AI-controlled zones determined to remain human-operated by Russians
In a recent statement, Yaroslav Bespalov, the head of the "Multimodal AI Architectures" group at the "Strong AI in Medicine" lab of the AIRI Institute, has emphasised the difficulties in creating artificial intelligence (AI) that can replicate the thinking patterns of high-class specialists.
Bespalov, a renowned figure in the field of AI, believes that the current AI systems fall short when it comes to learning from unique ways of thinking, such as that of researchers. This is due to the lack of data representing such diverse and complex thought processes.
AI primarily learns from large datasets that represent "average" thinking. However, fields that require deep creativity, conceptual thinking, ethics, empathy, and abstract understanding of the world, such as philosophy, fundamental science, art, and education, are the most challenging to automate. AI usually falls short of the best specialists in these areas, according to Bespalov.
Bespalov highlighted the need for larger and more diverse datasets to train AI models that can mimic the thinking patterns of high-class specialists. He believes that the volume of data needed to create a universal AI capable of thinking like a group of high-class specialists is too small, and the current data available is insufficient to train an AI model specifically on a unique way of thinking of researchers.
Despite the strides made in AI, Bespalov predicts that it will take approximately 20 to 30 years for AI to surpass human intelligence in fields that require context, intuition, and personal experience. This contrasts with Elon Musk's previous claim that AI would exceed human intelligence in just two years.
Bespalov's statement underscores the challenges in developing AI that can replicate human intelligence, particularly in areas that require deep thought, creativity, and understanding. While AI can outperform humans in certain tasks like predictive analytics and pattern recognition, it will not replace humans in fields such as philosophy, fundamental science, art, education, and other areas where context, intuition, and personal experience matter in the next 10 years.
Artificial intelligence (AI) struggles to replicate the thought processes of high-class specialists, including those in research, due to the lack of diverse and complex data available to train AI models. Furthermore, the current AI systems primarily learn from "average" thinking, making fields like philosophy, fundamental science, art, and education, which require deep creativity and understanding, difficult to automate.
