Bringing Equity in Mobility
Published in Parking & Mobility (May 2024)
Every new technological cycle leads to greater advancements but also brings significant disruptions to society’s social fabric. From the Industrial Revolution to the Digital Revolution, we have seen how, on one hand, technologies have uplifted societies worldwide; on the other hand, although unintended, they have created ever-widening social divides.
Now, we are at the dawn of another technology revolution: the Artificial Intelligence (AI) Revolution. This time, it is going to be different. The underpinnings of this technological revolution are very different.
In the past, different factions of society adopted every new technology at different levels, leading to significant advancement for some and further marginalization of the disadvantaged.
Three factors explain this disparity in the adoption of any new technology, and these factors can make it different this time around with the AI revolution.
- Usability gap. Unlike in the past when human-to-technology interfaces were built keeping common users in mind while disregarding the disadvantaged, the new interfaces for AI are going to be natural human expressions, including natural languages.
- Content gap. In the past, technology was merely a tool. The value one derived from using that tool varied with the skills of the person using it. The self-generative capability of AI changes the status of technology in this new cycle of technology revolution, from a “passive tool” to an “active and participating companion.”
- Innovation gap. In the past, due to high barriers to technological innovation, primary drivers for innovation were misaligned from the priorities of social good. It required significant resources, profit-driven entrepreneurs, or heavily funded public bodies to bring innovation. When AI takes over the role of the innovator, then the “profit for few” will give way to the “social good for all” as the primary driver of innovation. Of course, that greatly depends on the social values that we impart into the AI training models now while AI is in its infancy.
AI Makes the Technology Framework for Equity in Mobility Implementable
But how will this AI revolution change technology adoption, leading to social good in mobility?
In the May 2022 edition of Parking & Mobility magazine, I wrote an article proposing a comprehensive technology framework for defining adaptive public policies for social equity in mobility. Since then, various public bodies at different levels have done a lot of policy work. There has also been widespread recognition that solving social inequities is not just the government’s responsibility.
However, the technology framework defined in my earlier article was a resource-intensive proposition for any organization, whether public or private, to bring to fruition. Conceptually, the framework is a complete solution to the social equity in mobility problem, but since it was built upon pre-AI technologies, the framework suffered from the above-mentioned gaps in usability, content, and innovation, making its implementation very prohibitive.
Let’s look at each of the five layers of the framework—data collection, data privacy, quantification, policy definition, and policy execution—and see how AI bridges the above-mentioned gaps and makes the overall framework more implementable when applied to them.
Human-generated data collection is essential in any analytical model to effectively understand and solve social inequity problems. A data collection framework severely fails to collect human-generated data without natural human expression and natural language interfaces. AI bridges this usability gap with its natural interface languages.
A data privacy framework is essential for the overall trust in the system. Any leakage of private data can lead to long-lasting mistrust in the system and future impediments to cooperation and data sharing by individuals and organizations. AI is unparalleled in identifying private information patterns and automatically self-generating anonymization. AI efficiently bridges this content gap.
AI also bridges the content gap in the Quantification framework. With its sophisticated pattern recognition and content generation capability, AI can very efficiently enable granularization and localization of large and diverse datasets, leading to the analytics down to recognizing actionable patterns of information.
A policy definition framework requires not only very specialized data science skills but also strong domain expertise in policy making. Moreover, both technical and policy experts must work hand in hand to perform analysis and make predictions. Typically, there is a significant communication barrier between the two experts due to the language of their respective professions. AI bridges the usability gap by learning from policy experts and bridges the content gap by applying social AI models for predictions and performing the impact analysis of their predictions.
Finally, the policy execution framework is where bridging the innovation gap is most crucial. Significant resources are required to roll out social equity policies. With very few financial rewards to profit-seeking entrepreneurs and cash-strapped government bodies, traditional drivers of innovation are absent for individuals and organizations to execute these social equity policies. This is where AI becomes the innovator and flips the drivers from the “profit for few” to the “social good for all,” thus bridging the innovation gap.
Conclusion
The AI Revolution will be very different from any other technology revolution. It will eliminate the age-old fallacies of technology innovation, which have always led to unintended harm to the social fabric of society through the widening of the social divide and the irreparable marginalization of the disadvantaged.
Of course, that greatly depends on the social values we impart into AI training models. If the foundational models are trained on value systems, social inequities must be addressed first for society to thrive. Then AI will balance and maximize social good in its every prediction and prescription.
