Emerging Consumer Futures

Introduction

Consumer behaviour is entering a period in which the boundaries between human agency, algorithmic assistance, physical products and digital environments are becoming increasingly difficult to locate. Consumers now encounter brands through conversational agents, recommendation systems, synthetic research participants, connected products, virtual spaces and subscription-based access models. These developments do not merely create new communication channels. They can alter how consumers form preferences, understand identity, allocate trust, make decisions and define ownership.

For brand managers, the central question is not whether a particular technology will become popular. It is how technological systems will reorganise consumer–brand relationships. A generative AI assistant may select products on behalf of a consumer; a synthetic consumer may influence product development; a voice interface may remove visual brand cues; an immersive environment may make brand experience spatial and social; and a connected product may continue communicating with its owner after purchase. Each development introduces value, but also risks involving surveillance, manipulation, exclusion, bias, dependency and loss of human control.

This chapter evaluates these developments through the lenses of consumer psychology, marketing strategy and governance. It also introduces scenario planning as a disciplined method for anticipating alternative consumer futures.

Learning Objectives

By the end of this chapter, students should be able to:

1. Distinguish assistive, agentic and generative AI in consumer contexts.
2. Explain the uses and limitations of synthetic consumers.
3. Analyse immersive environments, voice interfaces and connected products as psychologically significant touchpoints.
4. Evaluate evolving ownership models, including access, subscription and product-as-a-service models.
5. Assess implications for trust, identity, decision making and brand relationships.
6. Apply scenario planning to emerging consumer futures.
7. Recommend governance and brand-management practices that preserve consumer agency.

1. From Digital Consumers to Distributed Agency

Traditional consumer models often assume that an individual recognises a need, searches for information, evaluates alternatives, chooses, consumes and evaluates an offering. Contemporary technology distributes these activities across people, platforms, algorithms, devices and communities.

A consumer may still express a preference, but an AI system may formulate the search query, filter alternatives, negotiate price, complete payment and monitor post-purchase performance. Agency therefore becomes distributed rather than purely individual. Distributed agency does not mean that consumers have no control. It means that control is shared with technical systems whose objectives, data and decision rules may not be fully visible.

Three changes are particularly important:

• Mediation: technologies increasingly filter what consumers see and consider.
• Delegation: consumers increasingly authorise systems to perform decisions or actions.
• Continuity: products and services increasingly collect data and interact after the transaction.

These changes affect the psychological processes of attention, learning, trust, perceived control and identity construction. A brand is no longer encountered only in an advertisement, store or package. It may be encountered as a recommendation, a voice, a default setting, a notification or an automated action.

2. Agentic and Generative AI in Consumer Behaviour

2.1 Generative AI

Generative AI produces text, images, audio, video, code or other content in response to prompts or contextual data. In marketing, it can create personalised messages, product descriptions, virtual try-on content, service responses and simulated experiences.

Its consumer-behaviour significance lies in the reduction of search and production costs. Consumers can ask for comparisons in natural language, request explanations adapted to their knowledge level, or generate alternatives that previously required specialist skills. At the same time, generated content can be inaccurate, stereotypical, overly persuasive or difficult to distinguish from human-created communication.

Generative AI changes the information environment in four ways:

1. Content abundance: consumers face more possible messages and options.
2. Personalisation: content can be tailored to inferred preferences and vulnerabilities.
3. Synthetic sociality: consumers may interact with apparently empathetic systems.
4. Provenance uncertainty: consumers may not know who created, verified or sponsored content.

2.2 Agentic AI

An agentic AI system goes beyond generating content. It interprets a goal, plans actions, uses tools, evaluates results and may act with limited human intervention. For example, an agent could monitor prices, select a flight, apply loyalty benefits and complete a booking within user-defined constraints.

Agentic systems introduce a distinction between decision support and decision delegation. Decision support assists a person who remains the effective decision maker. Decision delegation permits a system to choose or act on the person’s behalf. The latter may improve convenience but increases the importance of authorisation, reversibility and accountability.

