Content area
This experimental study investigated how generative AI authorship affects the perceived credibility and persuasion of B2B blog content. Using a 2×2 factorial design, 220 participants evaluated blogs that were either human-written or AI-generated, each accompanied by a label that either correctly or incorrectly identified the author (source). Statistical analysis revealed no significant differences in source credibility, message credibility, or persuasion ratings between human and AI sources, contrary to hypotheses grounded in source credibility theory. Surprisingly, 68.6% of participants failed to correctly recall the content's stated author, suggesting source attribution may be processed peripherally in digital environments. These findings challenge traditional assumptions about source credibility, indicating that in B2B marketing contexts, message characteristics may matter more than source attribution. For B2B marketers, the results suggest that using AI content tools may not harm, and may even modestly enhance, persuasion and perceived credibility, though these differences were not statistically significant. The observed (but non-significant) trend toward higher ratings for AI-labeled content may indicate evolving comfort with generative content among younger, tech-curious audiences. The non-significant differences between human and AI authorship represent an important departure from previous research on algorithm aversion, pointing to evolving audience perceptions as generative AI becomes increasingly integrated into communication practices. Future research should investigate cognitive mechanisms of source evaluation, track longitudinal changes in AI content reception, and examine industry-specific applications with professional audiences.