Robots and Ai Will Replace Most Human Work: a Discussion

Artificial Intelligence (AI) is a new technology to support human work on multi-area, it also could replace human on many physical works in society by mechanic arm or body. AI is defending as machine intelligence, which is intelligence exhibited by machines made by humans, it has similar functions with human minds such as learning. There are several of discussing points about AI replace human work is beneficial to society or not. Firstly, AI technology is still not stable and economic in most areas in recent years, which it costs too many recourses for a few achievements. Additionally, as a created program, AI cannot think like a human, which means it has no human nature and judge ability by self. However, AI is beneficial on the repeating work such as item handling and sorting, which increased the efficiency of some easy, high repetition rate mission. Lastly, the AI program is not successful in recently, further technology development would provide more chance and potential for developing AI programs.

First of all, AI programs are still in developing the process. Basically, AI is still a semi-finished product in recent years, the hard work of humans did not help the artificial program ‘intelligence’. Most released AI programs are still followed orders and calculated the each possible of steps, then start the work, it has no process for the AI “thinking” or “learning”, because those, AI is not completed. As an example, AI driving is a popular word around society. Normally it is known as a car receive the environment information by a variety of sensors it has, such as GPS, radar and sonar.(2) Thus, the AI program needs information input to the developed system, then calculate each possible for different movement, and output the steps to mechanical movement. It still followed orders in the process and spent time on information transport between sensors, main computer, and mechanical part, which are excess process compare to a human driver. Another example is Internet assistant, Siri from Apple Co. is a famous product. Siri is the commonest voice assistant program on Apple phones, which use a microphone to receive voice and use the internet database to find the best answer, then reply to the user with designed sentences. However, the assistant program cannot work without internet, which the database, also it may misunderstand users order or recognize multi languages in one sentence such as English character E and pronounce of One (Yi) in Chinese. There is no “learning” function of the product yet, except the record user’s daily information and gave designed suggestion. Not only above, AI program normally needs a supercomputer with high rate calculate-ability to help them collect data and choose the best answer, which has volume as a classroom, also need huge amount resources to work and maintenance. That is a high-risk investment before the AI program done some works that worth the machine it is based on. Thus, the AI program is not ready to be an important part of human life, it still uncompleted, also not cost-effective enough to be admitted by persons.

Additionally, AI programs are in trouble with “thinking”. As programs, they have no ability to judge right or wrong unless the developer gave them an exact answer. Even they have a big-database as support, some easy reading questions still led them to misunderstand. The main problem is AI programs we have is not like we thought, which could “think” like a human. Pronoun disambiguation is one example that AI “think” problem, such as “The city councilmen refused the demonstrators a permit because they feared violence. What does the word “they” refer to here? ”Humans could make decisions easily and fast, which AI did is “only a bit better than random.” Due to the natural language is elusive to understand, also has meaning in different environment or backgrounds, a designed program like AI is hard to recognise them well, which human could do simply. Problems also happened of AI using as a lawyer. The special programs may have a higher efficiency to censor case information and predict the possible in judgement, however, it still in trouble with cases which concern the interpretation of the law, such as “did someone Performance badly because he or she spent too long on wait?” or “a wrong judgement is made on purpose or not?” Those philosophical components will interfere judgements of the AI. In fact, not only the philosophical problems will bother the AI, but biased information also would lead AI to make a wrong judgement, which means AI is not reliable and fair enough if without real information input. Thus, AI programs cannot replace some human works on society because they are not ‘humanity’ enough, which they cannot think or feel like a real person, also will make mistakes that humans will not make.

