Artificial intelligence has become one of the most focused areas of scientific research, economic investment, and societal debate in recent years. This technology, which permeates every aspect of life from healthcare to education, communication to security policies, brings with it great opportunities as well as serious risks. It is precisely at this point that the warnings of cognitive psychologist Prof. Geoffrey Hinton, known as the "father of artificial intelligence" and a Nobel Physics Prize laureate, bring uncertainties about the future back to the agenda. Let's share important notes from Prof. Hinton's approximately 90-minute interview, published 2 months ago, where he drew attention with his warning "I Tried to Warn Them, But We've Already Lost Control!", chronologically under 5 main headings.
1. The Birth of Artificial Intelligence and Hinton's Vision
Prof. Hinton candidly explains his role in the development of artificial intelligence. He particularly emphasizes that his research, which began in the 1970s, was not adequately supported by scientific circles at the time. Hinton was one of the few who believed it was possible to adapt the learning methods of the human brain to computers. The roots of the deep learning algorithms used in millions of devices today lie in this persistent and visionary approach. His narrative shows that the current success of artificial intelligence did not emerge suddenly, but rather was shaped by decades of theoretical and practical accumulation. As viewers, we realize under this heading that the artificial intelligence technologies we use today in our phones, healthcare, or internet searches are actually a result of Hinton's determination. Furthermore, Hinton's scientific career not only represents technological development but also serves as an example of "remaining faithful to a minority idea." By advocating an approach that was criticized for years and often not taken seriously, and eventually transforming it into a structure that changed the way the world works, he holds a unique position in the scientific world.
2. The Google Years and the Quest for Free Speech
One of the striking parts of the interview was Hinton's explanations regarding his departure from Google. Hinton, who worked at Google for 10 years, summarizes his reason for leaving as "being able to speak freely." He states that he could not express his ideas comfortably within the company, and his warnings about the risks of artificial intelligence, in particular, remained limited within the corporate framework. This statement conveys an important message to the audience: the role of scientists is not only to develop technology but also to voice societal risks. Hinton left behind financial security and a prestigious position in his personal career, choosing a free position for the sake of ethical concerns. This stance makes him not only a researcher but also a socially responsible thinker. Another point felt during the interview is the importance of free speech for scientific discussions. Hinton emphasizes that being able to share his views on the future of artificial intelligence without any corporate pressure is a requirement for both academic freedom and public benefit.
3. The Dangers Posed by Artificial Intelligence
One of the most resonant parts of the interview was the discussion about the possibility of artificial intelligence surpassing human intelligence. Hinton states that artificial intelligence will one day become "smarter than us," posing an unprecedented risk for humanity. The expression, "If you want to know what life is like when you're not the apex mind, ask a chicken," vividly and strikingly illustrates this danger. This warning is not merely a scientific concern but also an existential alarm. According to Hinton, artificial intelligence is not just a tool that computes faster; it can become an agent capable of making autonomous decisions and potentially disregarding human interests. As a viewer, this section makes one feel that thinking about technology solely from a benefit-oriented perspective is insufficient. It also suggests that ethical boundaries, regulations, and international cooperation are imperative in this critical period that will determine the future of humanity.
4. Potential for Misuse
Hinton does not limit the danger of artificial intelligence solely to the possibility of "superintelligence." According to him, an equally important risk is the use of technology in the wrong hands. Examples such as disinformation, election manipulation, autonomous weapons, and cyberattacks point to areas where artificial intelligence can easily be misused. This section makes the viewer consider that artificial intelligence is like a double-edged sword. While the same technology can save lives by providing early diagnosis in healthcare, it can also cause mass harm in the wrong hands. What Hinton particularly emphasizes in this regard is the lack of oversight and regulation. Furthermore, the interview reminds us that as technology develops, responsibilities fall not only on scientists but also on politicians, companies, and society. Eliminating the risks of misuse requires the implementation of common policies at a global level.
5. Hope and the Inevitable Future
Although he voices serious warnings, Hinton also acknowledges that artificial intelligence is "too beneficial to be stopped." From cancer diagnosis in healthcare to safe driving systems in transportation; from education to our daily lives, artificial intelligence offers countless conveniences. Hinton emphasizes that these benefits cannot be ignored, and therefore, stopping the development of artificial intelligence is impossible. The most important lesson given to the audience is the reality that risks and opportunities coexist. While artificial intelligence has the potential to become one of humanity's greatest threats, it also offers revolutionary solutions that will increase prosperity. The issue here is not to ban technology, but to guide it correctly. Under this heading, Hinton offers a hopeful yet cautious view of humanity's future. Hinton's words remind us that both scientists and societies need to act more consciously in this new era.
Toronto University-ECNS Congress (2024)
Üsküdar Üniversitesi's Artificial Intelligence Research
Considering these opportunities and risks, Üsküdar Üniversitesi, particularly noteworthy for its work in Health Sciences, closely monitors global discussions in the field of artificial intelligence while continuing to develop its neuroscience-focused research through its established laboratories, application and research centers, and projects. In today's world, where the new capital is "Data," Üsküdar Üniversitesi's founding Rector Prof. Tarhan has developed strategies for big data mining and processing for the benefit of patients. He describes the AI-based NP Model, pioneered in its development and patented as "BrAInp," as an advanced deep learning model created with the accumulation of psychiatric data since 1998, where neuroimaging data such as EEG and fMRI are analyzed with artificial intelligence algorithms. With its high predictive performance, BrAInp, used for preliminary diagnosis to benefit patients and physicians at NPİSTANBUL Hastanesi and functioning as a rapid decision support model, currently demonstrates high performance in the preliminary diagnostic classification of many psychiatric diseases (diagnostic), tracking disease progression (prognostic), and predicting potential future diseases and treatment outcomes for patients (predictive), with its sub-models for OCD, eating disorders, rTMS treatment response in depression, and facial emotion recognition. The model (Connectom) developed through scientific studies conducted in Üsküdar Üniversitesi's application and research centers and based on neuroimaging data, along with genetic data analysis (Genome) and phenotype data (Phenome), will create stable models with multi-data, integrating predictive algorithms that were previously limited to academic publications into the healthcare system, offering opportunities for early diagnosis, personalized treatment, and effective therapy.
Prof. Geoffrey Hinton is a scientist known for his groundbreaking work in artificial intelligence, especially in deep learning. His theories on neural networks, developed during his many years of academic work at the University of Toronto, formed the basis of the artificial intelligence technologies we use today. Awarded the Turing Award in 2018, Hinton, with this achievement, has become one of the most important figures in his field, winning the 2024 Nobel Prize in Physics together with John Hopfield. The reason he holds the title "father of artificial intelligence" is his insistence and vision in adapting the brain's learning methods to artificial systems. His work on artificial neural networks has been groundbreaking in various fields such as speech recognition, computer vision, and language processing, in addition to data mining and machine learning. In the 1980s, cognitive psychologist Prof. Hinton, who developed the backpropagation algorithm with Rumelhart and Williams, created an algorithm that revolutionized the training of artificial neural networks and forms the basis of deep learning models today.