A useful managerial framework is the delegation ladder:

The key issue is not simply automation. It is whether the consumer understands what the system is doing, can intervene, and can obtain correction when an outcome is harmful.

2.3 Trust in AI-mediated decisions

Trust is not equivalent to liking or familiarity. In consumer settings, trust involves beliefs that a system is competent, benevolent enough to respect the user’s interests, and institutionally accountable. A highly fluent chatbot may appear competent while providing unreliable information. Conversely, a transparent system that acknowledges uncertainty may generate more durable trust.

Brand managers should therefore avoid treating anthropomorphic design as a substitute for reliability. Voice, names and friendly language can create social presence, but they can also encourage overreliance. Important practices include:

• clearly identifying automated systems;
• distinguishing facts, estimates and generated suggestions;
• explaining why recommendations were made;
• obtaining meaningful consent for consequential actions;
• preserving human escalation;
• recording decisions for audit and redress;
• testing systems across languages, disabilities and cultural contexts.

3. Synthetic Consumers and the Problem of Representation

A synthetic consumer is an artificial representation of a consumer, created from statistical models, behavioural data, generative systems or simulated personas. Synthetic consumers may be used to test advertisements, forecast demand, explore segments, simulate interviews or evaluate product concepts.

They can be valuable because they are scalable, available on demand and inexpensive relative to repeated fieldwork. They can help researchers explore hypotheses before investing in surveys or experiments. However, a synthetic consumer is not a human consumer. It does not possess embodied experience, personal history or genuine affect unless these are represented through model assumptions.

3.1 Appropriate uses

Synthetic consumers are most defensible when used for:

• early-stage idea generation;
• stress-testing possible responses;
• identifying questions for subsequent human research;
• simulating combinations of known characteristics;
• exploring scenarios where real-world experimentation is costly or risky.

They are less defensible as the sole basis for decisions involving vulnerable groups, high-stakes services, cultural interpretation or claims about actual consumer attitudes.

3.2 Validity risks

The principal risks are:

• Construct validity: the model may not represent the psychological construct being studied.
• Representation bias: training data may overrepresent some populations and erase others.
• Circularity: a model trained on existing marketing assumptions may reproduce those assumptions.
• False precision: numerical outputs may appear more certain than the evidence permits.
• Mode collapse: simulated consumers may produce similar, conventional responses.
• Reflexivity: consumers may change when they know they are being evaluated or targeted.

Synthetic research should therefore be treated as a form of model-based inquiry, not as a replacement for empirical observation. Outputs should be triangulated with qualitative research, behavioural data and controlled experiments involving real participants.

4. Immersive Environments and Consumer Psychology

Immersive environments include virtual reality, augmented reality, mixed reality, virtual worlds, game spaces and persistent digital environments. Their significance is not limited to visual novelty. They alter the consumer’s sense of presence, embodiment, social interaction and spatial behaviour.

4.1 Presence and embodiment

Presence is the subjective experience of being located within an environment. Embodiment concerns the feeling that a digital avatar or representation is connected to the self. Strong presence can increase emotional involvement and memory, while embodiment can influence self-perception and behaviour.

Immersive environments may support:

• product trial without physical inventory;
• spatial storytelling;
• collaborative shopping;
• training and demonstration;
• accessible previews of environments;
• social participation around brand meaning.

However, immersion can also intensify persuasion. The consumer may have fewer opportunities to distance themselves from the message, and behavioural data may include gaze, movement, voice, gesture and physiological responses. Such data can be more intimate than conventional clickstream data.

4.2 Identity and the extended self

Consumer identity is partly expressed through possessions, appearance, affiliations and environments. In immersive spaces, identity may be expressed through avatars, virtual goods, spatial design and membership. The self can therefore extend into digital objects that have no physical counterpart but carry social meaning.

