Perceptions of AI-Generated vs. Human-Created Marketing Content: A Literature Review

1. Introduction

The integration of artificial intelligence (AI) in marketing content creation has revolutionized how businesses communicate with their customers, particularly in the Software as a Service (SaaS) sector. This literature review examines how paying administrative customers perceive AI-generated marketing content compared to human-created materials, with a focus on the influencing factors of trust, relatability, cognitive bias, transparency, and cultural attitudes. The growing adoption of AI-powered tools in content creation raises important questions about how disclosure of AI involvement affects customer perception and engagement, especially across different cultural contexts such as American and Indian markets.

The literature reveals a complex relationship between AI-generated content and consumer perception, with studies indicating varying responses depending on multiple factors including content type, disclosure practices, and cultural backgrounds. This review synthesizes findings from recent research to provide a comprehensive understanding of how AI is reshaping the marketing landscape and influencing customer perceptions.

2. Perceptions of AI-Generated vs. Human-Created Marketing Content

Research indicates significant differences in how consumers perceive content based on its authorship. When the AI authorship of content is disclosed, there is often a significant negative effect on perceived quality. This suggests that transparency about AI involvement can influence consumer judgments about content value. However, studies have shown a general preference for AI-generated content in certain contexts, particularly when associated with greater perceived objectivity. This preference is especially notable in conventional contexts but diminishes in unconventional scenarios, indicating that unique or specialized content may reduce AI's perceived advantages.

The perception of AI-generated content is often context-dependent. Experimental research investigating whether authorship information (AI vs. designer) affects the perception and evaluation of graphic messages has shown measurable differences in how audiences respond. These studies typically employ questionnaires with Likert scales to assess factors such as aesthetic value, creativity, and overall impression, providing quantitative insights into perception differences. This research has significant implications for communication strategies and marketing value in shaping desired brand images.

In advertising contexts, [1]experimental studies testing how consumers react to advertising messages featuring content generated by an AI neural network demonstrated that potential donors responded differently to children's faces when they knew they had been generated by AI. The awareness of an image being AI-generated had a negative impact on donation intentions. [1]This negative impact was serially mediated by empathy, anticipatory guilt, and emotion perception, highlighting the emotional component of content perception. These findings suggest that AI-generated content may struggle to evoke the same emotional responses as human-created content, potentially limiting its effectiveness in contexts where emotional connection is crucial.

3. The Role of Trust and Relatability in Content Perception

Trust and relatability emerge as critical factors influencing how consumers engage with AI-generated versus human-created content. Research indicates that when AI influencers post marketing content with functional advertising appeal, they can significantly enhance people's perceived trust. Conversely, human influencers sharing marketing content with experiential advertising appeal significantly improve perceived empathy. This suggests a fundamental difference in how consumers relate to AI versus human sources, with AI being more effective for rational, functional messaging while humans excel at creating emotional connections.

The relationship between AI content and trust is further nuanced by research showing that there exists a strong positive correlation between the perceived realism of AI influencers and consumer trust. This indicates that as AI-generated content becomes more sophisticated and realistic, trust levels may increase accordingly. However, virtual AI influencers were found to be more effective in influencing purchase intentions for low-involvement products, while human influencers were more effective for high-involvement products. This differentiation suggests that the stakes of the decision impact the preferred source of influence, with consumers gravitating toward human sources for more significant purchasing decisions.

Several studies highlight the complementary roles that AI and human content can play in marketing strategies. While virtual influencers offer advantages like control and cost-effectiveness, human influencers provide authentic, personal connections. The effectiveness of either type varies based on product type and consumer perceptions, emphasizing the importance of strategic selection in marketing campaigns. [2]Although AI-generated content can be produced quickly, it should be vetted by experienced experts to ensure alignment with brand values and image. Even when AI promises to generate content and maintain brand voice, successful implementation requires collaboration with experienced professionals and thoughtful consideration of its use in building and maintaining an authentic and effective brand voice.

4. Transparency and Disclosure of AI Involvement

The disclosure of AI involvement in content creation significantly influences consumer perception. Research shows that disclosing the AI authorship of content significantly affects its perceived quality negatively, underscoring the impact of transparency on the acceptance of AI-generated materials. These findings highlight both the potential of generative AI in content creation and its limitations in certain contexts. This suggests that while AI can create content efficiently, transparency about its use must be carefully managed to avoid negative perception.

