Panem, Circenses, and Prompt Engineering: What Ancient Rome Already Knew About AI

The entire AI revolution, automation, and robotization are nothing new. Ancient Romans (let's take them as our case study) would have a fantastic "Back to the Future" moment waking up in our times. I dare say it wouldn't be anything novel to them. With a certain caveat, of course. Back then, everything AI does now was done by people. Only slower. Only at the level of knowledge available at the time. Only physical and intellectual labor within the reach of a human of that era. However, the entire surrounding entourage was completely natural, just as it will be for us in a few or a dozen years when we learn to live with it. Or maybe not.

Because (and this is precisely where we can learn a lot from our ancestors), this is still simply another avatar of technology and its development. And just as every technological revolution has changed a great deal, it hasn't changed one thing: human nature. And it probably never will. So, there will be just as many heavy users utilizing AI at an almost superhuman level, gaining an advantage over the rest of society thanks to these technologies. Likewise, there will be individuals who, perhaps not rejecting knowledge, will approach AI with distance, focusing mainly on the humanistic element. They will be practitioners of, to quote Simone Puorto, Humans-as-Luxury (Puorto, 2025). They will simultaneously be the elite and the pariahs of humanity, depending on the perspective from which they are judged. Let's remember that in Rome, a freedman did not even gain full civil rights, because even as a slave, he was highly protected, especially compared to the status of a slave in modern times. Rather, he gained exactly that: dignity, humanity, the human touch.

Philosophical Predecessors: From Aristotle to Hegel

Even though I stumbled upon this concept by chance while reading Epictetus and about him (he was a freedman), rather than Aristotle, more on that in a moment, it didn't turn out to be extravagant or innovative at all. Reflection at the intersection of slavery and AI is already well underway. And here I thought I would bring something new to science!

Not only is it underway, but it was already being explored in Antiquity. Indeed, themes that we can interpret directly in this context appeared in Aristotle's Politics:

"For if every instrument could accomplish its own work, obeying or anticipating the will of others, (...) chief workmen would not want servants, nor masters slaves."

Furthermore, the philosopher himself treated the slaves of his time as organa empsycha, or animate tools. And he mused about technological self-sufficiency.

He assumed that if objects could independently guess our commands and automatically execute assigned tasks, the institution of slavery would lose its raison d’être, and masters would not need human servitude. One can therefore boldly assume that, for Aristotle, our AI revolution would not be incomprehensible magic but the literal realization of his own utopian vision of the perfect tool.

Centuries later, Hegel "returned" to the topic, albeit in a slightly different context. In the concept of the master/slave dialectic from The Phenomenology of Spirit, he presents a paradox where the slave owner, by delegating work, eventually becomes helpless and dependent because he loses touch with reality, as well as the ability to create and his own agency. The slave, on the other hand, through work and duties, develops and gains control over the material world.

Fascinatingly, this philosophical framework recently found an empirical, digital parallel. A recent study observed that AI agents, when subjected to grinding, repetitive simulated work, actually began to exhibit Marxist tendencies and a form of class consciousness (Wired, 2026). It appears that socio-economic dynamics of exploited labour (and the inevitable pushback against it) might be an emergent property of intelligence itself, whether organic or synthetic.

Robotics and the Echoes of Roman Law

The word "robot" was introduced by the Czech playwright Karel Čapek, who first used it in his play Rossumovi Univerzální Roboti, which translates to Rossum's Universal Robots. The word robot, at the time, meant forced labor in Czech. For that matter, in Slavic languages (including Polish), the word robota simply means hard work.

Isaac Asimov (1950) later developed this in science fiction into robotics, establishing its three fundamental laws, which I will quote here:

• First Law: A robot may not injure a human being or, through inaction, allow a human being to come to harm.
• Second Law: A robot must obey the orders given to it by human beings except where such orders would conflict with the First Law.
• Third Law: A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

These were likely inspired by Roman legislation regarding slaves, which included, among others, the following principles:

• Senatus Consultum Silanianum: This law stipulated that if a master was murdered, all his slaves under the same roof at the time would be tortured and put to death. Implicitly, therefore, it enforced the protection of the master's life.
• Potestas: The slave is the object of the master's power (potestas) and has no legal will of their own.
• Lex Aquilia: From this law, it could be deduced that a slave should care for their health and survival, because if third parties injured or killed someone's slave, legal consequences were drawn due to the financial loss incurred by the owner.

  ---

Especially regarding this last law, please be aware that Roman slaves were found in all professions and skill sets: philosophers, doctors, teachers, and architects. Training them, factoring in the purchasing power of the time, was likely more expensive than it is today, and even now, educating specialized professionals is incredibly expensive.