Brand managers should distinguish between offering tools for self-expression and prescribing identities. Inclusive avatar systems, flexible customisation and portability of digital assets can support autonomy. Closed ecosystems that make identity dependent on one platform may create lock-in and reduce consumer control.

5. Voice Interfaces and the Reduction of Visual Branding

Voice assistants and conversational interfaces change the structure of brand choice. Visual environments can display many options simultaneously; voice interfaces typically present a sequential answer. This creates a scarcity of attention and a greater role for ranking, defaults and recommendation logic.

Voice interactions may feel intimate because they occur in domestic spaces and use conversational language. They can also be more accessible for consumers with visual, motor or literacy barriers. Yet voice systems may misrecognise accents, dialects or speech impairments. They may also make commercial intent less apparent than visual advertising.

Strategic consequences include:

• brands must optimise for usefulness, not only visual recognition;
• distinctive verbal identity becomes important;
• pronunciation, dialogue design and response timing affect experience;
• sponsored recommendations should be clearly disclosed;
• consumers need accessible transcripts, correction mechanisms and privacy controls.

The disappearance of the visual shelf does not eliminate competition. It relocates competition into the system’s ontology, ranking rules, data access and conversational context.

6. Connected Products and Continuous Relationships

A connected product can sense, communicate, receive updates or interact with other systems. Examples include wearables, connected vehicles, smart appliances, health devices and industrial products used by consumers.

The transaction consequently becomes a continuing relationship. Value may arise from software updates, predictive maintenance, personalisation and integration with other devices. The product may learn from usage, and the brand may learn from the product’s owner.

This creates a reciprocal data relationship with important psychological effects:

• Feedback: the product informs consumers about their behaviour.
• Self-monitoring: consumers may change behaviour in response to metrics.
• Dependence: routines may become linked to a particular ecosystem.
• Surveillance concern: consumers may feel observed in private settings.
• Identity reinforcement: usage data can become part of self-narrative.

A responsible connected-product strategy requires data minimisation, clear retention policies, security updates, understandable controls and functionality that does not disappear unfairly when a subscription ends. Security is also a brand attribute: a technically impressive product that exposes households to risk can destroy trust rapidly.

7. Evolving Ownership Models

Ownership traditionally provides possession, control, use and the ability to transfer or exclude others. Contemporary models separate these rights. Consumers may subscribe to music, lease vehicles, rent clothing, share mobility, access software or purchase products that depend on cloud services.

Ownership models influence the extended self. A purchased object may become part of personal history, whereas a subscription may encourage a more provisional relationship. Nevertheless, access-based services can also become identity-bearing when they organise routines, communities or status.

Managers should assess not only willingness to pay but also perceived control, fairness, continuity and exit costs. A low monthly price may be attractive initially but unacceptable if cancellation is difficult, data cannot be exported or essential functionality is remotely disabled.

8. How Emerging Technologies Change Decision Making

Emerging systems can reduce cognitive effort, but reduced effort is not always improved decision quality. Consumers may experience:

• choice delegation, in which a system selects among alternatives;
• automation bias, in which system recommendations receive undue weight;
• default effects, in which preselected options shape behaviour;
• algorithmic framing, in which the system defines what counts as relevant;
• personalisation effects, in which the same brand is presented differently to different people.

A useful distinction is between friction that protects reflection and friction that merely obstructs access. Removing unnecessary form fields improves experience. Removing all confirmation before an irreversible purchase may undermine autonomy. Good design places friction at consequential points and reduces it at routine points.

Decision quality should be evaluated through more than conversion. Relevant measures include comprehension, regret, reversibility, perceived autonomy, error rates, inclusion and longer-term satisfaction.

9. Scenario Planning for Consumer Futures

Forecasting attempts to predict what will happen. Scenario planning instead constructs plausible, internally coherent futures to test strategic robustness. It is particularly suitable for emerging technologies because uncertainty is high and historical data may not provide reliable extrapolation.