On the other hand, even AI programs are still in developing and completing, they are still having beneficial in some areas with easy order but hard works, which reflected as robots. With a mechanical arm or body, those AI robots could carry heavy items or Multiple pieces of stuff. They also could protect humans in different work environments. In 2016, Amazon deployed about 45,000 robots in its distribution centres and warehouses. Those robots could pick up and carry items weight over 340kg. With the robots help, Amazon warehouses decrease much time from order to shipment in the warehouse. As the experts presented, robots replace humans work is a “general trends”, with the automation developed, more and more robots been used into society such as the Amazon warehouse robots to carry and transfer items, which increased the warehouse collection and classification efficiency. Another example is car manufacturing robots, which usually have a long metal arm for assembly jobs such as spot welding and painting. They improved quality and reduce warranty costs, also protect workers from dirty, difficult and dangerous jobs. The robots will not drop parts or handle them in a way that causes damage. They reduce waste previously caused by human error. Additionally, they protect workers from the hazardous environment, such as pouring molten metal in a foundry. The customer service area is also touched by AI programs. The increased amount of company websites and customer server centres are using customer service robots, they instead real humans behind table or screens, reply questions from customers, which are known as chatbots. Compare to persons, those chatbots could receive a huge number of questions and search the best way to deal with in seconds, then reply a high accuracy answer back to the customers. They would not be affected by any strong or violented language, which could protect the rights and interests of both sides in the business. Consequently, AI robots are in beneficial in some area to replace humans in society, especially for those job with hazards and high-repetitions.

Lastly, even AI using is increased recently, this technology is still uncompleted. Technology companies are interested in it as much as possible. They spend a huge amount of money and human recourses in developing AI programs for different areas to use. In 2014, Google Co. acquired DeepMind, which a British AI development company with £400 million. They train the AI program to learns the navigate 3D spaces as a human dose, this research is aiming to solve intelligence first, then everything else. With that amount of resources cost, Google has a completed team working on the AI developing, they guide the special program beats humans on games, sciences, and bring the data to products where they might be useful. Alibaba Co. led funding of $600 million for AI-driving development in 2018, which is the largest venture capital investment in AI developing area around the world. They use the foundation to support seven research laboratory, which aiming to solve machine learning, network security, natural language processing and more. Not only those mega companies, but AI study is also popular in universities. University of Otago cooperate The Law Foundation, they provide a three-year foundation-funded project in New Zealand. Because the project is involved in AI use in law and public policy implications, the university is trying to lighten the potential hazards and risks, maximise benefits by educations, laws and product standards. Therefore, as a project with highly concerned, AI still has many potential functions to discover by society. With more recourses input, humans developed it more completed, AI programs also obtained more possible to be used in society.

In summary, there is controversy as to whether or not AI will replace most humans work. Some argue that because the AI is still an immature technology, it needs a huge cost to support. further, AI has not real life, they cannot think and learn, also has no ability to judge rights or nor by self, which has too many unknown hazards. Conversely, other support because AI robots are using in some areas and gains beneficiate in recent years, they work with high efficiency and protect workers from hazards in different work environments. Finally, AI is still in the developing process, companies spend recourses for more abilities and efficiency, which provides more potential functions in the future. Therefore, it can be decided that AI will instead humans in most works. It is recommended that with the AI developing, humans will use it in more areas than recently, it also will save more time and cost for society in the future. 

Works Cited

1. Russell, S., & Norvig, P. (2016). Artificial Intelligence: A Modern Approach (3rd ed.). Pearson.
2. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.
3. Brynjolfsson, E., & McAfee, A. (2014). The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. W. W. Norton & Company.
4. Ford, M. (2015). Rise of the Robots: Technology and the Threat of a Jobless Future. Basic Books.
5. Brynjolfsson, E., & McAfee, A. (2017). Machine, Platform, Crowd: Harnessing Our Digital Future. W. W. Norton & Company.
6. Bostrom, N. (2014). Superintelligence: Paths, Dangers, Strategies. Oxford University Press.
7. Arntz, M., Gregory, T., & Zierahn, U. (2016). The Risk of Automation for Jobs in OECD Countries: A Comparative Analysis. OECD Social, Employment and Migration Working Papers, No. 189, OECD Publishing.
8. Brynjolfsson, E., Rock, D., & Syverson, C. (2017). Artificial Intelligence and the End of Work. MIT Sloan Management Review, 58(3), 93-99.
9. Manyika, J., Chui, M., Miremadi, M., Bughin, J., George, K., Willmott, P., & Dewhurst, M. (2017). A Future That Works: Automation, Employment, and Productivity. McKinsey Global Institute.
10. Frey, C. B., & Osborne, M. A. (2017). The Future of Employment: How Susceptible Are Jobs to Computerization? Technological Forecasting and Social Change, 114, 254-280.