Author: Prof. Dr József Hajdú, University of Szeged, Faculty of Law and Political Sciences, Department of Labour and Social Law

This is a translated article. Original language is Hungarian. The translation was prepared on behalf of the Hungarian EPALE National Support Service.

Introduction

It is a common but true statement that artificial intelligence (AI) is constantly transforming everyday life, including the world of work. One interpretation says that AI operates along the three values of cost (reduction), quality (improvement) and capacity (increase), and will or may gradually replace people in the labour market. Conversely, it is also argued that AI is likely to create new jobs, reinforcing the need for a human workforce, which requires new knowledge and skills from workers. Perhaps never before has continuous and lifelong learning (LLL) been so timely.[1],[2]

The jobs disappearing with the mass application of AI and robots will directly worsen the labour market prospects of people, increasing the chances that a second wave of digi-technological unemployment(DTU) will come. DTU means that AI replaces human labour in such a way that workers who are laid off due to AI cannot enter (as entrants) or return (as unemployed) to the traditional labour market at all. The classical active labour market instruments (e.g. retraining, etc.) are not able to fulfil their original role.

The example of the dark factory: Advances in automation, robotics, artificial intelligence and machine learning have led to autonomous and self-service production lines and warehouses. These are called dark factories, where humans no longer work, only robots. According to Guy Courtin, vice president of industrial and advanced technology at Tecsys Inc., a digital supply chain design specialist, Machine Design, the advantage of fully automated dark factories is that “Robots don't need a break or get tired. The plant can be operated up to 24 hours a day, 7 days a week and all year round, except for maintenance. Robots do not need heating in winter or cooling in summer. Also, since they can see in the dark, lighting is not a prerequisite. Moreover, certain safety measures can be ignored, they don't need to go to the toilet, and they don't go on strike.” Jobs that are dangerous, repetitive and tiring can benefit most from this phenomenon.[3]

Another human rights approach is that AI and robotisation ultimately undermine the right to work as a fundamental human (social) right. If there is no human labour, then there is no reason to talk about the right to work. This is problematic because the right to work essentially has two dimensions: 

a) on the one hand, the right of people to participate in the productive and service activities of society, i.e. no one should be excluded from the labour market, and if excluded, should be re-integrated as soon as possible, and 

b) on the other hand, the right to a decent income from work which provides an adequate standard of living for oneself and one’s family members. If robots massively and irreversibly displace humans from the labour market, the substantive guarantees of the right to work will disappear. If this situation occurs at some point, society will have to provide the income needed for a decent living. In this context, the robotax and unconditional basic income are possible solutions.[4]

1. The impact of AI on the labour market

1.1 The gradual transformation of the labour market: short- and medium-term perspectives

Today, it is becoming clear that, with the development of AI, more and more work tasks can be automated. As a result, some – and a growing number of – jobs, tasks and even industries may disappear. Current projections suggest that the transition to an AI-dominated labour market will largely eliminate low-skilled, monotonous, repetitive jobs. In order to ensure that robots do not suddenly take over human jobs, the new jobs that will be created as AI develops will require the ‘upgrading’ and ‘retraining’ of the existing workforce.[5],[6]

The assumption that ‘AI and robots are taking all the work’ often stems from the disporportion of the ‘quantity of work that can be automated’. As far as we know at present, AI and robots will not lead to the creation of an unmanned labour market in the foreseeable future. A fundamental reason for this is that the amount of work to be done is not fixed. It is true that, in principle, machines can take over the tasks done by humans, but the labour-displacing effects of automation are counterbalanced by two forces that economist Daron Acemoglu calls the ‘productivity effect’ (a cycle in which the use of AI and robots instead of workers allows companies to produce more cheaply, thereby increasing the purchasing power of consumers, which, in turn, makes companies more profitable and creates more jobs) and the ‘reinstatement effect’ (automation creates new jobs). Acemoglu’s research shows that about half of the employment growth between 1980 and 2015 occurred in occupations where the jobs or tasks performed by workers have changed.[7],[8], [9]

While much of the literature on AI and unemployment and the labour market takes the view that automation can replace workers, it is also worth looking at how AI can create jobs and help job seekers avoid unemployment. According to the 2021 report issued by the World Economic Forum, AI is expected to create 97 million new jobs by 2025. Many of these jobs are unlikely to follow the traditional full-time employment model. Instead, more employers are expected to expand the number of teleworkers, or to outsource a significant part of their labour needs through platforms or contractors.[10]