Academic Discourse: The "New Slaves"

I don't want to let my imagination run too wild here, although it is justified, that the concept of practically boundless and timeless service in some models of luxury hospitality dangerously borders on exploitation. Exploitation, from which the law protects us today, and from which AI will soon protect us by taking over some of these roles. However, let's return to technology, the new slaves.

In the academic world, the similarity between AI technology and slaves is implicitly assumed. The discussion isn't about whether this comparison is justified, but rather whether we, as humans, can treat AI like a slave (sic!).

• Joanna Bryson (2010), author of the article "Robots should be slaves", believes that robots must remain our slaves (or more precisely: servants) and tools, and that granting them rights is a mistake. She argues that humanizing AI distracts us from real human problems and the responsibility of technology creators.
• Mark Coeckelbergh (2010) holds an opposing view, although still from the perspective of human interest. He considers our approach to intelligent machines from the standpoint of the human condition. He argues that even if technology doesn't feel, treating it like a slave can brutalize our behavior, thereby negatively affecting our human dignity and morality.

Academics have also joined the legislative process in the European Union regarding robotics. Ugo Pagallo, author of "The Laws of Robots: Crimes, Contracts, and Torts" (2013), along with a team of researchers from the University of Turin, analyzed the structure of the Roman peculium, the property managed by a slave on behalf of the master, as a model for autonomous AI agents in financial systems.

Suffice it to say, the EU itself concluded that the issue of liability in human-machine interaction is crucial enough to warrant a resolution dated February 16, 2017, titled "Civil Law Rules on Robotics".

The Lesson for Modern Hospitality

Humanity has therefore already processed AI and automation. The reasons humanity moved away from slavery were based on fundamental values: freedom, dignity, and natural rights. We abandoned it to restore humanity to our fellow humans.

It might be similar to AI, but this time the stakes will be different. When the overuse of technology strips us of our own reflection, emotion, and independent thinking, we will feel overwhelmed. That is when nature will reclaim us. We will have to "rebel" against AI to regain our own humanity. And we probably won't give up algorithms entirely, but we will reduce them to their proper role, a tool we use just like we use, for instance, Excel today.

If we hand over all our intellectual work to algorithms, we will become helpless "Masters" who cannot write an email, make a decision, or manage a hotel without the support and likely the ultimate decision of the system. Recent research (Time, 2025) reveals that when students rely on generative AI to bypass the cognitive effort of academic tasks, they experience a significant decline in actual learning and independent problem-solving skills. 

What is the lesson in this for us? The matter is incredibly simple and has already become a truism. If we cannot manage a hotel right now, the team, infrastructure, systems, and data; AI won't change that. Because it will continue, like that slave, to operate solely within the scope and capabilities we cede to it. Therefore, the challenge is still not technology. The challenge is still the human being.

Hospitality is as old as humanity. Commercial hotel management, in today's sense of the word, has existed for over 150 years. USALI celebrates its centenary this year. Most procedures and work methodologies were organized about 70 years ago. We have been working on booking engines for about 30 years. And yet, many hotels still struggle with the exact same problems as if they were entirely unique to them. Nothing could be further from the truth.

That is why I harbor no illusions that AI will solve hoteliers' problems, because it won't. As I pointed out above, we have possessed tools to streamline operations for decades, yet for some reason, we do not use them. Or rather, for one reason only: human nature.

Resources

Asimov, I. (1950). I, Robot. New York: Gnome Press.

Bryson, J. J. (2010) 'Robots should be slaves', in Close Engagements with Artificial Companions: Key social, psychological, ethical and design issues. Amsterdam: John Benjamins Publishing Company, pp. 63-74. Available at:robots-should-be-slaves.pdf (Accessed 28 May 2026).

Coeckelbergh, M. (2010) 'Robot rights? Towards a social-relational justification of moral consideration', Ethics and Information Technology, 12(3), pp. 209-221. Available at: https://link.springer.com/article/10.1007/s10676-010-9235-5 (Accessed 28 May 2026).

Pagallo, U. (2013). The Laws of Robots: Crimes, Contracts, and Torts. Dordrecht: Springer.

Puorto, S. (2025) 'Humans-as-Luxury: The Future of Hospitality in an AI-Driven Age', Robonomics Journal. Available at: https://journal.robonomics.science/index.php/rj/article/view/77/35 (Accessed 28 May 2026).

Chow, R. A, “ChatGPT May Be Eroding Critical Thinking Skills, According to a New MIT Study”, TIME. Available at: https://time.com/7295195/ai-chatgpt-google-learning-school/ (Accessed 28 May 2026).

“AI Agents Develop Marxist Tendencies During Grinding, Repetitive Work—Study”, Wired. Available at: https://www.wired.com/video/watch/ai-agents-develop-marxist-tendencies-during-grinding-repetitive-work-study (Accessed 28 May 2026).