GPT itself - also known as ChatGPT - is the most powerful chatbot the world has ever known. Yes: it's a form of artificial intelligence that gives real-time answers on an infinite number of topics that the model has learned, not just from the information it was fed when creating it, but from exchanges with its users.
It's a giant leap into the world of artificial intelligence language models — and that's not yet connected to the Internet. Still, it is likely to change the course of several writing, research, outreach, and translation industries.
It's a giant leap into the world of artificial intelligence language models—and that's not yet connected to the Internet. Still, it is likely to change the course of several writing, research, outreach, and translation industries.
It is so powerful that, according to the New York Times, Google has already raised an alert about its existence.
“While ChatGPT still has a lot of room for improvement, its release led Google management to declare it a 'code red'. For Google, this was similar to pulling the fire alarm, ” reporters Nico Grant and Cade Metz wrote in the New York newspaper.
“Some fear that the company is approaching a moment that the biggest companies in Silicon Valley fear: the arrival of a huge technological change that could disrupt their business,” Grant and Metz wrote. It makes sense: if you can ask a bot that answers everything in real-time, what good is googling? When is this powerful tool connected to the Internet?
But what is GPT?
If there is a clear answer that GPT can give, it is what it is. We asked him and he answered this:
GPT is the acronym for “Generative Pre-training Transformer”. It is an OpenAI language model that has been trained on a large amount of text from the internet to learn to predict the next word in a sentence or to generate coherent and natural text. GPT is capable of performing various tasks such as translation, troubleshooting, and text generation. It is used as the basis for many other language models and has been shown to perform outstandingly on many tasks.
GPT is a language model based on the Transformer architecture, which has proven to be very effective at processing natural language. The word “ generative ” in his name refers to his ability to generate new and original text, while “ pre-training ” refers to the fact that he has been pre-trained with a large amount of text to improve his performance in various tasks.
It's a giant leap into the world of artificial intelligence language models—and that's not yet connected to the Internet. Still, it is likely to change the course of several writing, research, outreach, and translation industries.
It is so powerful that, according to the New York Times, Google has already raised an alert about its existence. “While ChatGPT still has a lot of room for improvement, its release led Google management to declare it a 'code red'.
For Google, this was similar to pulling the fire alarm, ”reporters Nico Grant and Cade Metz wrote in the New York newspaper. “Some fear that the company is approaching a moment that the biggest companies in Silicon Valley fear: the arrival of a huge technological change that could disrupt their business,” Grant and Metz wrote.
It makes sense: if you can ask a bot that answers everything in real-time, what good is googling? when this powerful tool is connected to the Internet?
ChatGPT, which as of January 2, 2023, presents itself as Assistant (and insists that it has always been called that), has, as the two New York Times reporters rightly say, a lot of room for improvement. And as we pointed out before, it is not yet connected to the Internet.
In addition, as the same OpenAI site, the non-profit company dedicated to artificial intelligence research that developed it, acknowledges, the system can give incorrect answers. “It has limited knowledge of the world and events after 2021 and may occasionally produce harmful instructions or biased content,” the site says on its FAQ page.
GPT is known to be very good at the language completion task, which means that it can predict the next word in a sentence or complete a given sentence coherently and naturally. In addition, it is also capable of performing tasks such as translation, problem-solving, and the generation of coherent and natural text. It has been used in a wide variety of applications and has proven to be very effective in many of them.
If these lines were not italicized and were not anticipated to be written by artificial intelligence, it would be very difficult — if not impossible — to determine that they were not written by a real person.
The scope of such a powerful tool is multiple and even unsuspected. “ My first impression was of wonder and restraint,” says Matthew Carpenter-Arévalo, a specialist in technologies focused on climate change. “On the one hand, for the first time in my life, I felt that this technology could replace me, or at least many of the functions that I do,” explains Carpenter-Arévalo.
Yet, he says, it's clear that " as impressive as it is," he feels he's still limited. “He can write content, for example, but I don't think he'll be able to come up with an independent persuasive thesis without his ideas existing widely on the internet,” he says.
The discussion about the replacement of human creativity is already deep in the world of artificial intelligence. “Our contact with these technologies is going to change the standards and parameters of how we evaluate creativity and knowledge ”, says Luciana Musello, professor of communication and digital media at the University San Francisco.
As a concrete example, Musello explains, at the end of last semester she received what she believes were three essays generated by artificial intelligence. At that point, she thought that perhaps in the future she will favor essays that do not have “the best writing and spelling”, because what she is ultimately concerned with “ is the content and not the form, which these technologies produce with such precision ”.
So, a " more or little " essay will be preferable in terms of the format "but where I can perceive that human error and hot reflection, of flesh and blood, of humans in front of a cold and correct writing but that I intuit comes from a machine," says Professor Musella.
Marcus Du Sautoy, Professor of the Public Understanding of Science at Oxford (formerly held by Richard Dawkins) has written an extensive book on the subject The Code of Creativity.
In it, he looks at the future of artificial intelligence as creative minds — something they're still a long way from, but already on the way. What goes on inside our heads remains a mystery, but in recent years a new way of thinking about code has emerged: a shift from a top-down programming attitude to a bottom-up effort. for computers to design their path .” That is, they learn by themselves.
It's what ChatGPT has done—and what it will continue to do for millions of people: at least as long as its creators keep it free. “ The foundation that created ChatGPT has not yet determined its business model ”, explains Carpenter-Arévalo. “I suspect that for the public there will always be a free version and for businesses/apps, maybe they will have some free queries until they have to start paying,” she says. Her point is that OpenAI's business model will determine how pervasive it will be in our lives.
Still, 2023 will bring more innovations in artificial intelligence. The OpenAI co-founder has said that this year will be much more aggressive in innovations in this industry.
In this panorama, some see the future with hope and others with a certain suspicion. OpenAI CEO Sam Altman has said that technological progress is the only thing that produces real and sustainable economic progress.
Luciana Musella sees certain warnings in that statement. “ This is a position that in theory, we call technological solutionism, in which technological solutions are thought to be superior to social processes,” Musella alleges. “It assumes that it is a technology that drives the story when we humans are the ones that drive it,” she says.
For this reason, “ if these technologies must be viewed with suspicion or caution, it is not because of what they are in themselves, but because of who is shaping their development”, says Luciana Musella. We will see how this historical process unfolds, step by step, in real-time, every day of this present so intertwined with the future.
.png)
English (US) ·