At the moment, there are several reasons to believe that we are not yet living in a time of an exclusively robotised labour market. Some of these reasons are highlighted below: 

1. AI will rarely be able to replace an entire profession, because, in most cases, professions require much more flexibility and adaptability. 
2. At the same time, new technologies not only eliminate, but also create jobs, many of which did not exist before. 
3. The fact that it is technically possible for AI to replace an entire profession does not mean that it is going to happen, since a) in many cases, humans will still do the work cheaper than robots; b) legal liability is still to be regulated (e.g. who should be liable for the AI’s activities and the resulting damages, etc.).
4. New technologies contribute to increasing inequalities, both between different groups of workers and between business owners.
5. For the time being, they are changing the demand for certain skills and qualifications and contributing to a shift towards more flexible but more precarious, atypical ‘dependent work’.[11] In addition to these, there are a number of other arguments as well.

1.2 Horizontal and vertical dimensions of robots

Two further influential and transformative dimensions of the adoption of AI and robots in the labour market are 1) the horizontal dimension and 2) the vertical dimension.

1. The horizontal dimension (job substitution). This means that artificial intelligence is increasingly taking over tasks, jobs and, in some cases, entire professions in the labour market. It also means that more and more workers and jobs are constantly at risk in the labour market. Especially 3D (dirty, dangerous and difficult), low-skilled and monotonous jobs are easy to automate. In addition, tasks such as running a customer service/call centre or classifying, retrieving and archiving documents will increasingly rely on automation and will require less and less human work. The same is true for smart factories and warehouses, where humans are replaced by intelligent robots that can safely navigate through space, locate and move objects (such as products, parts or tools), or perform complex assembly operations.[12]
2. The vertical dimension of robotisation (generations of robots). There are differences between robots based on the level of AI use and the resulting applicability. Robots of the same level of development are regarded as belonging to the same generation. With the constant evolution of digitalisation, it is expected that new generations of robots will emerge. Another important feature is that generations of robots can work in complementary roles or alternately. The impact of robot generations on the labour market is that the more intelligent a robot is, i.e. the higher it is in the hierarchy of robot generations, the more jobs it can potentially fill. In the literature, several generations of robots are distinguished. We will discuss only the three main generations. These are:

a)   Industrial robots (image 1). These are known as first-generation robots. Basically, these are simple mechanical arms that can perform precise movements at high speed, many times in succession (repetitively) and for long periods of time, without stopping.

Photo: istockphoto

b)   Collaborative (cooperative) robots or cobots (image 2). Second-generation robots use algorithm-based artificial intelligence. For them, the most important asset is data. These robots are equipped with sensors that continuously collect large amounts of data (Big Data) about a relevant environment. These may be pressure sensors, proximity sensors, tactile sensors, radars, sonars, lidars or vision systems. Their name (collaborative) comes from their ability to perform work together with people. They are also collaborative in the sense that they can be programmed to perform certain tasks, can sense and appreciate changing work environments (human-machine) and can learn from their mistakes to some extent. For example, the Singrow: cobot, the UR20: cobot for small industrial spaces, the Whiz: cleaning cobot or Rita and Chippy: restaurant serving cobots, etc.: 

Photo: gettyimages

c)   Autonomous robots (image 3). Autonomy means freedom from external control or influence; independence. In the world of autonomous mobile robots, this means that the robot carries out its mission with no or minimal human intervention. For example, space probes or more modern examples of self-driving cars, etc.[13],[14]

Photo: gettyimages

1.3 The ‘race’ between robots and humans in the labour market 

It is already clear that there will soon be a surge in the market for robots, which will make high-tech developments commonplace; however, for now it seems that humans will also be needed. According to a 2017 report by the International Federation of Robotics (IFR), an industrial robot can outperform a human in at least the following areas: it does not expect to be paid for its work, it can produce consistent, high-quality products, it does not get tired, and it can work 24/7, therefore, its performance is actually equivalent to 3 full-time employees. As a result, production costs are reduced (for example, Sony uses robots to produce a PS5 game console every 30 seconds, while humans are responsible for placing bare motherboards on the production line and packaging the finished consoles at the end of the process), which is further aided by better materials management. In addition to these obvious benefits, automation has several hidden advantages as well. Since the latest robots are relatively easy to operate and programme, anyone can learn how to do it. While workers are largely limited to their job, and it is time-consuming to train them for a different one, robots can be quickly reprogrammed and thus learn in a short time to efficiently perform a completely new task. In addition, robots, especially the so-called light robots, take up much less space, allowing companies to optimise their use of space. All these advantages contribute to reducing production costs, which is perhaps one of the greatest strengths of robots. This has been recognised by several industries, leading to the rapid development of robotics.[15]

