Apple Podcasts  ·  Spotify  ·  Buzzsprout

Paper A

Consumer Trust in AI-Generated Marketing Content: A Systematic Literature Review and Research Agenda

Kirill Baryshkov, Yana Kuzina, Iryna Smuk, Mila Tkachuk — 2026 — American Impact Review

peer reviewed journal article  ·   ·  read now

Key findings

• Telling consumers that marketing content was made by AI often makes them trust it less — they become more suspicious that the brand is trying to manipulate them. But this doesn't happen every time; it depends heavily on the situation.
• The biggest reason AI disclosure hurts brands is that consumers feel the content is less 'real' or genuine. When people sense that no human creativity or emotion went into a message, they trust it less. A secondary pathway is that some consumers feel mild moral disgust — a gut-level discomfort — when they realize a brand used AI to communicate with them.
• Whether AI disclosure hurts a brand depends on several factors: emotional ads are more vulnerable than fact-based ads; consumers who already know a lot about AI are less bothered; cultural background matters; and making the AI seem more human-like (e.g., giving it a name or face) can partly offset the trust damage.
• The review identifies a significant gap — most research measures self-reported attitudes, not actual behavior (like purchases or clicks), so we don't yet know how much these trust effects translate into real-world consumer decisions.

Marketing implications

• If you're using AI to write emotional content — heartfelt stories, empathy-driven messages, sentimental ads — be more careful about slapping an 'AI-generated' label on it. The research suggests emotional content takes a bigger trust hit from disclosure than rational, fact-based content. Save AI disclosure for product specs and rational ads, not for emotional campaigns.
• If you need to disclose AI use (e.g., due to platform rules or regulations), experiment with how you frame it. Framing matters — 'AI-assisted by our team' likely lands differently than 'fully AI-generated.' Test which framing preserves trust best with your audience before rolling it out at scale.
• Don't assume your AI content strategy will work the same across different countries or demographics. Audiences with more AI knowledge are less bothered by AI disclosure. Consider adding short educational context or transparency notes in markets where AI awareness is lower.

Paper B

Algorithmic influence and consumer decision-making: empirical evidence on the limitations of predictive AI in marketing communication management

Pabllo Barcellos Soares Ferreira, Marcelo Henrique Neves Pereira — 2026 — Revista de Administração da UFSM

peer reviewed journal article  ·   ·  read now

Key findings

• The AI attention tool consistently got it wrong for Brazilian consumers: it over-predicted that people would look at visually flashy or high-contrast elements, while real Brazilian participants spent more time reading text and looking at contextually meaningful details.
• Brazilian consumers paid more attention to words and meaning than to visual 'pop' — the opposite of what the AI predicted — suggesting that AI tools trained mostly on Western audiences may not accurately model how people in other cultures process ads.
• Using an AI attention tool calibrated on European or American consumers as if it applies universally could lead marketers in Brazil (and likely other emerging markets) to design ads that nobody actually looks at in the intended way.
• Inaccurate AI attention predictions can make consumers work harder to find what they need in an ad or menu, potentially worsening their experience rather than helping them decide.

Marketing implications

• If you are running campaigns in Brazil or any market outside the US/Europe, don't trust AI attention-prediction tools at face value — run a quick local user test or eye-tracking study to check if the AI's predictions actually match how your real audience looks at the design.
• When designing ads or menus for non-Western audiences, prioritize clear, readable text over eye-catching visuals — local consumers may care more about what something says than how flashy it looks.
• Before using an AI design-optimization tool, ask the vendor: what data was it trained on? If the answer is mostly American or European users, treat its recommendations as a starting point, not a final answer, especially for emerging markets.

