Autonomous vehicles are coming. Recent company reports and surveys have revealed that tech giants, including Alphabet, Waymo, and Uber, as well as old-school car manufacturers like General Motors, have reported and demonstrated their plan to launch self-driving cars (Chan, 2017). In particular, Waymo, one of Google’s sister firms has set to launch its first autonomous car service. Using artificial intelligence, the company has so far produced prototype vehicles, which engineers consider to be at their advanced stages. However, some researchers and stakeholders from different sectors, including the technology, as well as automotive industries, argue that tech firms have only oversold the future attached to the self-driving cars (Johnsen et al., 2017). In their view and findings, these skeptics assert that driverless vehicles would not possess a variety of desirable qualities of the present-day automobiles, except being environmentally friendly. For instance, autonomous cars are most likely to be too slow to rely on and less smooth than people have been made to believe. Given skeptics tend to come to a wide range of disciplines, including security and safety agencies, each has presented their bear cases against the dangers and potential negative effects of self-driving cars.
Human vs. Machine Intelligence
The first bear argument against autonomous vehicles is that computers cannot still match human beings when it comes to smartness, commonly known as intelligence. While self-driving cars view the whole process of driving as one of the playing Go activities, opponents view this viewpoint as misleading and falsified (Wang & Li, 2019). On individual tasks, computers and associated technologies can outperform humans. Some of these activities comprise playing Go and the identification of a picture or related objects. However, opponents of self-driving cars argue that, despite these computer capabilities, people should make generalizations. In recent essays written and published by Rodney Brooks, one of the legendary roboticists and researchers of artificial intelligence, tech giants and likeminded car manufacturing firms should no place much emphasis on the short-term viability or benefit of self-driving cars (Faisal et al., 2019). Having directed MIT Computer Science, as well; as Artificial Intelligence Laboratory, Brooks is recognized for his arguments against relying on edge cases, such as unusual circumstances autonomous cars would have to handle, to determine the future car’s viability.
Delegate your assignment to our experts and they will do the rest.
Although current developers have argued that their guiding principles for the self-driving cars are both appropriate and would be subjected to frequent updates if the need arises, computer scientists and roboticists have projected the existence of perceptual challenges. Through artificial intelligence systems, deep learning networks, and more automated reasoning, autonomous car developers have attempted to solve perceived challenges (Faisal et al., 2019). However, Brooks and remains convinced through firsthand experience with computers and artificial technologies that people would want their cars to possess human qualities, more intelligent than a computer, as the only way to appropriately identify and handle the various edge cases. In as much as self-driving cars are set to replace human drivers from the roads, the process of supplanting humans would happen gradually, taking several years from now (Chan, 2017). Consequently, the success and guaranteed viability of autonomous cars would depend on the developers’ ability to design these vehicles in a manner that they match human smartness.
Vulnerability to Hacking
Besides the self-driving car developers’ failure to prove that these vehicles would operate like or surpass the driving capabilities attached to human drivers, autonomous devices, including cars, are susceptible to hacking. Typically, any given self-driving car uses advanced artificial intelligence technologies, which means information technology serves as the key factor. Recent studies have established that cybersecurity measures, such as anti-hacking tools or techniques, are less likely to be useful when it comes to protecting autonomous cars from black hackers (Gruel & Stanford, 2016). Given the high-level vulnerability level, it can take more time to solve the car data security; hence autonomous introduction delay completely. According to one of the acclaimed self-driving car security specialists, Tim Mackey, there is a data security threat already in the computer industry where the probability of hacking is fast spreading to autonomous cars (Fleetwood, 2017). The data security threat is yet to receive the much-needed proper solution, and this, in reality, has and continues to attract more danger on self-driven cars than assumed on the Internet.