The rise of robots is unstoppable, and there are obvious signs that the fears of workers are clearly discernible. The question often arises as to whether meaningful cooperation between robots and humans is at all possible. Currently, the most optimistic answer seems to be yes, but only if both sides contribute their own strengths to the success, meaning that the speed and precision of robots is combined with human creativity and intelligence. Thanks to their three-dimensional spatial recognition system, advanced robots adapt easily and almost instantly to new workstations and human co-workers. For example, cobots are specifically designed to interact with and learn from humans. If they are shown how to do a task by moving their arm, cobots can learn it in minutes – unlike traditional industrial robots that require 50 to 200 hours of programming before they become able to perform a new task. By contrast, second-generation cobots, in many cases, do not even need a programmer. They are also smaller, lighter, more flexible and mobile, and much cheaper than traditional industrial robots. Although they are generally slower, they can be used more flexibly and can be entrusted with a variety of interactive (human–machine) tasks.[16]

2. Digital technology unemployment as a possible scenario

A few years ago, researchers at the Massachusetts Institute of Technology (MIT) and Boston University[17] pointed to conflicting visions of the future in studies on the potential for technological unemployment. On the one hand, alarm bells are sounded when it is said that automation will mean the end of work. This approach is based on the assumption that many jobs will disappear in the coming decades as a result of the rapid growth of digitalisation (in the US, in certain parts of Europe and Asia).[18],[19] To take just one example, a Chinese factory replaced 90% of its workers with robots, resulting in a three-fold increase in productivity.[20] On the other hand, there are economists who argue that there have been technological revolutions in the past and yet the demand for labour in the labour market has not significantly decreased. They see no reason why the current transition to Industry 4.0 (or even Industry 5.0, according to some) should be viewed differently. Proponents of this belief start from the assumption that technology is only complementary to skilled labour, and therefore technological change can lead to a reduction in labour demand, but cannot completely eliminate human labour,[21] and may even create more jobs.[22],[23]

However, the more ominous consequences of automation for the labour market are also highlighted by a number of business leaders (e.g. Bill Gates or Elon Musk) and mathematicians/information scientists (e.g. Moshe Vardi or Bart Selman) who are clearly at the forefront of the theoretical or practical aspects of robotisation and AI developments. Vardi, Gates, Musk and Selman argue that AI is a threat to the world’s economies and societies, which should not be underestimated. They claim that the time is coming when machines will outperform humans in a multitude of tasks. In Vardi’s view, more than half of the world’s population will be unemployed in 30 years because of robots. A professor of computer science at Rice University in the US insists that robots will pose an existential challenge to humanity – even if they make life easier. This phenomenon is known as digital technological unemployment. It is a ‘one-way street’ where people are being pushed out of the labour market by automation, with absolutely no chance of getting back in. 

In contrast to the traditional industrial revolution, the AI revolution will not be about physical capabilities – at least according to Vardi –, but about who will ultimately win the intellectual war between human mind and AI.[24] For example, Foxconn in China, Samsung in Korea, and Amazon in the US are already using precision robots to replace humans. Bart Selman, professor of computer science at Cornell University, says that robots and other systems are increasingly becoming a part of everyday life, and people are starting to trust them. At the sme time, the artificial intelligence embodied in computers is basically starting to ‘hear’ and ‘see’ what humans are doing.[25],[26]

3. Examples of job substitution (the horizontal dimension of robots)

The above shows that certain jobs can be fully or partially automated. It can be see that, in this sense, many current jobs will indeed become unavailable for humans in the not too distant future. It goes without saying that we cannot predict the future; however, it is clear that more and more robots are already working in the labour market today. Without wishing to be exhaustive, I would like to highlight a few examples:[27]

