Conference Agenda

How can communications policymakers move beyond reactive, fear-driven responses to artificial intelligence—characterized by either uncritical adoption (FOMO: Fear of Missing Out) or reflexive prohibition (FOBMO: Fear of Being Made Obsolete)—toward stakeholder-informed governance frameworks that enable strategic, contextually appropriate, and democratically accountable AI integration?

The current AI policy landscape is dominated by two polarizing impulses. On one end, institutions rush to adopt AI wholesale to avoid falling behind, without deliberate evaluation of appropriateness or risk. On the other, legislators and community groups call for sweeping bans or moratoria, driven by fears of job displacement, loss of creative identity, and erosion of human agency. Both responses share a common root: they are shaped more by emotional reaction than by evidence of how AI is actually being used, experienced, and integrated by real stakeholders. This paper asks what a third path—grounded in empirical data and participatory governance—might look like.

This research employs a mixed-methods, evidence-based policy analysis approach. The primary empirical foundation is the Anthropic Interviewer dataset (Handa et al., 2025), which comprises 1,250 in-depth interviews with professionals across the general workforce, creative industries, and scientific communities. This unprecedented qualitative dataset—publicly released on HuggingFace for independent scholarly analysis—captures how workers across diverse sectors are actively integrating, adapting to, and feeling about AI in their professional lives.

The analysis proceeds in three stages. First, a systematic thematic analysis of the Anthropic Interviewer dataset is conducted, identifying patterns of AI use, emotional responses, and stakeholder concerns across professional categories. Second, these findings are mapped against current AI policy proposals and legislative actions at the federal and state level to identify alignment gaps—places where policy is responding to hypothetical fears rather than observed behaviors. Third, the paper synthesizes a stakeholder-driven policy framework using principles drawn from participatory governance literature and regulatory communications theory.

Quantitative survey data from the same study—including reported rates of AI-related time savings (86% of general professionals), satisfaction levels (65%), and anxiety indicators (55%)—are used to validate and contextualize qualitative themes. The divergence identified in the dataset between self-reported AI use patterns (65% augmentative) and observed Claude conversation data (47% augmentative) is treated as a methodologically significant finding that itself informs how policymakers should approach stakeholder consultation.

This research sits at the intersection of communications policy, science and technology studies (STS), and organizational communication. The primary disciplinary lens is communications policy analysis, examining how regulatory frameworks shape and are shaped by the communicative practices of stakeholder communities. Protection motivation theory, a threat response framework, is used for understanding how fear responses are amplified or moderated in public discourse—provides the conceptual vocabulary for diagnosing the FOMO/FOBMO dynamic.

This paper makes two novel contributions to the AI policy literature. First, it introduces and operationalizes the FOMO/FOBMO binary as an analytical framework for diagnosing the structural pathologies of current AI governance discourse. While scholars have noted polarization in AI policy debates, no prior work has systematically connected this polarization to well-established risk psychology constructs or applied them as a diagnostic tool for communications policymakers.

Second, the paper leverages the Anthropic Interviewer dataset as a novel empirical resource for AI policy analysis. This dataset is unique in the field: it is large-scale (1,250 interviews), professionally diverse, publicly available, and captures the qualitative texture of how workers actually experience AI integration. The finding that 69% of professionals navigate social stigma around AI use at work, and that only 8% of anxious workers lack any remediation strategy, directly challenges policy narratives that cast workers primarily as passive victims of automation.

The paper expects to demonstrate that real-world AI adoption patterns are substantially more nuanced than either enthusiastic or alarmist policy narratives suggest. The evidence indicates that most workers are not choosing between total adoption and rejection, but are actively using AI while negotiating contextually sensitive boundaries—often with vague or contradictory guidelines.This empirical reality should anchor governance frameworks rather than the extremes that tend to dominate legislative and regulatory debate.

Ultimately, the paper argues that moving from fear to informed choice is not merely a communications challenge—it is a governance design challenge. The tools to meet it, including scalable stakeholder interview methodologies and large-scale qualitative datasets, are now available.