9.1 A practical scenario method

1. Define the focal question. Example: How will consumers choose household products in 2032?
2. Map driving forces. Include technology, regulation, economics, culture, demographics, infrastructure and climate.
3. Separate trends from uncertainties. A trend has direction and evidence; an uncertainty has multiple plausible outcomes.
4. Identify critical uncertainties. Select forces with high impact and high uncertainty.
5. Construct scenario logics. Use contrasting, plausible combinations rather than optimistic and pessimistic stereotypes.
6. Develop narratives. Describe consumer routines, institutions, business models and emotional meanings.
7. Stress-test the brand. Examine value proposition, data practices, channels, capabilities and risks.
8. Identify signposts. Define observable indicators that suggest a scenario is becoming more likely.
9. Choose robust actions. Prefer initiatives that create value across several futures.

A two-axis framework can use degree of consumer delegation as one axis and strength of institutional trust and regulation as another. This produces four useful scenario types:

These are not predictions. They are analytical environments for testing strategy.

9.2 Signposts and strategic options

Signposts may include the percentage of purchases initiated by agents, adoption of digital identity standards, frequency of AI-related consumer complaints, regulatory requirements for disclosure, consumer demand for data portability and changes in subscription cancellation rates.

Robust strategies generally include transparent AI disclosure, interoperable data practices, human escalation, accessible design, flexible ownership options and investment in trust measurement. Such strategies are valuable in both highly automated and more privacy-conscious futures.

10. Illustrative Case: LumaHome Connected Appliances

Background

LumaHome, a fictional mid-market appliance brand, launches a connected washing machine. The machine detects load size, recommends cycles, orders detergent and sends maintenance alerts. The company also introduces a subscription that provides advanced optimisation and remote diagnostics.

Problem

Early sales are strong, but consumer research identifies three concerns. First, customers do not understand which functions operate locally and which require cloud processing. Second, the detergent-ordering agent sometimes selects a premium brand without clearly explaining the commercial relationship. Third, customers fear that cancelling the subscription will make the machine less useful.

Analysis

The product creates value through convenience and reduced cognitive effort, but it also changes the relationship from ownership to conditional access. The agent’s recommendation creates a conflict between assistance and persuasion. The unclear data boundary weakens perceived control, while subscription dependence threatens fairness and autonomy.

LumaHome maps the consumer journey and identifies four trust moments: initial setup, first automated recommendation, subscription renewal and product failure. It also conducts human-participant research rather than relying solely on synthetic simulations. The research measures comprehension, perceived control, recommendation acceptance, regret and willingness to continue using the product.

Approach and solution

LumaHome adopts five changes:

1. It labels local processing, cloud processing and data retention in plain language.
2. It separates sponsored recommendations from neutral maintenance advice.
3. It requires confirmation before purchases and provides a spending limit.
4. It preserves core washing functions after subscription cancellation.
5. It introduces a dashboard showing decisions made by the agent and allowing reversal.

The company also creates an independent review process for high-impact personalisation and tests the interface with older users, people with disabilities and households with limited connectivity.

Outcome

In this illustrative case, conversion to the subscription grows more slowly than under an aggressive default strategy, but cancellation-related complaints decline and long-term satisfaction improves. The brand develops a reputation for dependable assistance rather than covert automation. The result demonstrates that trust may require accepting some short-term friction in exchange for durable relationship quality.

Learning points

• Connected products extend brand relationships beyond purchase.
• Convenience is not sufficient when consumers lack control or understanding.
• Agentic recommendations require disclosure, boundaries and reversibility.
• Ownership models should be evaluated through fairness and exit rights.
• Human research remains necessary even when synthetic modelling is available.

11. Ethical, Regulatory and Governance Considerations

Emerging consumer systems raise several governance questions.

Privacy and data protection: Data collection should be necessary, proportionate, purpose-limited and understandable. Sensitive inferences require particular care.

Manipulation and vulnerability: Personalisation can exploit loneliness, urgency, financial stress or cognitive limitations. Firms should distinguish persuasion from exploitation.

