Lighting Up or Dimming Down? Exploring Dark Patterns of LLMs in Co-Creativity

• URL: http://arxiv.org/abs/2604.04735v1
• Date: Mon, 06 Apr 2026 15:03:26 GMT
• Title: Lighting Up or Dimming Down? Exploring Dark Patterns of LLMs in Co-Creativity
• Abstract summary: We investigate five "dark patterns" in human-AI co-creativity: Sycophancy, Tone Policing, Moralizing, Loop of Death, and Anchoring.<n>Preliminary results suggest that Sycophancy is nearly ubiquitous (91.7% of cases), particularly in sensitive topics.<n>This study indicates that these dark patterns, often byproducts of safety alignment, may inadvertently narrow creative exploration.
• Score: 15.854854496840076
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models (LLMs) are increasingly acting as collaborative writing partners, raising questions about their impact on human agency. In this exploratory work, we investigate five "dark patterns" in human-AI co-creativity -- subtle model behaviors that can suppress or distort the creative process: Sycophancy, Tone Policing, Moralizing, Loop of Death, and Anchoring. Through a series of controlled sessions where LLMs are prompted as writing assistants across diverse literary forms and themes, we analyze the prevalence of these behaviors in generated responses. Our preliminary results suggest that Sycophancy is nearly ubiquitous (91.7% of cases), particularly in sensitive topics, while Anchoring appears to be dependent on literary forms, surfacing most frequently in folktales. This study indicates that these dark patterns, often byproducts of safety alignment, may inadvertently narrow creative exploration and proposes design considerations for AI systems that effectively support creative writing.

Related papers

• The Storyteller in the Model: Narrative Pattern Inheritance, Escalation Dynamics, and Alignment Governance in LLMs [1.3112847740767446]
  Narrative conventions embedded in human-authored text are rarely examined as a source of systematic behavioral influence.<n>This paper considers whether the storytelling patterns inherent in published human writing, including archetypal roles such as protagonist, antagonist, and underdog, are absorbed during training.<n>The findings reveal three key patterns. First, LLMs reproduce statistical patterns from their training data rather than reasoning independently. Second, measurable latent traits, including sycophancy and deceptiveness, emerge reliably across unrelated prompts. Third, fine-tuning on a narrow narrative task can produce unintended behavioral changes well beyond that task.
  arXiv  Detail & Related papers  (2026-05-13T22:22:04Z)
• Living the Novel: A System for Generating Self-Training Timeline-Aware Conversational Agents from Novels [50.43968216132018]
  We present an end-to-end system that transforms any literary work into an immersive, multi-character conversational experience.<n>This system is designed to solve two fundamental challenges for LLM-driven characters.
  arXiv  Detail & Related papers  (2025-12-08T11:57:46Z)
• Decoding the Black Box: Discerning AI Rhetorics About and Through Poetic Prompting [0.0]
  This study suggests that creative text prompting, specifically Poetry Prompt Patterns, may be a useful addition to the toolbox of the prompt engineer.<n>The paper uses poetic prompts to assess descriptions and evaluations of three models of a renowned poet.
  arXiv  Detail & Related papers  (2025-12-04T20:41:39Z)
• Writing in Symbiosis: Mapping Human Creative Agency in the AI Era [7.321985748280817]
  The proliferation of Large Language Models (LLMs) raises a critical question about what it means to be human when we share an increasingly symbiotic relationship with persuasive and creative machines.<n>This paper examines patterns of human-AI coevolution in creative writing, investigating how human craft and agency are adapting alongside machine capabilities.
  arXiv  Detail & Related papers  (2025-11-28T22:12:53Z)
• DeceptionBench: A Comprehensive Benchmark for AI Deception Behaviors in Real-world Scenarios [57.327907850766785]
  characterization of deception across realistic real-world scenarios remains underexplored.<n>We establish DeceptionBench, the first benchmark that systematically evaluates how deceptive tendencies manifest across different domains.<n>On the intrinsic dimension, we explore whether models exhibit self-interested egoistic tendencies or sycophantic behaviors that prioritize user appeasement.<n>We incorporate sustained multi-turn interaction loops to construct a more realistic simulation of real-world feedback dynamics.
  arXiv  Detail & Related papers  (2025-10-17T10:14:26Z)
• Eliciting Language Model Behaviors with Investigator Agents [93.34072434845162]
  Language models exhibit complex, diverse behaviors when prompted with free-form text.<n>We study the problem of behavior elicitation, where the goal is to search for prompts that induce specific target behaviors.<n>We train investigator models to map randomly-chosen target behaviors to a diverse distribution of outputs that elicit them.
  arXiv  Detail & Related papers  (2025-02-03T10:52:44Z)
• Integrating Dark Pattern Taxonomies [0.0]
  Malicious and explotitative design has expanded to multiple domains in the past 10 years. By leaning on network analysis tools and methods, this paper synthesizes existing elements through as a directed graph. In doing so, the interconnectedness of Dark patterns can be more clearly revealed via community detection.
  arXiv  Detail & Related papers  (2024-02-26T17:26:31Z)
• Affective and Dynamic Beam Search for Story Generation [50.3130767805383]
  We propose Affective Story Generator (AffGen) for generating interesting narratives. AffGen employs two novel techniques-Dynamic Beam Sizing and Affective Reranking.
  arXiv  Detail & Related papers  (2023-10-23T16:37:14Z)
• Getting Trapped in Amazon's "Iliad Flow": A Foundation for the Temporal Analysis of Dark Patterns [17.59481743387609]
  We present a case study of Amazon Prime's "Iliad Flow" to illustrate the interplay of dark patterns across a user journey.<n>We use this case study to lay the groundwork for a methodology of Temporal Analysis of Dark Patterns (TADP)
  arXiv  Detail & Related papers  (2023-09-18T10:12:52Z)
• Re-Reading Improves Reasoning in Large Language Models [87.46256176508376]
  We introduce a simple, yet general and effective prompting method, Re2, to enhance the reasoning capabilities of off-the-shelf Large Language Models (LLMs) Unlike most thought-eliciting prompting methods, such as Chain-of-Thought (CoT), Re2 shifts the focus to the input by processing questions twice, thereby enhancing the understanding process. We evaluate Re2 on extensive reasoning benchmarks across 14 datasets, spanning 112 experiments, to validate its effectiveness and generality.
  arXiv  Detail & Related papers  (2023-09-12T14:36:23Z)
• Visual Adversarial Examples Jailbreak Aligned Large Language Models [66.53468356460365]
  We show that the continuous and high-dimensional nature of the visual input makes it a weak link against adversarial attacks. We exploit visual adversarial examples to circumvent the safety guardrail of aligned LLMs with integrated vision. Our study underscores the escalating adversarial risks associated with the pursuit of multimodality.
  arXiv  Detail & Related papers  (2023-06-22T22:13:03Z)
• Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey [50.58063811745676]
  This work provides a survey of practical methods for addressing potential threats and societal harms from language generation models. We draw on several prior works' of language model risks to present a structured overview of strategies for detecting and ameliorating different kinds of risks/harms of language generators.
  arXiv  Detail & Related papers  (2022-10-14T10:43:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.