Self-Driving Cars Have A Long Road Ahead (Part 2)
While tech companies race to make cars more autonomous, the technology is still unsolved, but exposes us to major regulatory, geological and environmental upheaval.
The 21st century ushered in the first realistic roadmap towards achieving Autonomous Vehicles (AVs). The concept first appeared in science fiction as early as 1935. Since then, AVs have transitioned from science fiction to reality, with companies like Google and Tesla at the wheel. The quest for full vehicular autonomy began with rudimentary systems like adaptive cruise control and lane-keeping assistance, gradually evolving into more advanced semi-autonomous systems like Tesla's Autopilot and Full Self-Driving suite. Yesterday, as of writing this, a two-year investigation led to Tesla recalling over 2 million vehicles to address Autopilot safety concerns.
In my previous letter, I explored the role of human error in road accidents, with over 1.5 million injury crashes in the US annually. It highlighted how self-driving cars stands to reduce these incidents, alongside the economic and social challenges they present, such as job displacement and retraining needs. Reflecting on the history of the automobile, the letter addressed the impact of urban development and environmental issues, framing autonomous vehicles as a nuanced solution requiring significant infrastructural and cybersecurity considerations.
The focus of this piece shifts from science fiction to reality for self-driving cars. This explores their evolution from basic automated systems to advanced semi-autonomous technologies, addressing societal needs for updated infrastructure and legal frameworks. It also explores the significant environmental impacts, especially in energy consumption and emissions from these vehicles and their AI systems.
The transition towards self-driving cars is not solely a technological endeavor but a societal one too. It requires significant alterations in infrastructure and legal frameworks. Modern cities are now contemplating the integration of smart infrastructure to accommodate autonomous vehicles. This includes the development of intelligent transportation systems, real-time traffic monitoring, and communication systems that allow vehicles to interact with each other and the surrounding environment. This change in the demands of contemporary society necessitates education on these topics too.
With that, we can look at the “five levels of driving automation” and see whether tech companies are more bark than bite when it comes to their progress. While I’m not a fan of a future where we rely on companies like Google and Tesla for transport, the benefits, as you automate the driving process, are hard to ignore.
Tesla, a market leader in electric and AVs, has equipped well over 800,000 vehicles with the standard Autopilot feature. Tesla has made a robust attempt at reducing the driver's workload. However, for an additional cost, Tesla owners can upgrade to what is dubbed Full Self-Driving (FSD), a feature promising a closer step towards autonomous driving. While the technology is impressive, empowering vehicles to maintain lane discipline, execute lane changes, and maneuver into and out of parking spots autonomously, it still only reaches Level 2 Automation. The driver must remain vigilant and use intervention when necessary. All Tesla vehicles manufactured since 2019 are hardware-ready to be upgraded to these higher levels of automation, or so the company claims.
Companies like Waymo (owned by Google/Alphabet) have started testing Level 4 autonomous vehicles in limited areas in the US, as well as Chinese tech companies, who have permission to operate fully driverless robotaxi services in some parts of of Beijing. In Europe, companies like Volvo and Mercedes-Benz are workings towards Level 3 and 4, though widespread deployment of ghost taxis is still in nascent stage.
Despite the promise of safety over human drivers, there have been fatal accidents involving autonomous vehicles. With the advent of autonomous vehicles, the legal landscape is shifting from driver liability to manufacturer liability. States and countries are scrambling to develop frameworks that govern the use of autonomous vehicles, ensuring safety and accountability. In the US, 29 states have introduced legislation concerning autonomous vehicle. Still, many open-ended questions remain about liability, insurance, and data privacy.
May 2016 bore witness the first passenger causality in an AV. This involved a Tesla in autopilot mode colliding with a large truck it failed to recognize as an obstacle. Then, two years later, a pedestrian was fatally struck by an Uber self-driving car that failed to stop—marking the first pedestrian causality. Both tragedies emphasized the crucial need for autonomous systems adept at recognizing and reacting to unforeseen road scenarios to safeguard both occupants and pedestrians… for the string of AV-related accidents didn’t stop there. Since then, Tesla has hundreds (nearly thousands) of accidents related to their self-driving technology. Then again, humans have always been a major threat to their fellow drivers on the road.
A report by IDTechEx revealed a startling statistic: 99% of autonomous vehicle accidents were triggered by human errors. Over-reliance on AV technology is another human error that can and will occur. Assumptions like an autonomous car halting timely could prompt risky behaviors like stepping in front of a moving self-driving car.
“Technology tends toward avoidance of risks by investors. Uncertainty is ruled out if possible. People generally prefer the predictable. Few recognize how destructive this can be, how it imposes severe limits on variability and thus makes whole populations fatally vulnerable to the shocking ways our universe can throw the dice.”
- Frank Herbert
Right now, China and the US are leading in AV innovation. The US has over 50% of its cities preparing for AVs. The Netherlands and Singapore are close behind, with well-established infrastructure and legislative frameworks that foster the adoption of AVs. The technological is always improving and consumer acceptance is growing, marking a significant shift towards a new era of transportation.
Since 2020, China's national strategy on autonomous vehicles took a significant stride. Various governmental bodies joined forces to outline a roadmap for the innovative evolution of this technology, setting ambitious targets for 2025. The plan includes mass-producing Level 3 vehicles and introducing Level 4 vehicles to the market in selected scenarios. Among Chinese consumers, acceptance of AVs is relatively higher compared to their Western counterparts. The burgeoning electric-vehicle (EV) market in China, more adaptable to autonomous functionalities, has positively influenced consumer perceptions toward autonomous driving. It also helps that China has access to the 70% of the world's known rare earth element reserve.
Globally, the journey toward full vehicular autonomy has been bumpy. Accidents involving autonomous vehicles have caused wide concern. There are many ethical quandaries like accountability during accidents, data privacy, and job displacement potential are among the pivotal concerns surrounding AVs. Different countries exhibit varying regulations and testing guidelines for AVs. In the US, the National Highway Traffic Safety Administration (NHTSA) outlines guidelines for automated driving systems. Conversely, the European Union operates under a framework provided by the United Nations Economic Commission for Europe (UNECE), showcasing the global effort in navigating the legal landscape of autonomous driving.
Regions endowed with the resources to invest in smart infrastructure may widen the technological divide. The disparity in the adoption of AVs between the Global North and the Global South will have far-reaching effects, as with other issues caused by technological inequity. Furthermore, the geopolitical arena could become even more precarious- the rare earth minerals needed for AVs and computers in general may lead to something like the Middle Eastern oil crisis one day, with countries rich in rare earth minerals, such as Bolivia, Chile, Argentina, Mongolia and various parts of Africa. The tension between Chinese and Western coalitions will escalate and many smaller developing nations will suffer for it.
In the next part of this series, our destination takes us deeper into the entangled concerns for our society, environment and ethics. In the mean time, why not learn more about human-AI centaurs and their unique advantages over both AI and humans alike?