The recent edition of Mobile World Congress stressed that the industry is in a state of flux, as many new technologies have yet to enter maturity despite their strong potential for disruption. In addition to 5G and the internet of things (IoT), artificial intelligence (AI) is one of these technologies whose usage is expected to grow to over USD36 billion by 2025. Application of AI will be wide-ranging across many sectors and industries, but the core of the technology remains the collection and processing of data from which an automated decision can be made, providing opportunity but also creating risk.
Artificial Intelligence and everyday use
AI was publicly recognized through a series of famous competitive wins against humans with Deep Blue beating the chess world champion in 1997, Watson winning Jeopardy in 2011, Deep Mind beating the Go world champion in 2016 and Libratus winning at poker in 2017. These events highlight the development of AI over the years with increased sophistication; an evolution that has progressed from singular task capabilities to more enabled capabilities, such as human-level abilities of perception, natural language and reasoning, and machine learning. With computers now capable of deep learning without the need for programming, it’s logical to assume that AI will continue to evolve, leading to the increased use of AI technology in different ways, across many industries.
Autonomous cars will only work with the help of AI. Cars will use multiple sensors to recognize objects, and the decision-making will be done through AI programming. There are several layers to the development, which includes the use of the network edge to make decisions quicker and with lower latency, as well as the use of machine learning. This will allow a fleet of vehicles to learn from a particular incident with one particular vehicle. The ability to collect and process data will be key, with NVIDIA already introducing that concept through its Xavier chip.
Telecommunications and Technology
Much of the consumer-facing AI has been performed through digital assistants. Apple Siri, Amazon Alexa, Google Assistant, Microsoft Cortana and the recently announced Aura service from Telefónica. The use of voice as a more intuitive interface is geared towards the collection of data, which will improve the service and allow companies to offer more personalized products.
Through AI systems, insurers will be able to predict risk using multiple data points and algorithms. Armed with this information, they will be able to determine the best premium pricing with greater granularity and have greater uniformity in decision-making.
The use of AI through automation and robots adds another layer to how machines can interact and make decisions. The use of AI can improve efficiency and production with better data analytics, allowing for greater visibility in terms of performance and maintenance costs.
Scouring through an abundance of legal agreements and documents is an arduous task. New digitization solutions exist that can easily scan and analyze these documents employing machine learning capabilities that can reduce the number of man hours by the hundreds of thousand per year. As a result, attorneys and other like-positions are at risk of being replaced by smart technology.
Artificial Intelligence comes with risks
Those five examples highlight the centrality of data in an AI system and the various ways it can be used in different circumstances. However, there are some key risks to consider with the growing usage of AI.
AI is often seen as a threat to the future of work, as greater reliance on the technology could lead to the elimination of some jobs across multiple sectors — from truck and taxi drivers being replaced by autonomous vehicles, new software programs impacting traditional attorney functions, and factory workers displaced as a result of automation. This has led to the development of new paradigms, such as taxing robots and universal income to replace funds lost by the lack of work. AI can also act as augmentation, meaning it will improve current capabilities while leaving humans in charge. While a robot will develop greater capabilities, there are certain applications it will never be able to master, such as inventiveness, assuring that human labor will not be completely overtaken.
Contrary to the vision of neutrality and objectivity that the technology industry may promote, there is always an element of bias when writing an algorithm. If AI is used for decision-making, such as predicting risk for insurers, it could lead to a biased decision for a particular individual policy, with the added difficulty of not being able to pinpoint where the bias comes from. The use of technology to make a decision gives an element of certainty to that decision, but conclusions drawn from machines should never be relied upon as being flawless or absolute. As the inventor of machines, human reasoning should not be cut from the equation decision-making.
Using machines to complete tasks traditionally performed by humans leads to the issue of responsibility. Who bears the burden of liability if something should fail? The machine manufacturer or the owner? In the automotive industry, autonomous vehicle manufacturers have said they will accept liability in the event of an accident. This is a major flip in the current system of the insurance sector that attributes responsibility to the driver. The legal issues surrounding liability of machines will continue to increase as AI develops.
The increased collection and processing of personal data as part of AI to make machines more intelligent raises the question of who should control that data, whether consent is required, and how it should be given. It will become easier to identify an individual, even with minimal data points, and many users may not be aware of the amount of personal data they give out in the course of their online activities. Too high a regulatory burden might hinder the progress of the market, but the lack of a strong framework may also lead to abuse.
AI is driven by data, as decision-making algorithms will get better through greater collection and processing, and regulation will have a clear impact on what can and cannot be done with data. The European Union has been the most active on the subject, proposing to give robots legal status as electronic persons, under the guise of ethics and responsibility. However, the more recent General Data Protection Regulation, due to become effective in May 2018, gives individuals the right to obtain an explanation from algorithm-made decisions and opt out. Global regulatory fragmentation can become a key risk when looking at the implementation of AI worldwide.
Click here to learn more about comprehensive cybersecurity from Willis Towers Watson.