Since (back in 1941) computers took shape they first filled entire rooms, then office desks, then pockets. But they all had one thing in common: they were designed by human minds. Over the years, many people have wondered: what if computers design themselves?
Someday soon, a smart computer could create a machine far more powerful than itself. That new computer would probably make another even more powerful, and so on. Artificial intelligence travels an exponential upward curve, reaching heights of cognition inconceivable to humans. This, in one word, is the technological singularity.
The term "technological singularity" dates back over 50 years, when scientists were just starting to tinker with the binary code and circuits that made basic computation possible. Even then, technological singularity was a stunning and formidable concept.
A new generation of super-intelligent computers could revolutionize everything. From nanotechnology to immersive virtual reality to superluminal space travel.
With the technological singularity, instead of exploiting only our bio-based brain, we could use artificial intelligence, interface with it. Improve or increase the performance of our brain with implants, or even digitally upload our minds for survive our bodies.
The result of the technological singularity would be a supercharged humanity, capable of thinking at the speed of light and free from biological worries.
A totally new world
The philosopher Nick bostrom he thinks this dynamic world could bring a whole new era.
In this world we would all be more like children in a giant Disneyland run not by humans, but by machines that we have created, or that they have created themselves.Nick bostrom, philosopher, writer and director of the Future of Humanity Institute at the University of Oxford
The starry sky above me, the moral law inside my processor
The technological singularity could lead us into a utopian fantasy or a dystopian nightmare. And this Bostrom knows it well. For decades he has been thinking about the emergence of a super-intelligent AI and knows well the risks that such creations entail.
There is the classic sci-fi nightmare of a robotic revolution, of course, in which machines decide they prefer to be in control of the Earth.
Even more likely, however, is the possibility that the moral code of a super-intelligent AI, whatever it may be, simply does not align with ours.
An AI in charge of self-driving car fleets or the distribution of medical supplies could wreak havoc if it no longer valued human life the same way we do.
The AI Alignment Problem, as it is called, has taken on a new urgency in recent years, thanks in part to the work of thinkers like Bostrom.
Technological singularity and differences of thought
If we are unable to control super-intelligent artificial intelligence, our fate may depend on future machine intelligence thinking like us. On this front, Bostrom reminds us that efforts are underway to “design AI in such a way that it actually chooses things that are beneficial to humans, and chooses to ask us for clarification when it is not clear to it what we mean. "
There are ways we can teach human morality to a nascent superintelligence. Machine learning algorithms could be taught to recognize the human value system, just like today GAN they are trained on a database of images and texts. Or, several AIs could argue with each other, under the supervision of a human moderator, to build better models of human preferences.
A question of respect
If the technological singularity were to create no longer simple machines, but real artificial thinking minds, we should also consider the ethical aspect. It would become, says Bostrom, a necessity to ask ourselves whether it is right to influence them, and to what extent.
In this era of conscious machines, in summary, the technological singularity could impose a new moral obligation on human beings: to treat digital beings with respect.