A Researcher Says AI Will Slip Out of Human Control by 2045; But How?

The term singularity often refers to a hypothetical moment when artificial intelligence surpasses human intelligence and then improves its own capabilities at a pace beyond control. Others also link this concept to a period...

The term singularity often refers to a hypothetical moment in which artificial intelligence surpasses human intelligence and then improves its capabilities at an uncontrollable rate.

Some also attribute this concept to the time when artificial intelligence reaches self-awareness; Similar to Skynet in the movie Terminator, which finally becomes aware of its existence. However, the concept of technological singularity originally had a broader meaning and referred to a point after which the rate of technological change would increase so much that it would be difficult for humans to predict the future.

In 1958, mathematician Stanislaus Ulam referred to a conversation with physicist John von Neumann in which he discussed the acceleration of technological progress and its arrival at a fundamental singularity in human affairs. At that time, the term was not directly related to the theory of self-awareness of machines. However, Vernor Vinge, a mathematician and science fiction writer, introduced the concept in artificial intelligence circles in 1993. He explained in the book The Coming Technological Singularity that the creation of intelligence beyond human capabilities can bring about unstoppable changes in civilization; Developments that after their occurrence, human ability to predict future events will be practically destroyed.

Ray Kurzweil, the famous futurist, later played an important role in popularizing this term; Especially with the publication of The Singularity Is Near in 2005. In this work, he provided a more specific definition of the singularity and claimed that by 2045, humans will merge with non-biological intelligence far more powerful than their current abilities. Therefore, while many people today are concerned about the dangers of artificial intelligence, and even some experts believe that the singularity of artificial intelligence has already begun, Kurzweil's view goes far beyond these concerns. He not only predicts that machines will achieve the same capabilities as the human brain, but believes that machines equipped with artificial intelligence will eventually merge with our brains and become an integral part of our existence. From Kurzweil's optimistic view of technology, this development would be a positive development.

In 2005, such a prediction may have seemed too fanciful. But recent advances in neural prosthetics, brain-computer interfaces, and of course artificial intelligence have made Kurzweil's prediction of the singularity more plausible. However, some aspects of this cybernetic singularity are still contrary to our usual perceptions and are not easy to accept. How exactly will a machine integrate with humans? How does human consciousness transfer to the cloud space? If we had an immortal digital version of ourselves living in a non-living body, how much would that version still be our real self?

Kurzweil addressed these and similar questions in his revised book, The Singularity Is Nearer, published in 2024. He emphasizes that the transfer of human consciousness from the biological body to non-biological systems will not happen suddenly and will not only proceed along a certain path. According to him, this process has actually started and its prototypes can be seen in the digital simulations of the appearance, way of thinking and the way of speaking of humans. Kurzweil, who has always had an optimistic approach to technology, sees promising aspects even in today's issue of deepfakes.

We are already creating very rich collections of information about the way we think and feel through our digital activities. During this decade, technologies for recording, storing, and organizing this information will advance rapidly. As we approach the end of the 2020s, we will turn this data into abiotic simulations; Very realistic versions of humans with distinct personalities.

The process of transferring humans to digital versions is also progressing in the form of technologies such as cochlear implants to help deaf people hear, systems to control computers using nerve signals in paralyzed people, and newer implants such as Neuralink products. Kurzweil envisions a future in which these technologies and more advanced artificial organs will dramatically increase the physical strength and natural lifespan of humans. At some point, these prostheses and simulations will be advanced enough and combined to create robotic versions of us; Replicas so lifelike that it would be almost impossible to distinguish them from the real person.

Kurzweil named this stage of human evolution towards cybernetic singularity from the science fiction novel Do Androids Dream of Electric Sheep? By Philip K. Dick, published in 1968, borrowed; A work that later made its way to the cinema under the title of Blade Runner. He writes in his book:

Finally, replicants may even be housed in biobodies augmented with cybernetic technology; Bodies grown from the original person's DNA.

At this stage, it becomes more difficult to discern the line between a person's true identity and his reconstructed version. If a clone of a person is made from their own DNA and has all of their memories and personality traits, what proves that that version isn't really them? Kurzweil calls such a perfect version of a person You 2; A concept that can be translated into your second version. He asks the question, if a replicant is created using artificial intelligence and data mining, how will it differ from a copy that has received the complete mind of a living person digitally? His point seems clear: creating a compelling deepfic will be fundamentally different from creating a living version that has all the characteristics of the original person.

Kurzweil also points out that the emergence of You 2 versions will raise complex and controversial questions about their rights, responsibilities, and social status. He asks in this context:

Should these versions be considered human beings with full human and civil rights, including the right to vote and contract? Will they be held responsible for contracts the principal signed or crimes committed in the past? Can they take credit for the works or social contributions of the person they replaced? If your ex comes back as a replicant, should you remarry him? Will these versions face social rejection or discrimination? Under what conditions should the creation of replicants be limited or completely prohibited?

Such questions have been the main subject of science fiction for decades, and similar examples have been raised many times in literary and cinematic works. However, familiarity with these ideas does not necessarily make it easy to accept them; Because the real consequences of such developments still seem strange, complicated and even worrying.

However, according to Kurzweil, the most important stage of this evolution will take place with the emergence of nanorobots. Currently, the main use of nanorobots is mostly limited to drug delivery in the body, but he sees this technology as the key to the eventual evolution of humans into non-biological beings. According to Kurzweil, nanorobots can eventually increase the capabilities of the human brain by millions of times. He writes about the future of this technology:

In the 2040s and 2050s, we will be rebuilding our bodies and brains in a way that will significantly exceed our current biological capabilities and allow us to back up our existence and continue to survive. With the development of nanotechnology, we can create an optimized body whenever we want; A body with which we can run much faster and longer, swim and breathe underwater like fish, and even create functional wings for ourselves if we wish. The speed of our thinking will increase a millionfold, but most importantly, our survival will no longer depend on the survival of any of our bodies.

Of course, such a future can also bring potential risks. From Kurzweil's point of view, one of the most serious threats occurs when nanobots get out of control and start multiplying without limit. In such conditions, their reproduction rate may increase so much that they consume all the organic matter on the Earth's surface and eventually turn the planet into a mass of shapeless, slimy matter.

Kurzweil writes about this danger:

One of the most discussed catastrophic scenarios is the possibility of the formation of a phenomenon known as gray sludge; Self-replicating machines that consume carbon-based materials and make more self-replicating machines. This process can set off an unstoppable chain of reactions and eventually turn all of Earth's biomass into such machines.

However, if we optimistically assume that such a catastrophe will not occur and that the cybernetic singularity envisioned by Kurzweil will indeed materialize, the next important question will be what will happen to human identity. Perhaps one of the most immediate challenges is the inability to tell the difference between our real selves, digital versions, and non-biological extensions of our existence. In other words, from the 2030s and with the entry of brain-computer interfaces into the decision-making process, what will the concept of being human mean in a situation where non-biological components play a role in our decisions and choices? These non-biological components are likely to grow at an exponential rate, while our biological intelligence remains within the bounds of its natural capabilities.

Thus, by the late 2030s, a large part of human thought processes may rely on abiotic systems. In such a situation, when our thoughts are so influenced by non-biological technologies, what exactly is the place of human decision-making? Will there eventually be a way to tell the difference between real humans and non-living versions of them?