Bias and discrimination: Training data, interfaces and ranking systems may disadvantage groups through inaccurate recognition, unequal access or discriminatory pricing.

Transparency: Consumers should know when they are interacting with AI, when content is generated, when recommendations are paid for and when decisions are automated.

Accountability: A brand cannot transfer responsibility to a vendor or algorithm. Consumers need a route to explanation, correction and compensation.

Security and resilience: Connected products create attack surfaces. Security updates and end-of-life policies should be part of product design.

Accessibility: Voice, immersive and agentic services should support varied abilities, languages, literacy levels and connectivity conditions.

A governance programme should combine legal compliance with ethical design review, impact assessment, red-team testing, incident reporting and ongoing monitoring. Compliance is a minimum standard; trustworthy brand management requires attention to lived consumer experience.

12. Managerial Best Practices

Brand managers preparing for emerging consumer futures should:

1. Design for agency: make delegation explicit, bounded and reversible.
2. Treat trust as a system property involving technology, policy and service recovery.
3. Build provenance into generated content and recommendation systems.
4. Use synthetic consumers for exploration, not as unquestioned substitutes for people.
5. Measure comprehension and autonomy alongside engagement and sales.
6. Offer meaningful alternatives to surveillance-based personalisation.
7. Make cancellation, data export and switching reasonably practical.
8. Test experiences with diverse and vulnerable consumers.
9. Use scenario planning to identify robust capabilities rather than betting on one forecast.
10. Establish clear ownership of AI-related decisions inside the organisation.

Chapter Summary

Emerging consumer futures are defined by distributed agency, continuous data relationships and the blending of physical, digital and social environments. Generative AI creates content and assistance, while agentic AI can plan and act on behalf of consumers. These systems can reduce effort but may also produce automation bias, opacity and overreliance.

Synthetic consumers support exploratory modelling but cannot replace empirical research with real people. Immersive environments influence presence, embodiment and identity; voice interfaces relocate competition into conversational ranking and trust; connected products transform one-time transactions into continuing relationships. Subscription, sharing and product-as-a-service models separate use from ownership and make control, continuity and exit rights central to consumer value.

Scenario planning helps managers examine plausible futures without pretending to predict a single outcome. Robust strategies emphasise transparency, accountability, accessibility, security, reversibility and consumer agency. The future of branding will depend less on technological novelty than on whether consumers experience emerging systems as useful, intelligible and worthy of trust.

Key Terms

Agentic AI: An AI system capable of pursuing goals through planning, tool use and action.

Algorithmic framing: The way a system defines which information or alternatives are relevant.

Automation bias: The tendency to over-trust or over-weight automated recommendations.

Connected product: A physical product capable of sensing, communicating or receiving digital services.

Consumer agency: The ability to understand, influence and control consumption decisions.

Delegation: Authorising a system or another party to perform a decision or action.

Embodiment: The felt connection between the self and a physical or digital body or avatar.

Extended self: The aspect of identity expressed through possessions, environments, affiliations or digital objects.

Generative AI: AI that produces new content such as text, images, audio or video.

Immersive environment: A digital or mixed environment designed to create a strong sense of presence.

Interoperability: The ability of systems, data or digital assets to work across platforms.

Ownership model: The arrangement governing possession, access, use, control and transfer.

Presence: The subjective feeling of being located within a mediated environment.

Scenario planning: A method for exploring multiple plausible futures and testing strategic robustness.

Signpost: An observable indicator that a particular future may be emerging.

Synthetic consumer: An artificial representation of a consumer generated through data or models.

Voice interface: A system through which users interact primarily through spoken language.

Trust: Belief that a system or organisation is competent, sufficiently benevolent and accountable.

Review Questions

Short Answer Questions

1. What is the difference between generative AI and agentic AI?
2. Why should synthetic consumers not automatically replace human research participants?
3. Define presence and embodiment in immersive environments.
4. How can subscription models affect perceived consumer control?
5. What is a signpost in scenario planning?