Robot police. A machine that went into operation in Palo Alto, California, in 2016 (Knightscope[28]) is the first robotic cop in the world to officially enter service. The robotic cop is on patrol duty – it is not a policing cop, but a system of surveillance. The rolling device is equipped with a number of distance-sensing sensors, enabling it to avoid pedestrians or parking cars. Its software keeps separate lists of the persons and vehicles that are allowed access to the area it guards,[29] and immediately alerts the company’s headquarters, if necessary. The unit also has a high-resolution infrared camera, is able to communicate with people using speech recognition, and can even recognise different noises. The device is equipped with software that is able to scan 300 license plates per minute.[30] Since the idea was to avoid scaring children or the elderly, the robot was designed to look like a charming rolling ‘rocket’. People who need help can run up to it and use a built-in emergency button to call the police headquarters immediately.[31] However, the developers stressed that robotic cops will never be able to replace real ones, and is not even fast and agile enough for, among other things, chase and capture operations. Other weaknesses include not being able to climb stairs.[32] In the US, however, remote-controlled robots are increasingly being used in live emergencies. In one incident in Los Angeles, this solution helped police officers avoid risking their lives against an armed robber. Robots can help make police officers’ jobs safer and easier, not to mention the fact that, by using machines, the scene can be approached from directions that were unavailable before, such as from the air, across a swampy area or through a narrow gap.[33]

The hotel industry. In hotel corridors, robot room service staff can sneak around almost unnoticed, bringing food orders, ironed shirts, etc., or acting as bellhops, escorting guests to their room and carrying luggage with ease. Even though robots (mechanical workers) are not very talkative, they are efficient in doing their jobs. The management of hotels that employ them ask their guests to leave feedback on Twitter about their level of satisfaction the assistance and service provided by the robots. Indeed, SaviOne robot[34] (also known asBotlrobot) assistants employed in a pilot operation in 2014 by Aloft Hotel in California can be just as helpful as their human colleagues. In 2016, in a bid to make the stay of guests more pleasant and hassle-free, the owner of the hotel, Starwood Hotels went beyond pilot programmes and put a whole army of robotic employees to work as porters, bellmen, maids, and so on.[35]Another advantage of robots is that they are fully discreet.[36] Hotel Trio, also based in Healdsburg, California, expanded its range of services with an unusual addition. In the hotel, a robotic attendant serves the wine – touch-free, in line with today’s trends. Rosé (nomen est omen) is extremely helpful, punctual, precise and reliable, and is also able to use the lift and find the right room without any human assistance.[37]

Restaurant services. It may seem a distant future, but there are already many robotic restaurants around the world. In Tokyo, the Susiro restaurant near Ogikubo train station is always busy. Even though it is difficult to book a table, the restaurant employs no staff. Another example: There is a restaurant in the amusement park of the city of Sasebo, where more robots work than humans. This is the same city that is home to the world’s first hotel that employs robots). One of the robots working in this restaurant prepares okonomiyaki[38], another one makes cocktails, while a third one warns guests when their time there is coming to an end. From cooks to waiters, everyone working there is a product of artificial intelligence. In Japan, they are forced to do this due to a shortage of available labour.[39] The mitigation of labour shortage with robots is another important aspect of this phenomenon that we have not touched upon yet. Several highly industrialised countries suffer from labour shortages due to demographic reasons, meaning that fewer new workers enter the labour market than leave it, as a result of factors such as reaching the retirement age. In such cases, there are mainly two possible solutions: migration from the external labour market (economic) and/or automation (AI and robots).

In Cambridge, a group of engineers[40] taught a robot to make an omelette, from cracking eggs to serving the prepared dish on a plate. In addition to solving the technological challenge, they have even managed to have it make food that tastes good. Cooking is a very interesting problem for robotics, because humans cannot be fully objective when it comes to food. The challenge is to solve the complex issues of computer perception and human–robot interaction. While robots are usually designed for monotonous and quantitative work, cooking is a qualitative job – and tastes will vary from person to person.[41],[42] On a related note, KFC has unveiled its restaurant of the future, where most processes have been automated to eliminate interaction between seller and customer. After the chicken and the side dishes have been prepared by KFC staff, the orders are placed on a tray and then on a conveyor belt. The tray is removed from the conveyor belt by robotic arms and placed in a drawer with a door, which can be opened from the outside by the customers after having paid by using a credit card or the facial recognition system. KFC’s newest restaurant has opened in Moscow. According to the restaurant chain, such outlets represent the future, as restaurants without human contact are much more hygienic.[43]

Skyscraper-cleaning robot. Skyline Robotics has deployed window-cleaning robots in several large cities. The main purpose of robotisation was to address the problem of labour shortages arising from the fact that around 74% of workers are over 40 and the job is not attractive for young people. This is a typical example of dangerous 3D work. The robot, named Ozmo, has the advantage of not needing a lunch or cigarette break, not going to the toilet and not getting tired, meaning that it can clean far more windows than a human during a given period of time – and all that while working on the outer surface of a skyscraper.[44]

