A future in which all cars are autonomous is going to have huge impacts on society. In some ways, this is not surprising: the advent of the automobile itself wrought huge changes in the way that our cities are designed, the way we work, and (unfortunately) the natural environment.
This future will bring many challenges, not least in understanding the ethical implications of AI. There also remain deep concerns that the world economy is not yet ready for mass AI implementation, since the technology is likely to eliminate many jobs, albeit while creating others.
On the other hand, a future dominated by autonomous cars will likely be a much greener, cleaner, more convenient one. It will also be much safer.
Before we look in detail at the ramifications of a future of autonomous cars, it's worth noting that this future might be some way off, and that it will not arrive overnight. There remain significant challenges to be overcome in rolling out truly autonomous cars.
Some of these occur at the technical level. In 2014, the Society of Automotive Engineers (SAE) defined six 'levels' of autonomy, from complete human control to complete autonomy. Between these extremes, there exist a range of differing levels of autonomy. At the moment, the most advanced cars on the market reach Level 4, 'pre-defined autonomy', which requires driver input in certain circumstances.
Moving beyond this level will be challenging, but not impossible. At the technical level, it's worth noting that the hardware to achieve complete autonomy is already available: the challenge will be the develop software that can safely drive autonomous cars in all circumstances. Nevertheless, given the huge rise in Software-as-a-Service (SaaS) companies offering machine learning and AI solutions, it is only a matter of time before this is realised.
Another challenge to the roll-out of autonomous cars, and one that is often overlooked, is the legal frameworks that will govern their use. At the moment, no government has put in place rigorous systems for controlling the usage of such technologies or – perhaps even more critically – for assigning blame if accidents occur.
All this said, once autonomous cars take over from their human-driven analogues, there are likely to bring some huge benefits to society:
All of these benefits outlined, it's also worth mentioning the risks that autonomous cars will bring. Or rather, the risk, singular: that they might be hacked.
One of the primary reasons why some industry groups and lawmakers are resisting the widespread adoption of autonomous cars is the security risk they pose. Spyware in the IoT, for instance, has been one of the major concerns of the last few years, and could easily affect autonomous cars.
Dealing with this security concern is not going to be easy, but some ways forward are visible. BlackBerry is a major developer of software solutions for autonomous vehicles, and the company has developed an advanced threat intelligence system that will also help to protect them. Using the same AIs that drive autonomous cars to improve their security is also feasible, and AIs are already being used to improve consumer VPN services that can block attempts to hack the control systems of autonomous cars.
It is almost inevitable that autonomous cars will completely replace human-driven vehicles within the next century. Though there remain some technical challenges involved in achieving this shift, the main resistance to this shift is likely to come from political actors.
Whilst it is encouraging that technical advances are being made in autonomous car technology, therefore, the most important work that needs to be done is in the social and political realms. Policymakers need to rapidly develop legal frameworks for working with this new technology, and governments need to step in to mitigate some of the risks. In this context, some have proposed that a U.S. Department of Cybersecurity could act as a centralised authority for reducing cyberattacks, both against existing targets, and emerging technologies like autonomous cars.