Organizations that treat AI adoption as a uniform process risk building compliance without genuine acceptance, according to researchers at Texas A&M University. A new framework from Mays Business School shows that employees and consumers experience generative AI in fundamentally different ways, and that mandates alone often fail to drive meaningful use.
Huachao Gao and Shrihari Sridhar developed the RISE framework to capture what they call AI heterogeneity-the variation in how people perceive, trust, and use tools like ChatGPT, Google Gemini, and Anthropic's Claude. Their paper, published in Customer Needs and Solutions, draws on a survey of 2,144 US adults and warns that conventional adoption measures can obscure the motivations and concerns beneath the surface.
Four dimensions of AI experience
The RISE framework organizes these differences into four areas: relational meaning (the personal and social significance people attach to AI), in-context segmentation (how attitudes shift depending on the task), the skepticism-usage paradox (using AI despite anxiety about it), and equitable AI integration (ensuring fair access and outcomes). The survey found that attitudes do not split neatly into pro- or anti-AI camps. Respondents associated the technology with productivity and creativity while also voicing concerns about job loss, privacy, and manipulation.
One striking pattern was the skepticism-usage paradox. Higher AI anxiety correlated with slightly more frequent use and longer daily sessions. The paper suggests perceived necessity, fear of falling behind, and workplace or social pressure as possible drivers, though the cross-sectional data cannot establish cause and effect. The researchers note that self-reported responses from US consumers provide illustrative rather than confirmatory evidence.
Why a one-size mandate fails
Sridhar applied the framework to organizational AI strategies, warning that mandates often produce compliance without genuine acceptance. "Start from the workflow. Start from the job to be done," he said. For leaders, this means looking beyond login counts and completed training modules to understand how employees actually use AI tools-a core concern in AI for Management.
This approach aligns with principles of AI for Executives & Strategy, where mandates without buy-in often stall. Sridhar recommends identifying a defined operational need before pushing broad AI adoption. Acceptance varies by context: a person might embrace AI for routine administrative tasks while resisting its use in areas they consider higher-stakes, such as quality decisions or regulated documentation.
Adoption metrics alone-logins, completed tutorials-can hide discomfort or misunderstanding. Supplementing those numbers with questions about which tasks employees use AI for, where they remain uneasy, and what guidance they need lets managers tailor training to specific workflows rather than presenting AI as a single, undifferentiated category.
Why this matters for management
Managers who assume everyone experiences AI the same way will misread their team's readiness. The RISE framework offers a practical lens to segment users by their motivations and concerns. Instead of issuing a general directive to increase AI use, start with a concrete job to be done. Pair usage data with direct feedback, address anxiety rather than dismissing it, and design rollout plans that respect how attitudes shift from one task to the next. The goal is not just logged hours on a tool, but genuine, informed adoption that improves work without breeding resentment.
Your membership also unlocks:
