Un camino en que no se escatiman los escasos recursos de la Tierra por un sueño de colonización.

Desde que ChatGPT se abrió al gran público en diciembre, las noticias de las IAs se multiplican. Desde la empresa china Netdragon que nombró a Tang Yung, una IA, como su CEO y ha conseguido subir su valor en un 10%, hasta imágenes del Papa con abrigo extra grande o la detención de Trump, creadas con inteligencia artificial, que inundan las redes con fake news, pasando por distintos personajes de la esfera pública que avisan de los peligros de las IAs y la necesidad de controlarlas.

Hoy me ha llamado la atención el caso de dos libros escritos e ilustrados por IAs: ChatGPT a los mandos de los textos y MidJourney a los pinceles de la ilustración.

Detrás de las IAs debe haber alguien que les pida hacer la tarea. En este caso son dos autores ingleses. Uno es, Ammaar Reshi que cuenta en su Twitter el proceso de creación con las IAs y cómo autopublicó el resultado de un álbum ilustrado: Alice and Sparkle (en el que reconoce la autoría a ChatGPT y MidJourney. El otro, Tinuviel Dixon, se atribuye la autoría al completo, también del álbum ilustrado If animals had jobs.

Los que me leéis, sabéis que soy de la opinión de que las IAs deberían estar contempladas de alguna forma en nuestra legislación. Tanto ellas como entidad (jurídica o de nueva forja) como las creaciones y trabajos que se realicen con ellas. Sin ese tipo de control las fake news se multiplicaran y siempre habrá quien haga mal uso a las IAs y quede escudado tras una autoría incierta de quién lo hizo realmente.

Qué no veremos en los próximos años…

Desde hace un tiempo estoy suscrita a la newsletter del físico Avi Loeb. La entrega de hoy me parece inspiradora y la comparto con vosotros. Igual que Orwell imaginó 1984, en 1949, Avi Loeb imagina nuestro 2084.

——

The Jewish Talmud says: “Since the destruction of the temple in Jerusalem, prophecy has been taken from prophets and given to fools and children.” Another way to put it is that as of right now, there is no way of knowing the future.

In physics, this statement boils down to the absence of a spacetime with “a closed timelike curve” that allows for time travel backwards in time. If time travel were possible, then we could have known the future. For example, the products of our future technologies could have visited Earth now and appeared to military personnel in a report from the Office of the Director of National Intelligence as “Unidentified Aerial Phenomena (UAP)”. However, in a 1992 paper, Stephen Hawking proposed the “chronology protection conjecture”, arguing that the laws of physics preclude time travel on human scales.

Nevertheless, we can imagine the future of humanity. In his dystopian novel titled “1984”, George Orwell forecast back in 1949 social phenomena that were realized since then. A timely example is what Orwell termed “newspeak”, a language which eliminates certain words, substitutes one word for another, inverts customary meanings and creates words for political purposes, such as doublethink as the belief in contradictory ideas simultaneously, as reflected in the Party’s slogan: “Ignorance is Strength.” As a matter of fact, I learned last week that a distinguished scientist stated recently: “It would be better if we did not know whether the COVID-19 virus came from a lab leak or the wet market in Wuhan.” Similarly, many scientists argue: “We do not want to know what UAP are,” and the SETI community banned discussions of UAP at their conferences. In other words, “Ignorance is Strength.” As Orwell noted: “In a time of universal deceit, telling the truth is a revolutionary act.”

The ability of Orwell to forecast the future is not a violation of Hawking’s chronology conjecture but simply a brilliant insight into human nature and social trends. Can we repeat Orwell’s spectacular insight to forecast what might happen a century after 1984, namely in 2084?

There is no doubt that by 2084 artificial intelligence (AI) will take over most routine tasks, currently assigned to humans.

Imagine a 2084 where teaching will be done by AI systems based on existing knowledge and where professors in academia will be assigned the sole task of exploring new knowledge rather than pursuing respect from students or honors from peers based on past knowledge.

Imagine a 2084 where universities will include only researchers and students, and not be dominated by administrators that keep adding rules and hiring more administrators to enforce them. Instead, we could all benefit from a limited number of AI bureaucrats whose sole task is to serve the intellectual engine of academia.

Imagine a 2084 economy where citizens will have their checking accounts secured fully by the government and not be vulnerable to the fate of local banks and government bailouts.

Imagine a 2084 where babies will be born on the Moon or Mars and will regard the Earth’s gravity g as imprisoning. Living in low-gravity allows us to lift massive objects and jump over obstacles with ease. Indeed, Sisyphus — as featured in ancient Greek mythology, would have had an easier time pushing a boulder up the mountain, only to see it roll down again. The absurdity of life, highlighted by the philosopher and writer Albert Camus, will be less absurd in low-gravity environments.

Imagine a 2084 where autonomous AI astronauts will be launched towards the habitable zone of every star in the Milky-Way galaxy. This goal can be realized within several decades from now by allocating our military budgets of 2 trillion dollars per year worldwide to space exploration. By prioritizing science over war, we could target all habitable zones in our galaxy with the hope that even if they do not possess intelligent civilizations as of now, they might give birth to them by the time our AI systems arrive there. And even if they remain infertile, our AI systems could serve as interstellar gardeners that plant the seeds of intelligent life there. Speaking about that option, did interstellar gardeners ever plant their seeds in our own backyard on Earth before we came to exist? Since most stars formed billions of years before the Sun, uncovering their deeds will reveal to us our own technological future without violating Hawking’s chronology conjecture.

Imagine a 2084 where the research team of the Galileo Project will find evidence for technological objects from extraterrestrial civilizations on Earth. Entrepreneurs would work around the clock on making a profit from the new technologies that were uncovered. Here’s hoping that after realizing that there is a smarter species in our cosmic neighborhood, we will start to treat each other as equal members of the human species because our differences are insignificant on the cosmic scene.

Imagine a 2084 where humanity will be active on the interstellar dating scene and find intelligent partners from whom it can learn. The interstellar community might give a meaning to our life. In other words, the cosmos will not appear “pointless” anymore, as argued by the Physics Nobel laureate Steven Weinberg in his book “The First Three Minutes”. Since it takes less than a billion years for chemical rockets to traverse the Milky-Way disk of stars, the last billion years might be far more consequential for our life than the first three minutes after the Big Bang.

Life could be a self-fulfilling prophecy. If we imagine a better future, we can attempt to make it a reality. However, if we obsess with our limiting past, we will end up with Orwell’s unfortunate society. Whether we seek progress or decline, is up to us. Our AI machines will serve the future we dare to imagine for ourselves.

According to Genesis 1:27, God created human beings in its image. The modern caveat is that if interstellar AI gardeners visited us, the credit goes to them. By now, we are creating AI systems in our image. Within a billion years, AI astronauts might plant our technological seeds throughout the Milky Way galaxy. Cosmic optimism is all about paying it forward.

+ Publicación original