Introduction
As artificial intelligence (AI) transitions from theoretical research to real-world deployment in traffic signal coordination, asset management, and autonomous systems, the focus often remains on technical feasibility or public acceptance. However, a critical yet understudied group sits at the center of this transition: the transportation professionals—engineers, planners, and policymakers—who design and manage our infrastructure. This paper investigates how these professionals perceive the impacts of AI, identifying distinct mindsets that could determine the speed and success of AI integration into transit systems.
Methodology
To move beyond simple averages and capture the nuance of professional attitudes, the researchers employed a two-step statistical approach on survey data:
- Factor Analysis: Fourteen attitudinal statements were condensed into three core "latent constructs": AI Efficiency Impact, AI Ethical Concern, and AI Equity Impact.
- Latent Class Cluster Analysis (LCCA): Unlike traditional clustering methods, LCCA is a probabilistic approach used to identify hidden subgroups (latent classes) within a population. This allowed the authors to profile specific segments of the workforce based on their underlying beliefs rather than just demographic data.
Data Collection
The study utilized a primary dataset from a 2023 survey of 354 transportation professionals in the United States. The sample was highly educated, with 71% holding postgraduate degrees, representing a mix of state DOTs, transit agencies, and professional organizations like the Transportation Research Board (TRB).
The Four Professional Mindsets
The LCCA revealed four distinct clusters of professionals, each with a unique outlook on the future of AI in their field:
- AI Neutral (51.4%): The largest group. These individuals are highly educated but report low personal knowledge of AI. They maintain a "wait-and-see" attitude, representing a massive segment of the workforce that remains undecided.
- AI Optimist (25.6%): Predominantly younger professionals (under 40). They hold strong convictions that AI will improve both operational efficiency and social equity.
- AI Pessimist (14.1%): Typically older professionals with moderate knowledge. They are skeptical of efficiency gains and harbor significant ethical concerns regarding AI deployment.
- AI Skeptic (8.9%): The smallest but most expert group. Paradoxically, these individuals reported the highest levels of AI knowledge but the lowest belief in its benefits, coupled with the highest concern for ethical risks.
Key Insights
The Knowledge Paradox
One of the study's most striking findings is that familiarity does not necessarily lead to favorability. The "AI Skeptics"—those who understand the technology best—are the most doubtful of its benefits. This suggests that for experts, a deeper understanding of AI may highlight the practical limitations and complexities of applying algorithmic solutions to messy, real-world physical infrastructure.
Efficiency vs. Equity
There is a clear consensus on "hard" metrics but a deep divide on "soft" ones. While 77% of respondents believe AI will improve efficiency and reduce costs, they are split on social outcomes. Approximately 40% of professionals fear that AI will exaggerate existing inequalities in transportation, and a staggering 88% agree that there is a limited understanding of AI ethics within the professional community.
The Role of Age
Age emerged as a significant predictor of attitude. Younger professionals are more likely to be Optimists or Neutrals, whereas older professionals are more likely to fall into the Pessimist or Skeptic categories. This generational gap suggests that as the workforce turns over, institutional resistance to AI may naturally decrease, though ethical concerns may persist.
Conclusion and Future Directions
The dominance of the "AI Neutral" group suggests that the future of AI in transportation is not yet written; successful early use cases could swing this majority toward optimism, while high-profile failures could solidify skepticism.
However, the study highlights a critical gap: the need for professional development focused on ethics and equity. Because even the most knowledgeable professionals express deep concern over these issues, technical training alone will not be enough. Future research should focus on qualitative interviews to understand the specific roots of expert skepticism and to design training programs that address the ethical pitfalls of automated transit systems.
While the study is limited by its US-centric convenience sampling, it provides a vital baseline for understanding the human bottleneck in the AI revolution.
Summary updated