Labor Economics | LLMs: All Humans are Intelligent. For What?

There is a new nickname for LLMs, glorified tape recorder.

In a recent article, AI fears overblown? Theoretical physicist calls chatbots ‘glorified tape recorders’,the physicist said, “It takes snippets of what’s on the web created by a human, splices them together and passes it off as if it created these things. And people are saying, ‘Oh my God, it's human, it’s humanlike.’ Mother Nature would laugh at us because Mother Nature does not use zeros and ones. Mother Nature computes on electrons, electron waves, waves that create molecules. And that’s why we’re now entering stage three.”

The problem of AI may not be that it is not original or has 0s and 1s, but about what it could mean to labor markets. Some people have said generative AI is nothing, understands nothing, some have called it applied statistics, others have said it is a stochastic parrot, some have said the name AI is a misnomer, some too have said that AI is not even as smart as a dog, let alone a human, some said it is not sentient, or does not have consciousness.

There is also chatter about how less threatening it would be to jobs, because other technologies in the past came, and people found other things to do. Maybe. There are, however, a few questions to consider: what is the competence of AI and where accurate, what can it do?

How much of human labor can be digitized and how much of data, within endeavors, is available for LLMs to be trained on? For some entrepreneurs or executives, can AI take on key tasks in the province of staff?

These questions are important to explore, to gauge how AI may permeate in the near future across sectors and continents, doing away with the need for humans, more drastically than anything else in the past.

The established human trajectory, within the last century, is education and work. Most of the work is procedural, requiring responsibilities for which training was provided. If AI can be trained, learning better than many, it may not make much sense to keep many people around.

There are several jobs where people have several free times, only working within intervals. There could be new possibilities with labor in other aspects that may pressure the existing latitude in other areas.

Humans are intelligent, the most among organisms. This intelligence, though special among living things, is common among humans. Human intelligence within society is so compartmentalized that the range of those that really matter is few.

Intelligence is at the heart of all that humans do. Nothing had come close to matching humans, from any time in history, until AI. The rise of machine intelligence may also null some characteristics associated with what it means for humans to be intelligent. There is news on LLMs acing some professional exams. Some say when the questions are tweaked that AI flunks, but for AI to do what humans do, is something else in the way it edges others, regardless of its training on human text data.

There may haveto be new directions for aspects of work and remuneration, around the world in the next few years, as AI as a job candidate may displace people and make the established trajectory of school then work almost null for many, which may necessitate new labor economics and education models.