Study Finds Wealthier Groups More Likely to Use AI, Raising Concerns Over Growing Social Inequality

In the evolving landscape of technology, artificial intelligence (AI) has woven itself deeply into the fabric of everyday life, yet this integration is generating a novel and insidious form of digital inequality, according to a groundbreaking study led by communication researcher Professor Sai Wang from Hong Kong Baptist University. Their analysis of a nationally representative sample of over 10,000 Americans highlights a divide not just in access to technology, but more critically, in awareness and familiarity with AI—factors intricately linked to socioeconomic status.

The study, published in the esteemed journal Information, Communication & Technology, challenges traditional notions of digital inequality, which often emphasize access and skill-based disparities. Instead, Wang and her colleagues reveal a nuanced digital divide where AI technologies operate ubiquitously but invisibly within everyday applications. People with higher education and income levels not only demonstrate greater AI awareness but are also more adept at recognizing AI’s presence across various platforms and understanding its implications. This phenomenon creates an uneven playing field in the digital era, where the advantages conferred by AI awareness accumulate to deepen social inequities.

A key conceptual advancement from Wang’s work lies in distinguishing between AI awareness and AI familiarity. Awareness is defined as the ability to recognize AI’s deployment in different contexts, while familiarity pertains to individuals’ perceived knowledge of AI irrespective of technical accuracy. Interestingly, the findings indicate that familiarity serves as a more robust predictor of AI awareness than actual AI use. This suggests that a subjective sense of understanding—feeling informed about AI—propels individuals to better identify when AI is at work, even if they are not direct users of specific AI technologies.

One striking implication of this study is how seamlessly AI integration in ubiquitous services such as social media algorithms, streaming platforms, and recommendation engines escapes the conscious recognition of many users. For instance, platforms like Netflix or Spotify leverage advanced AI recommendation systems finely tuned to individual user preferences, yet most consumers remain oblivious to this complexity. Rather than attributing personalized content suggestions to underlying AI, many perceive them as arbitrary or neutral, illustrating how AI’s hidden presence fosters unaware engagement.

This hidden AI integration presents a departure from conventional digital inequalities where users actively engage with technology. Instead, the AI divide functions subtly beneath user awareness, expanding disparities even in settings where physical access to technology is universal. Such invisibility complicates efforts to bridge the gap, as simply enhancing access to AI-powered tools does not equate to fostering meaningful awareness or understanding among marginalized social groups, particularly those from lower socioeconomic strata.

Wang and her team underscore the societal risks accompanying this divide. Individuals with limited AI awareness are vulnerable not only to missing out on opportunities but also to potential harms such as manipulation through AI-generated misinformation or deepfakes. Conversely, those with heightened awareness are better equipped to navigate AI’s benefits and pitfalls, highlighting the ethical imperative of widespread AI literacy to avoid exacerbating social marginalization.

The empirical backbone of this research lies in data garnered from the American Trends Panel conducted by the Pew Research Center. The research leverages the respondents’ education and household income levels as proxies for socioeconomic status (SES), revealing that education level exhibits a stronger correlation with AI use than income. This finding aligns with previous literature linking higher SES to greater digital skills and confidence in employing innovative technologies, factors that facilitate engagement with AI tools.

Perhaps most compelling is the unexpected emphasis on familiarity as a gateway to awareness. This psychological dimension implies that outreach focusing solely on providing access or technical tools may miss the broader necessity of cultivating an informed mindset toward AI. Educational initiatives must therefore pivot to demystifying AI concepts, clarifying its pervasive but veiled presence, and cultivating critical awareness among populations traditionally underrepresented in technology discourse.

To remediate this burgeoning inequity, Wang proposes multifaceted strategies including community-based workshops, public awareness campaigns, and the integration of foundational AI concepts into formal education curricula. Such programs should utilize accessible language and relatable examples to illuminate AI’s role in everyday life, particularly in lower-SES communities. Elevating baseline AI literacy will empower more individuals to recognize AI’s influence, enhancing their capacity to leverage the technology and guard against its potential abuses.

Moreover, AI literacy initiatives must prioritize training on identifying and critically assessing covert AI applications. Understanding how AI shapes information streams, personal recommendations, and decision-making processes can empower marginalized groups to reclaim agency in increasingly automated environments. Failing to address this awareness gap risks entrenching differential outcomes in sectors like employment, where AI-driven resume screening might favor candidates who understand and strategically navigate algorithmic biases.

Wang and her colleagues caution that while their study’s US-centric data offers valuable insights, cross-cultural variations in AI adoption and awareness necessitate further research. Notably, prior studies contrast nations such as South Korea, China, and Finland, which exhibit comparatively high AI awareness, with countries like the Netherlands where awareness lags. This geographic dimension highlights the global complexity of AI-driven disparities and the pressing need for inclusive, context-sensitive policy responses.

In sum, this pioneering study reframes digital inequality through the lens of AI, revealing how socioeconomic disparities manifest not only in tangible digital access but also in the subtle dimensions of awareness and familiarity. By exposing the hidden mechanisms of AI’s everyday integration and the unequal distribution of knowledge, the research charts a critical pathway toward an inclusive digital future, urging concerted efforts to democratize AI literacy and ethical understanding across society.