Introduction of an AI Model Capable of Reasoning Like Humans

The company OpenAI, a leader in AI development, announced the launch of its new product; a model of artificial intelligence that has the capability for complex reasoning and can solve difficult problems in mathematics, programming, and sciences. This new product, known as 'O One', represents a significant step towards achieving human-like understanding in machines. Mira Murati, the Chief Technology Officer of OpenAI, stated in an interview with the Financial Times, 'This new technology allows us to observe the model's thought process step by step.' She also emphasized that the 'O One' models achieved remarkable results in the International Mathematical Olympiad, scoring 83%, compared to only 13% for previous models like GPT-4. According to OpenAI, these models were integrated into the ChatGPT Plus service starting Thursday, September 13, and are specifically designed to assist scientists and developers. These models have a higher cost due to the use of more complex techniques and longer processing times for analyzing requests, but their responses are more consistent and accurate. Mark Chen, the Vice President of Research at OpenAI, demonstrated the new model solving several problems that GPT-4 could not, including an advanced chemistry question and a complex mathematical puzzle, showcasing the superior logical reasoning capabilities of this model. Murati also pointed out that this set of models could revolutionize areas such as online search. She mentioned that OpenAI is testing a new search pattern using its SearchGPT tool that performs better in research and information retrieval. Yoshua Bengio, a computer scientist at the University of Montreal and a recipient of the prestigious Turing Award, emphasized the importance of true reasoning power in AI: 'If AI systems can provide real reasoning, it allows for the continuity of facts, arguments, and results presented by AI.' He also added that the lack of this capability in AI has been a major barrier to achieving 'Artificial General Intelligence.' Artificial General Intelligence is a model of machine intelligence that can successfully perform any intellectual task that a human can do. Scientists predicted in 2023 that Artificial General Intelligence will be available to humans within less than 15 years. Advances in AI model development are significant not only technologically but also in terms of economic and social impact. These technologies are increasingly being utilized across various industries and have the potential to transform how humans interact with machines and work processes. Concurrently with the rapid and increasing advancement of AI technology, concerns about its detrimental effects on democratic elections, facilitating embezzlement and fraud, and the loss of many jobs have also risen.