The term Artificial Intelligence evokes strongly contrasting emotions. Enthusiasts envision a future where humans will simply supervise the work of machines, while skeptics predict a global cultural and employment catastrophe.
What is certain is that, regardless of individual opinions, we are facing another industrial revolution, with consequences that seem destined to change the world.
In reality, the origins of this “sorcery” date back more than eighty years. The first concepts of AI date back to the early 1940s, when Warren McCulloch and Walter Pitts presented the first mathematical artificial neuron. This would later form the basis for today’s neural networks.
It goes without saying that the modern era of AI is marked by the resounding boom of ChatGPT, the chatbot that allows users to communicate with the machine, providing input (the infamous prompts) to obtain output, initially textual and then also visual.
But what is the real impact of such a powerful technology on our way of life? There are various interpretations, and at present, it’s difficult to predict what will happen in the future.
Artificial Intelligence, training and copyright
Evolution is rapid, so fast that there are no systems capable of stemming this Big Bang by setting limits with appropriate laws and regulations.
In such a context, the first obvious criticality is copyright. Tech giants are hoarding material available online to feed their AI models. This has happened, and likely continues to happen, in defiance of any kind of copyright.
While it’s true that large companies can attempt to defend their data archives with lawsuits, “small fish” are afforded very little protection. In the future (or perhaps even now), copyright may be the prerogative of only those with strong backing.
Model Collapse and Privacy
As early as 2025, AI-generated texts online will surpass those created by humans.
Those who generate content by hand will soon be considered a modern-day Hiroo Onoda, their material systematically “plundered” by AI Overview and Perplexity, resulting in desolate lands that confirm the theory of a dead internet, which until recently was considered a fantasy. Those working in the SEO sector are already perceiving this radical change, thanks in part to their ongoing work to understand the mechanisms of algorithms.
The results, at least for now, are clear: visits plummeting and, especially in the digital space, portals seeing their very existence at risk. All of this will inevitably lead, within a few months, to what experts call Model Collapse, meaning an AI training itself with data generated by other AIs, in an endless cycle that will lead to an increasingly defunct web populated by junk content.
Privacy and cyberattacks
Not only that, but a sensitive issue like privacy must also be considered.
The excessive use of AI models, without reviewing the tools’ policies, can pose a risk to the confidentiality of personal data.
As we’ve already mentioned, those operating in the sector have an extreme need to continuously feed their models. Some companies, therefore, may not hesitate to access chat history to obtain what they so vehemently seek.
Already, to obtain their precious loot, several companies are tapping into social networks to fuel their AI. And we’re not talking about some dubious platform, but prominent names in the industry like Meta and LinkedIn.
Cybersecurity is a separate issue, with AI models becoming potentially devastating tools in the hands of hackers and criminals. This technology, combined with deepfakes, is paving the way for a higher level of attacks, leaving cybersecurity experts in a much more difficult position than their counterparts.
AI and Work: Apocalypse in Sight?
AI impacts not only the internet but, more generally, everything related to work.
Despite reassurances, large companies are already implementing massive layoffs among their employees.
For those who remain, staying up-to-date and studying specific topics like prompt engineering or related topics isn’t just an option, it’s a must if they hope to keep their jobs and secure their salaries. The introduction of AI makes work easier, but it doesn’t necessarily lead to increased income.
All this, however, leads to a logical consideration: if tech giants are wiping out their employees, if the labor market is already in deep crisis… who will buy the products?
This short circuit offers a glimmer of hope that, despite the pachydermic slowness of institutions, a way will be found to balance AI and employment in the future, possibly in the near future.
The real benefits of AI: scientific research and medicine
While I’ve painted a disturbing picture of AI so far, there are undeniable positive aspects associated with this technology.
Beyond large corporations, even small entrepreneurs can benefit from these solutions. Especially in Italy, where tax pressure is overwhelming and hiring is nearly impossible, AI can help the classic over-stressed jack-of-all-trades.
Going beyond the economic aspect, the benefits are undeniable for medicine. There’s talk, for example, of a marked reduction in diagnostic errors. Significant results are already clearly visible in oncology, with marked improvements in patient survival.
Moving on to more science fiction-inspired aspects, we can talk about the Da Vinci Surgical System, which combines AI with robotics to create automated surgeons. These, already used in millions of procedures, have demonstrated greater precision than human intervention.
The impact is even more evident when it comes to research. In these cases, AI can significantly reduce the time required for research. An example of this? AlphaFold, developed by Google DeepMind, is a platform that allows for protein structure searches.
Compared to other methods such as X-ray crystallography, it allows results to be achieved in a few months rather than years.
The right tool, but only in the right hands
As evidenced by the adoption of AI in medical and research projects, this technology is nothing less than a very powerful tool.
This technological science is still in its infancy (as evidenced by the many errors, the infamous “hallucinations“), open to various types of abuse, but it fundamentally holds enormous potential.
If humanity is not yet mature, it is the authorities who must set limits for its proper use. The stakes are enormous, extending far beyond the workplace.
As with all great revolutions that mark human history, the emergence of AI presents both negative and positive aspects. Once again, it remains in human hands to determine, properly manage, and control this new era, which could be either a new era or its very decline.
SviluppoMania
Stay Tuned