Self-driving tractor with mobile control. John Deere unveiled its self-driving tractor at the 2022 Las Vegas technology trade show. The manufacturer has not created a new tractor, but has added a self-driving mode to its 8R 410 model by installing 12 stereo cameras to the front and the rear of the vehicle. The tractor can be controlled, started and operated with a mobile phone. The price for a single 8R tractor and the associated plough is USD 500,000, or about HUF 160 million.[45]

Cobot forklifts. Jungheinrich’s self-driving forklift can work around the clock without any human intervention. Among other places, Gyermelyi pasta factory in Hungary has deployed several such intelligent forklifts, which can travel up to 300 kilometres a day without interfering with each other or endangering the physical safety of the people working in the factory. Thanks to advanced sensors, forklift trucks can work perfectly with ‘human’ colleagues, who can be prepared for working with the machine with a brief training course. [46]

The above examples contain instances of physical work carried out by robots. At the same time, it is safe to say that robots will also appear in the service sector, as quasi-white-collar workers. Consequently, the idea that only physical jobs can be replaced by automated systems is wrong. Robots are increasingly employed to do intellectual work. In the following, we are going to present a few examples that already exist, without wishing to be exhaustive.

Robotic translators. There are a number of online services that use artificial intelligence for translating texts – just think about the free-to-use Google Translate or ChatGPT. These systems usually learned languages through a long process of programming and human intervention. Usually, they were based on a large number of texts that had already been translated from one language into another by human translators. By reading these, machines can quickly learn how to translate sentences. When they mess up, programmers show them where they went wrong, making them ‘smarter’ next time. In comparison, algorithms developed by the University of the Basque Country (Euskal Herriko Unibertsitatea) and Facebook employees based in Paris can now learn languages on their own. Experts say that, even though the two algorithms have performed less well than Google Translate or humans, the technology looks promising. Especially when you consider that both algorithms were able to learn the languages needed for translation in a single day. Thanks to developments, it is becoming more and more realistic for robotic translators to quickly learn languages for which only a very limited number of translators are available.[47]

Robotic lawyers. According to Richard Susskind, perhaps the best-known expert in the field of legal innovation, the exponential growth of computing capacity, artificial intelligence and the ability to process an infinite number of lawsuits will bring unimaginable changes to the application of law and the administration of justice in the next decade. He argues that ‘people don't want lawyers, but solutions to their legal problems.’ According to Susskind, while more and more people are realising that automation can significantly improve the efficiency of legal work, the profession still tends to ignore the threat that digitalisation poses to lawyers. He predicts that quantum computing capabilities will lead to tremendous advances in the processing of past cases and existing legislation, and that artificial intelligence could enter the predictive era. The importance of this issue is reflected in the growing debate among lawyers in the US as to whether AI will replace lawyers entirely in the near future, whether it will be a tool to make their work easier, or whether the definition of a lawyer will change completely and will also include AI. Other experts agree with Susskind’s findings, but believe that he draws the wrong conclusions. For example, Mark Cohen[48] argues that developments such as free chatbots are very useful, but they are not suitable for substituting humans. He argues that AI will not fully replace lawyers. Robots will perform paralegal work where cases are uncomplicated and the case value is low. Currently, such work is still done by lawyers. While robots cannot solve more complex cases entirely on their own, they will help lawyers predict possible outcomes and collect data.[49]

Summary

Robots are likely to get better and better at handling uncertainty in perception, decision-making and action. At the same time, we hope that robots will only do the jobs that humans delegate to them. It also seems certain that robots will change the traditional workplace; however, there may still be time to prepare for the change. Technology has progressed in recent decades at astonishing speed. It is safe to say that nothing can stop it. This is a fact that must be accepted. While we might have to work with robots instead of humans, that is not necessarily a bad thing. However, if people are not careful, knowledge will become obsolete sooner than it would be comfortable. Fortunately, people are still in control, and are able to take steps to ensure that automation does not take over the workplace before the workers are prepared for the new situation.

Go to next chapter

Back to the MOOC home page

Internet resource

1. https://www.youtube.com/watch?v=qDCW3PFANYI

2. https://www.youtube.com/watch?v=9J7G0Ea_-b0 

Selected resources

Acemoglu, Daron – Restrepo, Pascual: „Artificial Intelligence, Automation and Work”, National Bureau of Economic Research, 2018, Working Paper No. 24196. 

