Sensory evaluation plays a pivotal role in the world of cigars. But as technology advances, a question arises: can artificial intelligence (AI) replace the traditional human panels used to assess these products? The answer is complex and depends on the purpose of the sensory evaluation — whether it’s for commercial/promotional purposes or for analytical/predictive modeling.
Sensory Inputs for Analytical and Predictive Modeling: Where Accuracy is Paramount
For businesses, sensory data often drives critical decisions, such as new product development, portfolio optimization, or targeting specific markets. Accurate sensory information is crucial in these scenarios, and this is where sensory science tools come into play.
AI is showing promise in this field. Researchers are actively exploring how AI can assist panel leaders in designing sensory tests, identifying suitable panelists, and developing training materials (Odello L., Sensorialita’ Artificiale, Feb 2023). Some preliminary findings suggest that AI could streamline these processes and enhance the efficiency of sensory testing. Additionally, the integration of electronic noses and taste sensors with AI is on the horizon, offering exciting possibilities for the future. While AI has the potential to assist in data interpretation and report generation, it cannot fully replace human expertise and nuanced understanding.
Some companies are experimenting with correlating gas chromatography (GC) outputs — the chemical analysis of products — with consumer responses. However, bypassing a trained sensory panel is risky. GC is precise in identifying and quantifying the types and concentrations of molecules in a sample, but it doesn’t directly translate to sensory perception.
The Challenge of Interpreting Chemical Data and Consumer Preferences
The same molecule can evoke different sensory characteristics at varying concentrations. For example, isoamyl acetate smells like green apple at one level but like nail polish remover at another. Similarly, 1-octen-3-ol, a compound found in mushrooms, has a mushroomy aroma at high concentrations but can smell like raspberries at lower ones. This complexity requires the nuanced understanding of a trained human panel.
Consumer preferences are subjective and influenced by factors beyond chemical composition, including personal experiences, cultural background, and expectations. This further underscores the importance of human sensory panels in accurately assessing how products will be perceived by consumers.
Unless companies are willing to sacrifice accuracy or have AI systems trained on massive datasets that include sensory panel data, relying solely on AI for important decisions is a gamble. At Cigar Sense, we prioritize the accuracy of our human panel and haven’t found a cost-effective AI alternative that matches it.
Sensory Inputs for Commercial and Promotional Purposes: AI’s Creative Potential
For marketing and promotion, AI is already making waves. Tools like ChatGPT can generate compelling and informative text based on vast amounts of data. Gas chromatography has no application here.
However, brands should be cautious about scraping existing content for promotional copy, even if authorized. This can lead to generic, uninspiring messaging that fails to resonate with consumers.
Instead, AI-powered social listening, if used carefully, can help brands understand consumer sentiment and identify trends.
The challenge of social listening lies in filtering out the noise to find meaningful insights. Sarcasm, cultural nuances and hate speech can all obscure valuable feedback. Therefore, a nuanced approach is needed to leverage AI effectively in understanding consumer and reviewer sentiment.
The Human Element: An Irreplaceable Ingredient
While AI and GC offer exciting possibilities, they cannot fully replicate the human element in sensory evaluation. The nuanced understanding, cultural context, and interpretation that humans bring to the table are irreplaceable, especially when accuracy and consumer trust are at stake.
As technology continues to evolve, the relationship between AI, GC, and human sensory panels will undoubtedly become more intertwined. But for now, the human element remains an essential ingredient in the recipe for meaningful sensory insights.