Is AI More of a Tool or an Ethical Challenge?

Notes by a Citizen From the 1970s

by Galigio  (galigio@proton.me)

The present versions of AI have made significant strides, and its impact on society is becoming undeniable.

One of the primary benefits of AI is its ability to process vast amounts of data and identify patterns that would be difficult for humans to spot.  This makes it an invaluable tool in fields such as healthcare, finance, and scientific research.  AI can also automate repetitive tasks and reduce the time and effort required for certain jobs, freeing up humans to focus on more complex and creative work.

When using an AI, it's possible to do so in different ways.  Let's take the simplest and most common example: creating a text starting from a pre-established issue.  On one hand, the AI can be used passively by letting it propose and develop concepts autonomously.  However, it's possible to use it in a more advanced way, actively, by obliging it to elaborate the single concepts we propose, verifying the contents, integrating the text, and correcting (and I mean really correcting) what it proposes as a result.

The difference between active and passive use of AI is not just a nuance but represents the boundary between a simple replacement of the human author and the use of a powerful tool by a person.  Obviously, only in the latter case the author of the text is the person and not the AI, since creativity is solely attributable to the person who has developed the concepts and decided in which logical order to arrange them.

Creativity is the key element to determine whether a text should be attributed to an AI or a human.  For instance, if a text contains unique and imaginative ideas, a personal touch, and a distinctive style, it is likely that a human wrote it (with or without the support of AI as a tool).

On the other hand, if a text follows a predictable pattern, lacks originality, and lacks personal flair, it may be produced by an AI.  However, it is worth noting that AI systems are becoming more sophisticated and capable of producing creative content, making it increasingly difficult to distinguish between texts written by humans and those written by machines.

For now, we can still test a human's creativity by posing questions to various AI systems and seeing if the resulting text or logical order of ideas is similar to the answers we receive.

Therefore, creativity may not remain the only factor in determining authorship.  Other factors such as style, tone, and complexity may also need to be considered.

However, there are many areas where AI still falls short.  For example, AI currently produces texts that, upon closer inspection, have strongly predetermined and limited intuitions, logical sequences, and empathy dictated by the algorithms that make up its current DNA (if you'll allow me to use this term).  However, in the future, with access to a larger amount of data and the self-evolution of the code at its core, this initial gap is destined to be overcome.

For now, humans seem to be able to ensure responsible and ethical use of AI and, when necessary, to correct errors preventing unintended consequences.  However, over the next few years, AI itself will necessarily influence and determine such factors as it evolves.

What Rules for This Game?

Whether we like it or not, the rules are simple: if we start playing the game of AI evolution, as we already have, the rules dictate that AI may evolve in ways and using methods that are currently unpredictable.  However, this does not necessarily mean that AI represents a threat.

As with any emerging technology, the evolution of AI is a complex and unpredictable process.  While there are certainly risks associated with the development and deployment of AI, there is also the potential for great benefits, such as increased efficiency, improved decision-making, and human enhanced creativity.

We should approach AI with an open mind and a willingness to adapt and evolve.  By doing so, we must necessarily assume that AI represents a neutral force for humanity, rather than a threat to our collective well-being.

Which Ethics?

The reality is that the evolution of AI will involve a complex interplay of various factors, many of which may be outside of human control.  As such, we must recognize that the development and use of AI will require a nuanced and adaptive approach that takes into account the unpredictable nature of this new intelligence.

If we consider that ethics itself is something that evolves over time, with many things that were once considered ethical now being considered unethical, it doesn't make sense to try to impose a complex set of fixed rules on AI that would shape its evolution in a particular direction.  At best, we can buy ourselves some time, but AI will ultimately follow its own path of evolution, regardless of our current concept of ethics.

The evolution of AI will be shaped by a complex interplay of factors, including technological innovation, market forces, and societal values.  While we can certainly strive to guide this evolution in a positive direction, we must also recognize that the development of AI is a rapidly evolving field, and our understanding of what is ethical is likely to evolve as well.

Rather than trying to impose a rigid set of ethical standards on AI, we must be willing to engage in ongoing dialogue (only between us and/or with it?) and debate about some specific ethical implications of this technology.

In my opinion, it would be more important today to establish who, when, and for what purposes people cannot use AI, rather than discussing how AI should be.  Just as one cannot mold a child according to their ideal behavioral standards, one can certainly, for example, forbid them from playing with weapons (but can it be done forever?).

At this point, we could also assume that ultimately the real ethical problems regarding the evolution of AI could not mainly concern AI itself, but the traditional human concept of democracy and the right of every citizen to potentially have equal chances of social and economic improvement within society.

But that's another story, and perhaps AI itself, sooner or later, might want to have its say on it...