Acemoglu, Daron – Restrepo, Pascual: „The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment”, American Economic Review, 2018, 108/6. 1488–1542.

George Krasadakis (2018) How Artificial Intelligence Disrupts Employment and the Workforce - the Types of Jobs to be Impacted; https://medium.com/innovation-machine/artificial-intelligence-3c6d80072… (2024.11.23.)

Gregory, Terry – Salomon, Anna – Zierahn, Ulrich: „Racing With or Against the Machine? Evidence from Europe”. U.S.E. Research Institute Working Paper Series 18-07, 2018, 24–30.

Hajdú, József (2021) A mesterséges intelligencia hatása a munkaerőpiacra, avagy elveszik-e a robotok az ember munkáját INFOKOMMUNIKÁCIÓ ÉS JOG 17: 2 pp. 3-9.

Hajdú, József: Munka újragondolva - Lehetséges-e együttműködés ember és gép között? In: Kapitány-Fövény, Máté (szerk.) Holnap: Hogyan őrizzük meg lelki egészségünket a gyorsan változó jövőben? 13 trendkutató és 13 pszichológus párbeszéde a jövőről, Budapest, Magyarország HVG Könyvek (2023) pp. 153-169., 17 p.

International Federation of Robotics: „Executive Summary World Robotics 2017 Industrial Robots”, 2017.

International Federation of Robotics: „Executive Summary World Robotics 2018 Industrial Robots”, 2018.

James Pethokoukis: (2022) The Case against Mass Technological Unemployment. (And What Happens If I’m Wrong.), https://www.aei.org/articles/the-case-against-mass-technological-unempl… (letöltés: 2024.11.23.)

Jeremy Bowles: The computerisation of European jobs. Bruegel Institutte, 2014 http://bruegel.org/2014/07/the-computerisation-of-european-jobs/ (2024.11. 23.)

Kónya Ádám: (2023) Kiszámolták, hány ember veszítheti el a munkáját az MI miatt, https://haszon.hu/megkeresni/munka/mesterseges-intelligencia-munkakor (letöltés: 2024.11.23.)

Liu, Zhihao at al.: Robot learning towards smart robotic manufacturing; A reviewRobotics and Computer-Integrated Manufacturing, Volume 77, October 2022, 102360

Mándó Milán (2021) 5 munkahely, amit már robotizáltak. A szakácstól a targoncásig; https://minner.hu/5-munkahely-amit-mar-robotizaltak-a-szakacstol-a-targ… (2024.11.23.)

Robot Generations: 1st, 2nd, 3rd, and 4th Generations (2022) https://www.electricalterminology.com/robot-generations/ (letöltés: 2024.11.23.)

Susskind, Richard: Tomorrow’s Lawyers: An Introduction to Your Future. Oxford University Press, 2013.

Távoli jövő vagy a jelen zenéje a sötét gyár? (2023) https://autopro.hu/gyartosor/a-tavoli-jovo-vagy-a-jelen-zeneje-a-sotet-… (letöltés: 2024. 11. 23.)

Waldal, Espen: „Building a Robot Journalist”, Medium, 2016.

Reference source

[1] The Impact of Artificial Intelligence on Unemployment (2022) https://www.technology.org/2022/09/17/the-impact-of-artificial-intelligence-on-unemployment/ (letöltés: 2024.11.23.)

[2] How artificial intelligence is reshaping our lives (2018) https://www.technology.org/2018/04/19/how-artificial-intelligence-is-reshaping-our-lives/ (letöltés: 2024.11.23.)

[3] A távoli jövő vagy a jelen zenéje a sötét gyár? (2023) https://autopro.hu/gyartosor/a-tavoli-jovo-vagy-a-jelen-zeneje-a-sotet-gyar/952051 (letöltés: 2024. 11. 23.)

[4] Hajdú, József: Munka újragondolva - Lehetséges-e együttműködés ember és gép között? In: Kapitány-Fövény, Máté (szerk.) Holnap: Hogyan őrizzük meg lelki egészségünket a gyorsan változó jövőben? 13 trendkutató és 13 pszichológus párbeszéde a jövőről, Budapest, Magyarország HVG Könyvek (2023) pp. 153-169., 17 p.

[5] The Impact of Artificial Intelligence on Unemployment (2022) https://www.technology.org/2022/09/17/the-impact-of-artificial-intelligence-on-unemployment/ (letöltés: 2024.11.23.)

