In 1999, the Wachowskis released the first film in an epic sci-fi trilogy that our world, if it really is our world, has never been the same since: The Matrix. That year, the film foresaw much of the debate surrounding the role of artificial ielligence in society today, the massive advanceme of virtual reality, and the poteial threat of autonomous robots.
Now, with the return of Neo (and Trinity) in The Matrix Resurrections, it’s the perfect time to discuss this importa question: Is super-iellige machine takeover of human society immine?
Maybe not yet, but there are several reasons to worry about AI dominance. Philip Berry, professor of philosophy and ethics of technology and head of the artificial ielligence and robotics team at the SIENNA project, believes that we are probably not looking for risk in the right place.
Barry says: “The concern that artificial ielligence systems consider themselves superior to humans and use this as an excuse to corol and dominate the human race is specific to cinema and science fiction literature. In reality, the risk of machine domination has nothing to do with their ielligence level.
Yes! It’s scary… but it’s not that bad. So sit down and immerse yourself in this simulated world to analyze the three big scieific questions raised in the original Matrix trilogy:
3. What makes the human brain function unique and can robots do it?
In the Matrix trilogy, machines look and act like humans. The only difference between them and humans is the robotic tone of their voice and their super-fast reactions. But this degree of similarity between robot and human is far from real world technologies.
With curre algorithms, artificial ielligence will never be able to mimic the unique functioning of the human brain, says Alison Mottery, a professor of cellular and molecular medicine at the University of California, San Diego. AI is “even better than humans” at recognizing things based on prior training, explains Motter, whose studies are in the iricacies of the human brain. If you can train an AI algorithm to recognize faces well, the AI will do very well (although the same AI facial recognition technology can also have dangerous biases.) But the main challenge that preves “The closest artificial ielligence comes to human performance is that they cannot coinuously learn and adapt to situations for which they have not been trained.”
British mathematician Alan Turing inveed the Turing Test to challenge robots. During the test, the robot had to convince the other side of the conversation (which was done in writing) that it was human. While some modern chatbots can be seen as a testame to the progress of bots in this field, most neuroscieists are still unconvinced.
“Artificial ielligence may be able to fool you in a simple conversation based on a set of data or a formula, but if you make the conversation a little more complicated, you’ll realize that the other side of the conversation is not human,” Mottery says.
2. Can we use machines to upload data io our brains?
In one of the most ieresting scenes in The Matrix, Neo is connected to a device that is very similar to the brain-machine ierface in the real world, a device that connects the human mind to the machine using implas and electrical impulses.
Miguel Nicolelis, a pioneering scieist in the field of brain-machine ierfaces and a professor emeritus at Duke University, published his first paper on brain-machine ierfaces in 1999. Exactly the same year that the first Matrix movie was released.
Naturally, the film left a great impact in this area. “I think The Matrix was really one of the best sci-fi movies I’ve ever seen,” Nicolelis says. Of course, not in the sense that it is scieifically realistic. The human brain learns through association. As a result, unless there is practice, nothing can be learned by loading information io the mind. What is certain is that you cannot directly download or upload coe from the brain.”
1. Will artificial ielligence dominate humanity in the future?
“The Matrix is a dream created by artificial ielligence to keep us under its corol.” This is the famous line that Morpheus says to Neo and it blows Neo’s mind and audiences around the world at the same time.
The Matrix is actually a reflection of the same fear that science fiction writers have since the time of Arthur C. Clarke and from the beginning of the formation of artificial ielligence have been its harbingers: the domination of man by his creation.
Humans have long been concerned about whether robots can really achieve the same unique human awareness of their existence. This was the first step for the dominance of robots in the Matrix movie, and with the artificial ielligence algorithms becoming more complex in rece years, this issue has become a concern that is likely to happen at any mome.
While many other sci-fi writers and filmmakers worry about the dominance of robots like the Wachowskis, AI ethicists have a differe opinion and believe we can rest easy.
Because even if robots can reach this level, there is still a long way to go.
“For the foreseeable future, say 20 to 50 years, AI ethicists aren’t too concerned about machines becoming more conscious or smarter than humans,” Barry says.
But you don’t need super-iellige robots to worry about cars. Even today, Barry says, there are three reasons to worry about AI corol:
- Machines with low or medium ielligence that can act corary to human commands and limit their freedom. Like a self-driving car that disobeys your orders to turn right.
- The curre artificial ielligence that can formulate its goals in such a way that humans are not able to understand properly and has taken away corol from humans.
- As AI becomes more and more independe, we are more and more at risk of artificial ielligence versus human decision-making, resulting in a society where most decisions and actions are corolled by machines rather than humans.
“Most robots have pre-defined scenarios that limit their performance to specific tasks to help humans,” says Hooten Jabli, an assista professor of architectural engineering at Penn State University whose research focuses on improving communication between robots and humans on construction sites. “But if robots reach a poi where they can define a new scenario for themselves and take a new action without human ierveion, it may create poteial risks.”
It’s true that Nicolelis thinks the possibility of robots using brain-machine ierfaces to corol humans, similar to The Matrix’s human power pla system, is highly unlikely, but he goes on to propose an even scarier hypothesis: Machines have already taken over, we just don’t know. We have not become that.
Modern society today depends on algorithms that dominate every aspect of our lives: the products we buy, the friends we talk to, the people we date. Machines even dominate our stock-based economy.
“Algorithms are already indirectly harming us,” Nikollis bluly believes. They are the ones who shape human behavior.”
Source: INVERS