Safety
The technology on the autonomous vehicle is at its initial stage where the physical field tests are being carried out in various addresses to ensure their effectiveness and public safety, at this stage, crushes have been reported in US public road which brought a significant concern on whether they actual safety standard meets the proclaimed design safety. As compared to commercial aeroplane autopilot that is considered safe, the autonomous software relies on machine learning algorithms which are harder to test since they are made to be self-determined from the recorded data (Faisal et al., 2019). These data cannot be uniform from various roads, and this can lead to wrong safety analysis hence catastrophic events in self-driving cars. As far as safety is concerned, people assume the rule of “try not to hit something” as the safety, but in reality, the safety rule should ensure that the machine operates precisely as per the design parameters. Therefore, with self-driven cars, security remains an important issue since the software workability is not 100% accurate.
The decision of whether self-driven cars should be allowed on the public roads remains with the regulators, where they will make decisions based on the proofs by the firms manufacturing self-driven vehicles. So far the safety evidence by Waymo, a self-driven car manufacturing company, shows that though simulations have met some safety thresholds given the deaths which have been caused by self-driven cars, the safety standards thresholds are supposed to be set higher to avoid more deaths from self-driven cars. The need for such sophisticated security measures revolves around the fact that data lapses, identity left, as well as hacking have all failed to bar people from serving active consumers of the Internet. In most cases, online users tend to overlook these problems (Wang & Li, 2019). However, all the responsible stakeholders, including developers, should acknowledge and appreciate the fact that self-driven cars can potentially pose a far greater danger than surfing the Internet and being cyberbullied.
Transportation Service versus Reality
Available empirical evidence has shown that most companies working as developers of self-driving cars remain focused on earning revenues from the autonomous vehicle service. In this respect, users are expected never to own a car ( Gruel & Stanford, 2016 ). People, In the near future, would not only rely a great deal on robo-cars for their rides but also trust their movements and right schedules on Lyft, as well as other providers, such as Waymo. Equally important, the current transportation-service firms, which are yet to commit their resources to the development of self-driving cars, would not be in a position to realize any benefits or profits from operating their businesses. In this sense, these companies would lack the much-needed capacity to gain and maintain a competitive advantage in the market (Fleetwood et al., 2017). Moreover, another cost that is most likely to come as a result of a world characterized by autonomous vehicles involves the removal of human drivers from the equation of the transportation industry. Although a driverless world would save on expenses incurred by car service providers, the cost of readmitting these drivers to their respective countries’ labour system would be costlier in the long run.
Apart from the far-reaching negative economic effect of developing, launching, and formalizing the use of autonomous vehicles, recent studies have shown that the equipment necessary for converting physical reality and associated elements into data would be extremely expensive. In particular, Hancock, Nourhakhsh, and Stewart (2019 ) report that consumer vehicles, for safety reasons, would require each of the developers to install advanced and effective consumers, as well as lasers, which, preliminary findings have shown to be considerably costly. On the same note, the whole process of maintaining these installations and associated gadgets has been projected by economists to be prohibitively expensive. Concisely, robo-cars are most likely to become worse car-service providers than Uber.
References
Chan, C. (2017). Advancements, prospects, and impacts of automated driving systems. International Journal of Transportation Science and Technology, 6 (3), 208-216.
Faisal, A., et al. (2019). Understanding autonomous vehicle: A systematic literature review on capability, impact, planning, and policy. The Journal of Transport and Land Use, 12 (1), 45-72.
Fleetwood, J. (2017). Public health, ethics, and autonomous vehicles. American Journal of Public Health, 107 (4): 532–537.
Gruel, W. & Stanford, J. (2016). Assessing the long-term effects of autonomous vehicles: A speculative approach. Transportation Research Procedia, 13 , 18-29.
Hancock, P., Nourhakhsh, I., & Stewart, J. (2019). On the future of transportation in an era of automated and autonomous vehicles. PNAS, 116(16), 7684–7691.
Johnsen, A., et al. (2017). Literature review on the acceptance and road safety, ethical, legal, social, and economic implication of automated vehicles. Technical Report . https://www.researchgate.net/publication/325786957_D21_Literature_review_on_the_acceptance_and_road_safety_ethical_legal_social_and_economic_implications_of_automated_vehicles/citation/download
Wang, S. & Li, Z. (2019). Exploring the mechanism of crashes with automated vehicles using statistical modeling approaches. PLoS ONE, 14 (3), e0214550.
