Notes on Puntoni et al. (2021) – Consumers and Artificial Intelligence

Paper: “Consumers and Artificial Intelligence: An Experiential Perspective,” Journal of Marketing, 85 (1), 131–51.

Main Topic or Phenomenon

This paper addresses consumer experiences with artificial intelligence (AI) systems across various touchpoints in their daily lives. The authors examine how consumers interact with AI-enabled products and services, moving beyond technical evaluations of AI efficiency to understand the social and psychological implications of these interactions.

Theoretical Construct

The paper develops a framework centered on four distinct consumer AI experiences:

1. Data Capture Experience: The experience of providing personal data to AI systems, either intentionally (sharing) or unintentionally (surrendering), including data collected from behavioral shadows.
2. Classification Experience: The experience of receiving AI-generated personalized predictions based on perceived categorization as a certain type of consumer.
3. Delegation Experience: The experience of involving AI in production processes to perform tasks that consumers would otherwise do themselves (decisions, digital actions, physical actions).
4. Social Experience: The experience of reciprocal communication with AI, either knowing it’s AI (like Siri) or initially unaware (like chatbots).

Each experience is characterized by psychological tensions between positive and negative outcomes:

• Data capture: served vs. exploited
• Classification: understood vs. misunderstood
• Delegation: empowered vs. replaced
• Social: connected vs. alienated

Key Findings

1. Dual Nature of AI Experiences: Each AI experience type creates both benefits and costs for consumers, leading to psychological tensions that must be managed.
2. Personal Control as Central Concern: Loss of personal control emerges as a fundamental issue across data capture and delegation experiences, leading to negative affect, moral outrage, and psychological reactance.
3. Identity Concerns in Classification and Social Experiences: These experiences particularly affect consumer self-identity through inaccurate categorizations or perpetuation of harmful stereotypes.
4. Sociological Context Matters: Popular narratives (surveillance society, unequal worlds, transhumanism, humanized AI) shape how consumers interpret their AI experiences.
5. Individual Differences Affect Reactions: Factors like socioeconomic status, cultural norms, education, and group membership influence how consumers experience AI interactions.

Boundary Conditions and Moderators

Several moderators influence the main effects:

Individual Differences:

• Socioeconomic status affects feelings of exploitation
• Cultural norms influence control preferences
• Awareness of discrimination affects misunderstanding perceptions
• Uniqueness vs. belonging motives affect classification reactions

Contextual Factors:

• Physical context of data collection
• Device type used for interaction
• Frequency of data capture (intermittent vs. continuous)
• Nature of tasks (subjective vs. objective, identity-relevant vs. not)
• Timing of AI disclosure

Task and Content Characteristics:

• Type of data collected (environmental vs. personal)
• Perceived “humanness” of activities being automated
• Instrumental vs. symbolic consumption contexts
• Focus on process vs. outcome

Building on Previous Work

The paper extends existing literature by:

1. Integrating Multiple Perspectives: Combines computer science efficiency focus with sociological and psychological insights about consumer experiences.
2. Moving Beyond Binary Views: Challenges simple positive/negative evaluations of AI by revealing simultaneous benefits and costs.
3. Connecting Micro and Macro Levels: Links individual psychological processes with broader sociological narratives about AI in society.
4. Experiential Focus: Shifts attention from technical capabilities to consumer subjective experiences, building on customer experience literature.
5. Comprehensive Framework: Provides systematic organization of diverse AI consumer interactions under four experience types.

Major Theoretical Contribution

The primary theoretical contribution is the development of a comprehensive experiential framework that:

1. Reveals Hidden Costs: Exposes psychological and social costs of AI that efficiency-focused approaches miss.
2. Identifies Fundamental Tensions: Shows that AI experiences are inherently conflicted, not simply positive or negative.
3. Bridges Disciplines: Integrates insights from psychology, sociology, and marketing to understand AI adoption and resistance.
4. Provides Organizing Structure: Offers a systematic way to categorize and understand diverse consumer-AI interactions.
5. Highlights Control and Identity: Identifies personal control and self-identity as key theoretical constructs for understanding consumer AI experiences.

Major Managerial Implications

Organizational Learning:

• Conduct empathetic listening to understand consumer exploitation and alienation
• Collaborate with diverse experts (psychologists, sociologists, activists)
• Diversify hiring to include marginalized groups in AI design
• Sponsor research on AI’s impact on vulnerable populations

Experience Design:

• Implement personalized privacy defaults using AI itself
• Provide easy transitions from AI to human representatives
• Design “bubble-bursting” features to avoid filter bubbles
• Create less anthropomorphic rather than more human-like AI
• Allow consumer validation of AI inferences
• Preserve meaningful human control in delegation experiences

Industry-Level Action:

• Develop algorithm bills of rights
• Share organizational learning across firms
• Create ethical guidelines specific to marketing applications

Unexplored Theoretical Factors

Several potential moderators and theoretical factors remain unexplored:

Individual Differences:

• Personality traits (need for control, openness to experience)
• Technology anxiety or technophobia
• Prior negative experiences with technology
• Cognitive load and decision fatigue

Relational Factors:

• Trust in the brand/company providing AI
• Perceived similarity between consumer and AI
• Social proof from others’ AI usage
• Relationship length with the service provider

Temporal Dynamics:

• Adaptation effects over time
• Expectations vs. reality gaps
• Learning curves with AI systems
• Generational differences in AI acceptance

Contextual Moderators:

• Social presence of others during AI interaction
• Economic pressure or necessity to use AI
• Alternative availability (human vs. AI options)
• Perceived reversibility of AI decisions

Motivational Factors:

• Goal orientation (performance vs. mastery)
• Regulatory focus (promotion vs. prevention)
• Need for cognition
• Desire for efficiency vs. experiential consumption

Reference

Puntoni, Stefano, Rebecca Walker Reczek, Markus Giesler, and Simona Botti (2021), “Consumers and Artificial Intelligence: An Experiential Perspective,” Journal of Marketing, 85 (1), 131–51.