The firm that created ChatGPT, OpenAI, has released an updated version of their well-liked chatbot with the goal of addressing the recurring problems of incorrect math, malfunctioning coding, and sporadic fabrications. With the use of cutting-edge AI technology known as OpenAI o1, this improved ChatGPT is able to “reason” its way through complex tasks in the fields of science, coding, and maths.

The new ChatGPT may deliberate, analyse, and strategise its solutions, boosting the accuracy and dependability of its outputs, in contrast to prior models that responded quickly. With examples of the chatbot resolving challenging riddles, responding to PhD-level chemical queries, and making medical diagnoses using patient data, OpenAI demonstrated this enhanced capabilities.

The industry’s general push towards AI systems with sophisticated reasoning skills is reflected in this development. While Microsoft and GitHub are aggressively incorporating OpenAI’s breakthroughs into their businesses, Google and Meta are also exploring related technologies.

The final aim is to develop AI systems that can replicate human-like reasoning and methodically solve issues one step at a time. Scientific research, education, and computer programming could all be revolutionised by this.

Experts warn that while the new ChatGPT appears promising, artificial intelligence reasoning is still in its infancy and has limitations. It is also unclear if these capabilities will completely overcome the drawbacks of earlier models and how they will be applied in the actual world.