{"id":76972,"date":"2025-05-20T00:01:41","date_gmt":"2025-05-20T04:01:41","guid":{"rendered":"https:\/\/world.edu\/?p=76972"},"modified":"2025-05-20T00:01:41","modified_gmt":"2025-05-20T04:01:41","slug":"will-the-day-really-come-when-we-are-served-by-real-robot-waiters","status":"publish","type":"post","link":"https:\/\/world.edu\/will-the-day-really-come-when-we-are-served-by-real-robot-waiters\/","title":{"rendered":"Will the day really come when we are served by real robot waiters?"},"content":{"rendered":"

\"\"<\/span><\/p>\n

We are witnessing a progressive growth of artificial intelligence (AI). Actually, what has grown and been massively deployed since November 2022 is one of its branches:\u00a0<\/span>generative AI (GenAI)<\/span><\/a>\u00a0.<\/span><\/p>\n

This artificial intelligence is capable of generating text, images, and videos, adapted to the format and purpose we desire. For example, texts to promote products on social media, videos based on an idea, and images based on other images.\u00a0<\/span>ChatGPT<\/span><\/a>\u00a0(texts),\u00a0<\/span>MidJourney<\/span><\/a>\u00a0(images), and\u00a0<\/span>Sora<\/span><\/a>\u00a0(videos) are three of the most representative examples of this generation of tools.<\/span><\/p>\n

Perhaps due to its rise, predictions about the automation of work in the service sector are becoming more frequent. One of its consequences would be the replacement of workers by AI tools and robotics. This automation would be total, if the technology replaces the worker, or partial, when the replacement occurs only in some of the job’s activities.<\/span><\/p>\n

Job destruction and automation<\/strong><\/h4>\n

The reality is that there were already predictions about job destruction due to automation long before. In one of the most cited studies\u2014\u00a0<\/span>The Future of Employment: How Susceptible Are Jobs to Computerization?,<\/span><\/em><\/a>\u00a0from 2013\u2014it was estimated that, in 10 to 20 years, 47% of jobs in the United States would be automated. And multiple studies estimate that\u00a0<\/span>between 5% and 50%<\/span><\/a>\u00a0of jobs will be\u00a0<\/span>automated by 2030<\/span><\/a>\u00a0.<\/span><\/p>\n

Despite these predictions, a look around shows that these figures are not being met, and they don’t seem likely to be met anytime soon. We’re not seeing hundreds of robots in restaurants, or machines serving us in hotels.<\/span><\/p>\n

It’s true that self-service kiosks are very common in fast-food establishments or airports, for example. Or robots in restaurants, performing very specific tasks. In any case, these are highly standardized operations with little variability in very specific locations.<\/span><\/p>\n

But when we step back from these highly structured environments, we find that in restaurants, the people who take orders are workers, and in hotels, the people who clean the rooms are people.<\/span><\/p>\n

Why is there such a gap between the projections made, both in the academic and professional spheres, and the reality we find ourselves in? In our opinion, there are several reasons for this.<\/span><\/p>\n

Global analysis and generic results<\/strong><\/h4>\n

Most projections are based on analyses based on generic representations of jobs, as they investigate the labor market or an economic sector as a whole. Therefore, projections are made that do not go into the details of the tasks performed in each job.<\/span><\/p>\n

To avoid these types of errors, we have proposed\u00a0<\/span>a new framework for researching<\/span><\/a>\u00a0work automation.<\/span><\/p>\n

When assessing the potential evolution of the labor market from a general perspective\u2014all digital technology (AI, IoT, robotics), the labor market, the economy as a whole, or a specific economic sector\u2014we can make very interesting, but not very precise, predictions.<\/span><\/p>\n

On the other hand, if you work at lower levels (a specific technology, a specific task, a specific company or industry), the predictions will be more reliable, but they will not inform the entire labor market.<\/span><\/p>\n

In general, most of the analyses conducted to date have been conducted at the higher education level. Therefore, their results on possible developments in the labor market are very striking but, at the same time, imprecise in both magnitude and timing.<\/span><\/p>\n

Machines\u00a0for everything<\/em>\u00a0?<\/strong><\/h4>\n

Another factor to consider is that the vast majority of projections assume that, given the current advances in digital technologies and the significant capacity for progress they have demonstrated, there will always be a technology capable of performing the tasks performed in a workplace.<\/span><\/p>\n

Our research shows that reality is far from this situation. If, for example, we try to find robotic systems capable of performing the tasks performed by waiters, chambermaids, or receptionists, we will find that there are machines and systems capable of performing\u00a0<\/span>some<\/span><\/em>\u00a0of these tasks. For example,\u00a0<\/span>in the hospitality industry<\/span><\/a>\u00a0, there are already robot concierges in hotels, or robot waiters that bring towels to guest rooms, or\u00a0<\/span>a robot <\/span>bartender<\/span><\/em><\/a>\u00a0serving drinks at a bar.<\/span><\/p>\n

However, it’s impossible to find a machine that can perform\u00a0<\/span>all<\/span><\/em>\u00a0the tasks performed in these positions. It’s not even feasible to find a set of machines capable of performing all of these tasks.<\/span><\/p>\n

The other costs of automation<\/strong><\/h4>\n

In work automation processes, other additional considerations must be taken into account, such as:<\/span><\/p>\n