Artificial intelligence negative impacts

 

Artificial Intelligence ai Future Society Prospects and New Risks

From chatbots to self-driving cars, AI technology is transforming society as it spreads widely. With the development of artificial intelligence (AI) technology, automation is promoted in the manufacturing and service industries, and productivity and work efficiency are expected to improve. According to Accenture research results 1) , artificial intelligence technology is estimated to improve the profitability of companies in 16 industries in 12 countries by an average of 38% by 2035.

Existing AI applications showed cognitive abilities in some specific areas, such as driving a car, solving puzzles, or recommending products, but the area has since expanded and is being realized in many other industries. expectations are increasing day by day. Until now, AI has been passive in solving given problems by interpreting given data, but with the advent of deep learning, it has moved to the stage of 'Strong AI', which has human-like intelligence and makes decisions like humans. entry is progressing rapidly.

The development of artificial intelligence may improve the quality of life and productivity, but it may cause serious social and ethical problems if it becomes out of control or is abused by a group with a specific purpose according to the advancement of technology and the improvement of the level of automation. there is. According to the 'Allianz Risk Barometer 2018', the impact of new technologies such as artificial intelligence has already been classified as the seventh corporate risk ahead of political risks and climate change.

1.Impact of AI in Major Sectors Artificial intelligence AI

·      economic impact

AI is expected to increase GDP per capita by performing existing tasks more efficiently and especially by compensating for the decline in the labor force caused by an aging population, but will potentially lead to widespread unemployment by displacing many low-income jobs. The World Economic Forum predicts that within five years, AI robots will eliminate about 5 million jobs worldwide or replace more than 45% of them. Discussions about a new robot tax in 2017 have been actively developed out of concerns about a significant increase in human unemployment by replacing production by robots. The European Parliament, which first led the discussion, adopted a resolution against the introduction of the robot tax after discussion, but recognized the need for legislation regarding the ethical and legal responsibility for the development and spread of robots. This was passed. Currently, in Japan, there is an active discussion about the form of taxing corporations and individuals using artificial intelligence, respectively, corporate tax and income tax.

·      impact on mobility

AI self-driving cars have been attracting attention for their efficiency and the fact that they can greatly reduce the possibility of traffic accidents caused by human error. It is estimated that autonomous vehicles can help reduce the number of traffic accidents by up to 90%, but on the other hand, recent incidents related to autonomous vehicles have raised questions about who is responsible for accidents and the ethical standards for designing autonomous vehicles. is raising In other words, according to the classification of autonomous driving stages, it is possible to interpret that the manufacturer may be held responsible for accidents caused by defects inherent in autonomous vehicles, where there is little room for driver negligence.

·      Health-related impacts

Among the fields of application of artificial intelligence, healthcare is a field with high growth potential and is highly valued in terms of social benefits. The healthcare industry is expected to have the highest CAGR of 60.3% among various application fields of the overall artificial intelligence market. As the medical cost burden increases due to an aging population worldwide, faster and cheaper medical services are required, and artificial intelligence technology is expected to create new value in the healthcare market. The use of artificial intelligence makes it possible to identify rare diseases that are difficult for doctors to detect by analyzing vast amounts of medical data. In addition, it is expected to be able to care for patients in areas with underdeveloped medical infrastructure through an artificial intelligence remote reading system.

However, on the other hand, there are concerns about the use of patient personal information when AI uses medical records to research new diseases, raising the need to better manage personal data privacy and personal information.

·      environmental impact

Artificial intelligence is already being used usefully in the field of environmental science to protect the earth, such as climate change and the protection of various species. AI with visual information processing capabilities can predict the weather and increase productivity. It is possible to predict the weather by looking at the amount of clouds or adjust the value of agricultural products by predicting crop yields from photos of the farm taken by satellite. that is an example In addition, it is expected that the disaster damage caused by climate change can be minimized by predicting temperature changes on a daily and weekly basis and deriving vulnerable points through deep learning. Artificial intelligence is also being used to monitor deforestation. The World Resources Institute is partnering with a big data startup to monitor and predict deforestation based on tens of millions of high-resolution satellite images.