[6] Over 30 million U.S. workers will lose their jobs because of AI (2019) https://www.marketwatch.com/story/ai-is-set-to-replace-36-million-us-workers-2019-01-24 (letöltés: 2024.11.23.)

[7] Daron Acemoglu, Pascual Restrepo (2018) Artificial Intelligence, Automation and Work, NBER WORKING PAPER SERIES Working Paper 24196; http://www.nber.org/papers/w24196

[8] James Pethokoukis: (2022) The Case against Mass Technological Unemployment. (And What Happens If I’m Wrong.), https://www.aei.org/articles/the-case-against-mass-technological-unemployment-and-what-happens-if-im-wrong/ (letöltés: 2024.11.23.)

[9] Kónya Ádám: (2023) Kiszámolták, hány ember veszítheti el a munkáját az MI miatt, https://haszon.hu/megkeresni/munka/mesterseges-intelligencia-munkakor (letöltés: 2024.11.23.)

[10] The Impact of Artificial Intelligence on Unemployment (2022) https://www.technology.org/2022/09/17/the-impact-of-artificial-intelligence-on-unemployment/ (letöltés: 2024.11.23.)

[11] Will robots and AI cause mass unemployment? Not necessarily, but they do bring other threats, https://www.un.org/en/desa/will-robots-and-ai-cause-mass-unemployment-not-necessarily-they-do-bring-other (letöltés: 2024.11.23.)

[12] George Krasadakis (2018) How Artificial Intelligence Disrupts Employment and the Workforce - the Types of Jobs to be Impacted; https://medium.com/innovation-machine/artificial-intelligence-3c6d80072416 (2024.11.23.)

[13] Liu, Zhihao at al.: Robot learning towards smart robotic manufacturing; A reviewRobotics and Computer-Integrated Manufacturing, Volume 77, October 2022, 102360 

[14] Robot Generations: 1st, 2nd, 3rd, and 4th Generations (2022) https://www.electricalterminology.com/robot-generations/ (letöltés: 2024.11.23.)

[15] International Federation of Robotics: „Executive Summary World Robotics 2018 Industrial Robots”, 2018.

[16] Hajdú, József (2021) A mesterséges intelligencia hatása a munkaerőpiacra, avagy elveszik-e a robotok az ember munkáját INFOKOMMUNIKÁCIÓ ÉS JOG 17: 2 pp. 3-9.

[17] DARON ACEMOGLU, PASCUAL RESTREPO: Artificial Intelligence, Automation and Work, NBER Working Paper No. 24196, Issued in January 2018 p.1.

[18] JEREMY BOWLES: The computerisation of European jobs. Bruegel Institutte, 2014 http://bruegel.org/2014/07/the-computerisation-of-european-jobs/ (2024.11. 23.)

[19] CARL BENEDIKT FREY - MICHAEL A. OSBORNE: The future of employment: How susceptible are jobs to computerisation? Technological Forecasting and Social Change Volume 114, January 2017, Pages 254-280

[21] ACEMOGLU, DARON and PASCUAL RESTREPO: The Race Between Machine and Man: Implications of Technology for Growth, Factor Shares and Employment, American EconomicReview, 2017.14. pp. 

[22] TERRY GREGORY (ZEW Mannheim) - ANNA SALOMON (Utrecht University) - ULRICH ZIERAHN (ZEW Mannheim): Racing With or Against the Machine? Evidence from Europe. U.S.E. Research Institute WorkingPaper Series 18-07, 2018 pp. 25-27

[23]https://www.elektro-net.hu/uzlet/7325-robotok-miatt-tobb-lesz-munkahely?acm=_103 (2024.11.24.)

[24]https://www.newsweek.com/2018/11/30/ai-and-automation-will-replace-most-human-workers-because-they-dont-have-be-1225552.html (2024.11.23.)

[25]http://faktor.hu/faktor-munkanelkuli-lesz-a-vilag-nepessegenek-a-nagy-resze-30-ev-mulva (2024.11.23.)

[26]https://futureoflife.org/ai-researcher-bart-selman/ (2024.11.23.)

[27] Hajdú, József: Munka újragondolva - Lehetséges-e együttműködés ember és gép között? In: Kapitány-Fövény, Máté (szerk.) Holnap: Hogyan őrizzük meg lelki egészségünket a gyorsan változó jövőben? 13 trendkutató és 13 pszichológus párbeszéde a jövőről, Budapest, Magyarország HVG Könyvek (2023) pp. 153-169., 17 p.

