Psychology of Gen AI
- Research & Data Quality
- Article
Why Synthetic Respondents Flatten Consumer Sentiment
A new ARF Psych of GenAI experiment reveals that large language models apply a rigid, rule-driven logic when evaluating privacy scenarios—even when humans typically shift their reasoning based on framing, emotion and social context. Unlike consumers, who blend intuition, feeling and social perspective into their judgments, GPT-4o relied on a single internal rule across all testing conditions: data use is acceptable only with explicit consent. This consistency offers value for certain analytic tasks but exposes limits for advertising research that depends on emotional nuance and context-sensitive consumer insight.
Member Only Access
- Research & Data Quality
- Article
Steering AI Bias: How Persona Prompts Unlock Nuance in Gen AI Responses
Large language models mirror human cognitive biases—but can those biases be guided? New ARF and MSI research reveals that while loss aversion remains deeply ingrained in AI responses, introducing persona information, such as demographics or personality traits, can increase variability and make outputs more nuanced. For advertisers and researchers, this opens the door to design strategic prompts that spark richer and more nuanced, human-like responses.
Member Only Access
- Research & Data Quality
- Article
Decoding AI’s Bias: What Loss Aversion Means for Advertising Research
This experiment, part of the ARF’s “Psychology of Gen AI” series, reveals how models like ChatGPT not only replicate human cognitive biases, such as loss aversion, but also compress variability into uniform patterns. This raises concerns for advertising researchers who rely on authentic insights into consumer behavior.
Member Only Access