In addition, by utilizing artificial intelligence technology, smart cities and smart grids2 ) are expected to reduce gas emissions, replacing areas that are difficult for humans to do, such as placing AI robots in harsh environments exposed to toxic or harmful substances. expect to be able to

2) New risks posed by artificial intelligence AI

·      business risk

As robots replace human tasks, companies will face a new responsibility scenario in which the subject of responsibility can change from humans to machines. If an accident occurs due to an artificial intelligence making a wrong decision, the AI ​​cannot be held responsible because it has no legal personality, which raises the question of who is legally responsible among the creators, sellers, and users. Countries around the world are rushing to enact legislation, but discussions about the legal relationship and attribution of responsibility for the use of artificial intelligence and robotics are still in their infancy.

While AI can be used to detect and prevent cyberattacks as AI advances at a rapid pace, the reverse is also possible. There is a possibility that users with malicious intentions will launch automated hacking attacks, and there is also a possibility that AI will be abused to increase malicious cyber behavior and spread and spread fake information impersonating humans in a future society where interconnectedness increases. . As a result, even minor, unintentional errors caused by AI can lead to huge accidents, which can damage the company's reputation and reduce profits.

·      Collapse of the existing labor market

With the rapid development of artificial intelligence, concerns about future jobs are rising, causing chaos in the job market. While the prospect that AI will replace a significant portion of the world's jobs is widespread, there is also a strong opinion that technological innovation has only changed the form of jobs, but has never reduced jobs. Jobs that are simple and repetitive and require relatively little communication with people, such as courier service workers, environmental cleaners, and accounting clerks, are highly likely to be replaced by automation.

Bill Gates, U.S. Senator Bernie Sanders, and British Labor leader Jeremy Corbyn have argued that the use of robots to replace humans should be taxed at the level of income tax. The rationale for the robot tax they argued was that a government-level program should be created to support workers who lost their jobs and to retain some of the jobs that would be rapidly disappearing due to artificial intelligence automation. The introduction of the robot tax was rejected on the grounds that the robot tax hinders innovation and that the use of robots provides new jobs. It can be said that a stroke was drawn.

·      non-compliance with regulations

As the interconnectivity increases due to the use of advanced technologies such as artificial intelligence, convergence with other technologies is also actively occurring. Therefore, the traditional method of regulation is ineffective due to the high uncertainty of risks related to new technologies and the uncertainty of regulatory effects. In addition, these regulations are more likely to hinder the development of the market related to new technologies. In addition, the lack of expertise of regulatory agencies may limit the ability to access information and deepen information asymmetry.

·      responsibility

With the advent of artificial intelligence, as the subject of action moves from humans to machines, it is highly likely that responsibility will be transferred to manufacturers or suppliers. AI is not yet a person (corporate) who is the subject of legal rights or obligations. Therefore, when it is impossible to hold AI legally responsible, it is not easy to determine who will be held responsible if the natural person (corporate) involved in it is held responsible. It is still controversial as to how far the manufacturer should be held responsible if the driver or pedestrian is injured due to a malfunction of the autonomous vehicle, and what priority should be applied to the autonomous vehicle when an accident occurs while autonomous driving is unavoidable. am. In addition, the burden of the cost of an accident is a controversial issue, which can lead to confusion about who will pay the cost of the accident among stakeholders such as program developers, machine manufacturers, owners, and designers when a robot malfunctions. Even in the case of robotic doctors, disputes may arise over who is responsible for the death of a patient due to a misdiagnosis.

In addition, if there is a possibility of a large-scale accident due to an error in autonomous driving technology or external hacking, a large-scale recall may occur in several manufacturers and countries.

·      implication

At present, it is difficult to accurately predict the future of artificial intelligence. Although artificial intelligence is spreading rapidly, it will take a long time to completely change almost all fields. However, if the AI ​​in the past only supported human judgment, it is expected that AI in the future will make the same level of judgment as humans. Therefore, it is time to think deeply about the social ripple effect that artificial intelligence will bring and to discuss in-depth how to improve the problems caused by it. In the insurance industry, the introduction of artificial intelligence has already started in earnest in each area of ​​the insurance industry’s value chain, such as consulting through chatbots, asset management and insurance claims using AI, and insurance premium calculations using driving habit data to automate business processes. The efficiency of work performance is increasing. However, on the other hand, the effects of AI also coexist with light and dark, such as cases in which artificial intelligence replaces functions previously handled by designers is increasing. It is time to prepare for the counterattack of new technologies by preemptively thinking about changes in the industrial structure in the artificial intelligence society to come.