AI ads match humans in quality but lag in performance
A groundbreaking study conducted by Ipsos in collaboration with Syracuse University's S.I. Newhouse School of Public Communications reveals a critical distinction between artificial intelligence and human creativity in advertising. While AI-generated ads are now visually indistinguishable from human-made work, they consistently underperform when predicting short-term sales impact. The research tested 20 advertisements across 10 major brands with 3,000 U.S. respondents to determine the true efficacy of automated creative tools. The study design was rigorous, pairing existing human-made ads produced before 2021 with fully AI-generated counterparts created from the identical strategic brief. This ensured that the only variable was the creator. Consumers were then surveyed to gauge both their ability to identify the source of the ads and their perception of the ads' effectiveness. The first finding confirms that the line between human and machine creation has blurred significantly. Only 13% of viewers were confident they could identify an AI-generated ad, a figure identical to the percentage who suspected human-made ads were artificial. With 40% of all participants uncertain regardless of the ad's origin, visual detection is no longer a viable metric for distinguishing AI content. However, perceptual similarity did not translate to performance. Despite looking the same, human-made ads outperformed their AI counterparts. Using sales-validated measures, human ads over-indexed against the benchmark by an average of 11 points, whereas AI ads under-indexed by five points. In practical terms, this suggests that human creativity drives stronger short-term sales results. Performance varied based on the complexity of the creative challenge. AI tools performed adequately when the brief was straightforward and product-driven. However, they struggled significantly when the task required storytelling, emotional resonance, or a genuine point of view. The most effective pairing in the study involved Cheerios, where a deeply human-centric brief yielded the highest combined effectiveness scores across both human and AI versions. Adam Peruta, who oversaw the technical process, and Carrie Riby, who provided advertising strategy, led the academic side of the project. Peruta noted that the AI was tasked with performing every function of a creative team, from interpreting strategy to producing the final spot. Riby highlighted that the study was inspired by her students' lack of enthusiasm for their own AI-generated work, observing that if creators are unimpressed, consumers are unlikely to be swayed. The data validates this intuition, showing that human creators produce work that moves the needle more effectively. Ryan Barthelmes of Ipsos, who guided the research, stated that the findings provide a necessary framework for an industry grappling with the future of creative agencies. While chief marketing officers face pressure to replace agencies with AI, the data indicates that human capacity for emotional connection remains a competitive advantage. The study concludes that AI is a powerful tool but not a replacement for human-led creativity. The industry is advised not to settle for average output generated by machines, but to leverage AI as a partner in a hybrid model where human oversight ensures the emotional depth required for successful advertising.