Long Answer Questions

1. Analyse how agentic AI may change consumer decision making and brand competition.
2. Discuss the psychological and strategic implications of connected products.
3. Evaluate the benefits and risks of immersive environments for brand management.
4. Compare ownership, subscription and product-as-a-service models from the consumer’s perspective.
5. Explain why transparency and reversibility are central to trustworthy AI-mediated consumption.

Analytical / Application Questions

1. A retailer wants an AI agent to select products automatically for customers. Design a governance framework covering authorisation, disclosure, ranking fairness, human escalation and cancellation.
2. Construct two contrasting scenarios for consumer choice in 2032 using consumer delegation and institutional trust as the critical uncertainties. Identify three signposts for each scenario.
3. Evaluate whether a synthetic consumer study should be used to launch a health-related product. Specify what additional human research and safeguards would be necessary.

Multiple Choice Questions with Answers

Question 1

Which feature most clearly distinguishes agentic AI from generative AI?

A. It can produce text or images.

B. It can plan and execute actions toward a goal.

C. It always operates without data.

D. It is used only for entertainment.

Correct Answer: B. It can plan and execute actions toward a goal.

Question 2

What is a major validity risk of synthetic consumers?

A. They always require physical prototypes.

B. They cannot produce any output at scale.

C. They may reproduce assumptions and biases in their training data.

D. They eliminate the need for scenario planning.

Correct Answer: C. They may reproduce assumptions and biases in their training data.

Question 3

Presence in an immersive environment refers primarily to:

A. The subjective feeling of being located within the environment.

B. The price of a virtual product.

C. The legal ownership of an avatar.

D. The speed of an internet connection.

Correct Answer: A. The subjective feeling of being located within the environment.

Question 4

Why can voice interfaces alter brand competition?

A. They display unlimited alternatives simultaneously.

B. They eliminate the importance of recommendation ranking.

C. They often present options sequentially, increasing the importance of defaults and ranking.

D. They prevent all forms of commercial persuasion.

Correct Answer: C. They often present options sequentially, increasing the importance of defaults and ranking.

Question 5

Which practice best supports consumer agency in an agentic purchasing system?

A. Hiding the system’s commercial relationships.

B. Removing all confirmation steps.

C. Providing explicit constraints, monitoring and reversal options.

D. Using only premium recommendations.

Correct Answer: C. Providing explicit constraints, monitoring and reversal options.

Question 6

A product-as-a-service model generally gives consumers:

A. An outcome or continuing performance rather than only a one-time transfer of possession.

B. Permanent control without any provider involvement.

C. A guarantee that no data will be collected.

D. Automatic ownership of all supporting infrastructure.

Correct Answer: A. An outcome or continuing performance rather than only a one-time transfer of possession.

Question 7

What is the main purpose of scenario planning?

A. To guarantee an accurate single forecast.

B. To explore plausible futures and test strategic robustness.

C. To replace all consumer research.

D. To select the cheapest technology.

Correct Answer: B. To explore plausible futures and test strategic robustness.

Question 8

Which is an example of automation bias?

A. Rejecting every recommendation made by a system.

B. Over-weighting an automated recommendation despite contrary evidence.

C. Requesting a human explanation of a decision.

D. Comparing several independent sources.

Correct Answer: B. Over-weighting an automated recommendation despite contrary evidence.

Question 9

Why is reversibility important in AI-mediated consumption?

A. It allows consumers to undo or correct consequential automated actions.

B. It guarantees that every recommendation is correct.

C. It eliminates the need for privacy protection.

D. It makes all products free to use.

Correct Answer: A. It allows consumers to undo or correct consequential automated actions.

Question 10

Which use of synthetic consumers is most defensible?

A. Making final decisions about vulnerable consumers without human research.

B. Replacing all field experiments in high-stakes services.

C. Generating early hypotheses to be tested with real consumers.

D. Proving that a cultural group has one fixed preference.

Correct Answer: C. Generating early hypotheses to be tested with real consumers.