Studies have emphasized the importance of declaring the role of humans in the AI text generation process, as awareness of a text's origin can influence consumer perceptions. This indicates that the perceived level of human involvement in AI-generated content impacts how audiences respond to it. In some cases, [1]research has investigated various motives for employing AI-generated images and indicated that entities employing those images can benefit by making their ethical motives salient. Under extraordinary circumstances, the use of AI images is considered acceptable by consumers and is likely to lead to similar outcomes as the use of real images. This suggests that context and purpose can override negative perceptions of AI-generated content when properly communicated.

The transparency issue extends beyond simple disclosure to include how AI is implemented in the creative process. Studies reveal that generative AI promotional content impacts purchase intention due to its perceived entertainment, transparency, and usefulness. This indicates that consumers value knowing how AI is being used and why, rather than just whether it is being used at all. [2]Although AI can generate content quickly, achieving satisfactory brand-voice results requires careful analysis and extensive, representative data that informs the AI system. AI-generated content should be vetted by experienced experts to ensure it aligns with brand values and image, suggesting that transparent hybrid approaches may be most effective.

5. Cultural and Demographic Influences on AI Content Perception

Cultural background significantly influences how consumers perceive AI-generated content. Cross-cultural analysis has shown numerous significant differences in how different cultural groups evaluate AI-generated advertisements. For example, one study found that Colombians consistently determined that AI-generated ads were more visually appealing than Austrians. Such cultural variations suggest that AI content perception is not universal but is instead shaped by cultural contexts and preferences.

Research has shown that cultural relevance plays a significant role in content perception. Colombians ranked culturally relevant advertising, such as those about skiing and the Alps, higher in visual appeal and found that they caught attention more successfully. They were also more inclined to interact with the ads regarding engagement in general comparison. When considering culturally specific ads, Colombians rated the overall quality higher than Austrians, suggesting they found the culturally relevant ads more enjoyable overall. These findings highlight how cultural background influences not just general perception of AI content but specifically how cultural relevance within that content affects engagement and quality assessment.

Demographic factors beyond national culture also influence AI content perception. Research has found that demographic characteristics showed significantly different results in their response to AI-generated promotional content, with status motivations having a stronger impact on purchase intention. Consumer knowledge levels also play a role, as those with lower levels of knowledge are more susceptible to the matching effect of influencer type and advertising appeal type on the persuasive effect of marketing content than those with higher levels of knowledge. This suggests that education, expertise, and other demographic factors interact with cultural background to shape perceptions of AI-generated content.

6. Methodological Approaches in Studying AI-Generated Content Perception

The research field employs diverse methodological approaches to study AI-generated content perception. One comprehensive approach involves developing applications that generate ads based on AI systems like GPT-4 and DALLE-3, creating persona generators to define target demographics, and using structured questionnaires to assess various aspects of perception. These questionnaires typically ask about message clarity, trustworthiness, visual appeal, interest matching, attention attraction, and potential interaction on a 5-point scale, along with free-text questions about elements encouraging interaction and emotional responses. Statistical analyses such as T-tests are then used to determine the significance of response differences.

Neuromarketing techniques provide deeper insights into consumers' subconscious responses to AI-created content. Studies have used eye tracking, galvanic skin response (GSR), and facial expression analysis to investigate the perception of visual, textual, and audio content created by AI. These techniques provide deeper insights into attention, emotional engagement, and consumers' overall perception of AI-generated content. The integration of these various measurement techniques enables a comprehensive understanding of consumer responses and provides practical recommendations for research and use of AI in marketing communications.

Mixed-methods approaches are increasingly common in this field. Some studies adopt a mixed-method research approach that integrates quantitative surveys and structural equation modeling analysis with qualitative Natural Language Processing (NLP). Other research has focused on understanding the usage of AI in marketing from both the perspectives of marketing professionals and consumers through in-depth interviews, surveys, and extensive study of existing literature. Topic modeling approaches (such as BerTopic) have been used to analyze online reviews and identify topics included in consumer conversations about AI-generated content. This methodological diversity reflects the complexity of studying perceptions, which involve both conscious and unconscious processes.