Paper C

The Dynamics of Customer Engagement Within an AI-Driven Marketing Environment

SH.M. Fakhar-e-Alam Siddiqui, Maham Abid — 2026 — ACADEMIA International Journal for Social Sciences

peer reviewed journal article  ·   ·  use cautiously

Key findings

• When people feel that AI-powered marketing is effective — meaning it gives them relevant, useful suggestions — they engage more with the brand (click more, spend more time, participate more).
• Trust matters a lot: when people trust that AI recommendations are honest and transparent, they are more likely to interact with personalized content and feel closer to the brand.
• AI systems that keep learning and improving over time (rather than staying static) also drive higher engagement — people respond better to tools that seem to 'get smarter' about their preferences.
• Engagement acts as a bridge: it explains why AI capabilities lead to satisfaction. In other words, AI doesn't directly make customers happy — it first gets them engaged, and engagement then drives satisfaction.

Marketing implications

• If you're using AI-powered recommendations or chatbots, make sure they're transparent about how they work — telling users 'we suggested this because you liked X' builds trust and keeps them coming back.
• Don't just deploy AI and leave it static. Make sure your AI tools are actually learning from customer behavior over time (e.g., improving recommendations based on purchase history). Customers notice when suggestions get more relevant.
• Think of AI as an engagement tool first, not a satisfaction tool. Focus on getting customers to interact — click, respond, explore — and satisfaction follows naturally from that engagement.

Paper D

AI-Driven Marketing Personalization and Customer Loyalty

Manish Satpal — 2026 — Scriptora International Journal of Research and Innovation (SIJRI)

peer reviewed journal article  ·   ·  use cautiously

Key findings

• The more personalised the marketing a customer receives (recommendations, offers, messages tailored to their behaviour), the more likely they are to buy again, feel satisfied, and tell friends about the brand.
• AI-powered personalisation works partly because it makes customers feel understood and valued — that sense of being 'seen' by a brand builds trust and emotional attachment, not just repeat purchases.
• Over-personalisation is a real risk: if customers feel a brand knows too much about them or they can't control what data is used, their trust can drop and they may disengage, reducing the loyalty gains.
• Combining AI automation with human judgment — rather than letting algorithms run completely unchecked — is associated with better long-term loyalty outcomes, according to the paper's practical recommendations.

Marketing implications

• If you run email or product recommendation campaigns, audit how personalised they actually are — customers who feel seen and understood are more likely to come back, so generic blasts are leaving loyalty on the table.
• Set a 'creepiness check' before launching any personalisation feature: if a customer would feel surveilled rather than helped by a recommendation, pull back. Feeling over-targeted can destroy the trust you're trying to build.
• Don't let your personalisation run fully on autopilot. Build in human review points — a marketer spot-checking AI-generated segments or offers — because unchecked automation is where trust breaks down.

Paper E

A Study on Consumer Perception Towards AI-Based Marketing Chatbots

M. Geetha, R. Jayasri — 2026 — Journal of Advance and Future Research

peer reviewed journal article  ·   ·  use cautiously

Key findings

• 80% of respondents use AI chatbots either regularly (38%) or frequently (42%) while shopping online — only 6% have never used one.
• The top-rated chatbot feature was speed: consumers ranked 'responds quickly to my queries' as the most important attribute (average score 4.08 out of 5).
• Providing accurate and relevant information ranked second (3.74/5), followed by ease of use (3.72/5) — meaning speed matters more to consumers than accuracy or convenience.
• Despite generally positive attitudes, consumers still worry about data privacy, lack of human interaction, and chatbots giving wrong answers.

Marketing implications

• Speed is what customers actually care about most when talking to a chatbot — if your chatbot is slow, fix that before anything else. Accuracy matters second, not first.
• Privacy concerns are real blockers for chatbot adoption. Add a one-sentence plain-English privacy note at the start of any chatbot conversation ('We don't store your personal data') to reduce drop-off.
• Offer a clear 'Talk to a human' escape hatch in your chatbot flow — consumers still worry about the absence of human touch, and making it easy to escalate builds trust without removing the chatbot.

Apple Podcasts  ·  Spotify  ·  Buzzsprout