[28] A Knightscope az első robotzsaru a világon, amely hivatalosan szolgálatba állt. Az eszközből két modell érhető el jelenleg: a K3 és a K5 típusú. Az egységet számos szenzorral szerelték fel és ezáltal képes kikerülni a járókelőket. Csak akkor van probléma, ha az emberek gyorsabban reagálnak a robotnál. Stacy Dean Stephens egykori rendőr alapította meg a Knightscope nevű startupot azért, hogy a korszerű technológiákon alapuló megoldásokat kínáljon a rendőrség számára.

[29]Van fehér-, szürke- és feketelista. Az utóbbin azok nevei, rendszámai és információi szerepelnek, akik bírósági döntések alapján nem tartózkodhatnak bizonyos területeken. Ilyen lehet például egy egykori alkalmazott, aki haraggal vált el a volt munkahelyétől vagy egy távoltartási végzést kapott elvált férj. Amennyiben valaki mégis megjelenik, akkor a Knightscope tájékoztatja a hatóságokat és riasztja a legközelebbi rendőröket.

[30]https://sg.hu/cikkek/119201/munkaba-allt-az-elso-robotzsaru (2024. 11. 23.)

[31]https://m2mzona.hu/biztonsag/eles-bevetesen-az-elso-robotzsaru (2024. 11. 23.)

[32]https://sg.hu/cikkek/119201/munkaba-allt-az-elso-robotzsaru (2024. 11. 23.)

[33] https://m2mzona.hu/kozszolgaltatas/robotzsaru-fegyverezte-le-a-bunozot (2024. 11. 23.)

[34] https://www.savioke.com/ (2024. 11. 23.)

[35] Ld. részletesebben a következő linken: https://www.travelandleisure.com/slideshows/hotel-technology-starwood-robot-butlers (2024. 11. 23.)

[36] https://m2mzona.hu/gyartas/a-robotok-is-betornek-a-munkaeropiacra (2024. 11. 23.)

[37] https://roadster.hu/ebben-a-szallodaban-robotinas-viszi-ki-a-bort/ (2024.11. 23.)

[38] Japán palacsinta vagy japán pizza, általában ezzel a két névvel próbálják leírni a japán okonomiyakit. Igazából mindkettőnek lehet mondani: az alapja hasonlít a palacsinta tésztához, a kinézete pedig a pizzához (valamennyire).

[39] https://hvg.hu/gazdasag/20181201_A_robot_nem_szol_be_a_fonokenek_japan_receptek_munkaerohianyra (2024. 11. 23.)

[40] A Cambridge-i Egyetem kutatói a Beko háztartásigép-gyártó céggel együttműködve gépi tanulást használtak a robotséf kiképzéséhez és az IEEE Robotics and Automation Letters című folyóiratban mutatták be eredményeiket.

[41] https://nuus.hu/tech/0603/robotsef-omlettet-keszit/ (2024.11.23.)

[42] https://www.ibtimes.sg/scientists-train-robot-make-omelet-46135 (2024.11.23.)

[43] https://www.portfolio.hu/uzlet/20220627/ilyen-lesz-a-jovo-gyorsetterme-robotok-futoszalagok-automatak-a-kfc-nel-438968?utm_source=hirkereso_es_kapu&utm_medium=portfolio_linkek&utm_campaign=hiraggregator (2024. 11.23.)

[44] Mándó Milán (2021) 5 munkahely, amit már robotizáltak. A szakácstól a targoncásig; https://minner.hu/5-munkahely-amit-mar-robotizaltak-a-szakacstol-a-targoncasig/ (2024.11.23.)

[45] Mándó Milán (2021) 5 munkahely, amit már robotizáltak. A szakácstól a targoncásig; https://minner.hu/5-munkahely-amit-mar-robotizaltak-a-szakacstol-a-targoncasig/ (2024.11.23.)

[46] Mándó Milán (2021) 5 munkahely, amit már robotizáltak. A szakácstól a targoncásig; https://minner.hu/5-munkahely-amit-mar-robotizaltak-a-szakacstol-a-targoncasig/ (2024.11.23.)

[47]https://m2mzona.hu/kozszolgaltatas/egyedul-tanul-nyelveket-a-mesterseges-intelligencia (2024.11.17.)

[48] A Legalmosaic alapítója.

[49] https://jogaszvilag.hu/uzlet/jogaszok-helyett-robotok/ (2024.11.23.)