7. Limitations and Research Gaps

Despite growing research interest, several limitations and gaps exist in our understanding of how customers perceive AI-generated versus human-created marketing content. Many studies analyze responses from participants from limited cultural backgrounds. While this allows for some cross-cultural insights, the sample size and cultural diversity are often limited, which may affect the generalizability of findings. Future studies with more diverse participant pools could provide a broader understanding of how different cultures perceive AI-generated content. This is particularly relevant given the focus of our research question on American and Indian contexts, which are underrepresented in the current literature.

Technical limitations also impact research quality and findings. The capabilities of AI models used, such as GPT-4 and DALL-E 3, constrain studies. These models, while advanced, still need improvement in fully understanding and replicating human creativity, particularly in areas such as appropriate design, cultural references, and the integration of text and visuals. The lack of interaction between text and image generation phases often results in inconsistencies between content and visuals. Overusing certain words in captions or over image text is a common constraint of current models, which tend to stick to words like "Enhance" and "Elevate" regardless of the product or service, deteriorating the quality of the final output.

Most notably, there is a significant gap in research specifically addressing how paying administrative customers of SaaS products perceive AI-generated content. While some studies look at the challenges and opportunities involved in implementing AI in marketing and how AI technologies can accelerate higher levels of engagement, better ROI, and strengthen relations between consumers and brands, they typically don't focus on the specific context of B2B SaaS customers. Similarly, studies on how AI influences brand perception based on consumer behavior towards brands adopting AI into their marketing strategies often focus on general consumers rather than business customers making purchasing decisions for their organizations. This represents a significant gap in understanding the unique considerations of SaaS administrative customers.

8. Future Research Directions

Several promising directions for future research emerge from this review. First, integrating text and image generation models could significantly improve the coherence and quality of AI-generated ads. Using tools like ChatGPT-Vision to offer feedback on generated DALL-E images to GPT could be a step forward in automating the whole content creation process. This technical advancement would enable more sophisticated research into perception differences.

Second, there is a need for more research specifically examining the perceptions of administrative decision-makers in B2B contexts. Future studies should investigate how companies can select the more appropriate promoter between AI influencers and human influencers based on their marketing content, and how enterprises can design more persuasive marketing content according to the type of influencer. Such research could assist enterprises in scientifically matching influencer types with advertising appeals, maximizing marketing effectiveness through refined strategies.

Third, as AI is expected to develop more human-like characteristics and behavior, testing theories centered on human influence in an AI context is crucial for marketing theory and practice. Future research should examine the human characteristics of perceived expertise, transparency, and entertainment in an AI context in light of status consumption motivation and other psychological factors. Additional research could identify the applications of AI in digital marketing, evaluate its impact on competitive advantages, explore the risks associated with its use, and recommend effective and ethical implementation of AI technologies in marketing strategies. This could help address concerns about data privacy policies and employee training on ethical AI use in marketing contexts.

9. Conclusion

This literature review has examined how customers perceive AI-generated marketing content compared to human-created content, considering factors such as trust, relatability, cognitive bias, transparency, and cultural attitudes. The research indicates that perception is highly context-dependent, with AI content sometimes preferred for its objectivity but often viewed less favorably when its AI origin is disclosed. Trust and relatability emerge as critical factors, with AI generally stronger at fostering trust in functional contexts while humans excel at creating empathetic connections in experiential contexts.

Transparency about AI involvement significantly impacts perception, generally negatively when AI authorship is disclosed without proper context or ethical justification. Cultural and demographic factors further complicate these perceptions, with evidence of significant differences across cultural groups and knowledge levels. Current research employs diverse methodological approaches, from experimental studies to neuromarketing techniques, but is limited by technological constraints, sample diversity, and a lack of focus on B2B SaaS contexts.

Future research should focus on improving AI content generation technologies, examining B2B decision-maker perceptions specifically, and investigating how increasingly human-like AI characteristics affect perception. As AI continues to evolve in marketing applications, understanding these perception nuances will be crucial for effectively leveraging both AI and human-created content in marketing strategies.