Artificial intelligence and algorithm (AI) is transforming our lives and work, offering immense potential for productivity and innovation. However, it has become clear that AI’s benefits are not distributed equally and can worsen social and economic disparities, especially along demographic lines such as race.
Leaders in business and government are urged to ensure that AI advancements benefit everyone. Yet, each day reveals new ways in which AI contributes to inequality, often resulting in reactive and fragmented solutions, or sometimes no response at all. To effectively address AI-driven inequality, a proactive and comprehensive approach is necessary.
To make more equitable, policymakers and business leaders should acknowledge three key forces that contribute to AI-driven inequality. We propose a straightforward, high-level framework that encompasses these forces while focusing on the intricate social mechanisms that underlie AI’s creation and perpetuation of inequality.
This framework offers dual advantages: it is adaptable across various contexts, from manufacturing to healthcare to art, and it sheds light on the often-overlooked interplay of factors influencing the demand for goods and services, a significant channel through which spreads inequality.
Technological forces: Algorithmic bias
Algorithmic bias arises when algorithms make decisions that systematically harm specific groups of people. This bias can have severe consequences, especially in critical areas like healthcare, criminal justice, and credit scoring. For instance, a widely-used healthcare algorithm was found to significantly underestimate the healthcare needs of Black patients, resulting in inadequate care.
This isn’t merely unfair; it has deeply harmful implications. Algorithmic bias often emerges because certain populations are inadequately represented in the data used to train algorithms, or because existing societal prejudices are ingrained in the data itself.
While mitigating algorithmic bias is a crucial component, regrettably, it alone is insufficient to ensure equitable outcomes. Beneath the surface, intricate social processes and market dynamics come into play, creating a landscape of winners and losers that can’t be solely explained by algorithmic bias. To gain a comprehensive understanding of this unequal landscape, it’s essential to grasp how AI influences the supply and demand for goods and services, perpetuating and sometimes even creating inequality.
Supply-side forces: Automation and augmentation
AI often reduces the costs associated with providing certain goods and services by automating or enhancing human labor. Research by economists like Erik Brynjolfsson and Daniel Rock demonstrates that some jobs are more susceptible to automation or augmentation by AI than others. A notable analysis by the Brookings Institution revealed that “Black and Hispanic workers… are disproportionately represented in jobs at high risk of being automated or substantially altered.”
This isn’t due to biased algorithms but rather because certain jobs involve tasks that are more easily automated or are financially attractive for AI investment. However, since people of color often occupy these roles, the automation and augmentation of work through AI and digital transformations at large can lead to demographic-based inequality.
Demand-side forces: Audience (e)valuations
The integration of AI into professions, products, or services can influence how people perceive their value. In essence, AI also impacts demand-side dynamics.
Why AI-augmentation can reduce demand
Recent research has shed light on the reasons behind the reduced demand for various goods and services when AI-augmentation is involved. Our findings revealed that individuals often perceive professionals who offer AI-augmented services as having lower value and expertise. This devaluation of AI-augmented services was observed across a range of professions, including coding, graphic design, and copyediting.
However, it’s essential to note that people hold diverse views when it comes to AI-augmented labor. In a survey we conducted, 41% of respondents were categorized as “AI Alarmists” – individuals who expressed reservations and concerns about AI’s role in the workplace. In contrast, 31% fell into the “AI Advocates” category, enthusiastically supporting the integration of AI into the workforce. The remaining 28% were “AI Agnostics,” individuals who remained neutral and recognized both potential benefits and drawbacks. This diversity of perspectives underscores the absence of a clear, unified mental model regarding the value of AI-augmented labor.
While these results are based on a relatively small online survey and may not represent the entire spectrum of societal views on AI, they do highlight distinct differences in how individuals socially evaluate the use of AI and its users. These evaluations significantly influence their demand for goods and services, which forms the core of our continued research efforts.
Read our more blogs on retailinsights