts, researchers, developers, and linguists.

Daniel Nest’s Why Try AI?: The most amazing updates on AI tools and techniques

Jason Gulya’s The AI Edventure: An important exploration of cutting-edge innovations in AI-responsive curriculum and pedagogy.