openai (3) OpenAI has taken a bold step forward in AI development with the release of its latest model, o1, which it claims can “think” and “reason” in a way that mimics human cognition. This shift marks a new phase in the evolution of generative AI, with the company positioning the model as a game changer for solving complex problems in fields like science, coding, and mathematics. But what does it mean for AI to “think like a person,” and how transformative is this new technology? Thinking Like a Human
The o1 model, unveiled in September 2024, is designed to address one of the major limitations of previous AI models: their inability to truly reason. While earlier versions of ChatGPT were capable of generating responses by drawing from vast amounts of training data, they often struggled with tasks that required logical connections or multi-step reasoning. OpenAI claims that o1 can now solve these kinds of problems by thinking through them in a way similar to humans.
Read More: OpenAI’s o1 is here to solve the world’s toughest challenges
For instance, the o1 model excels in handling follow-up questions that require reasoning. When asked, “Who is Tom Cruise’s mother?” (Answer: Mary Lee Pfeiffer), and then asked, “Who is Mary Lee Pfeiffer’s son?” the model correctly answers Tom Cruise—demonstrating its ability to make connections across queries, something previous models often stumbled on. A Slower, More Thoughtful AI
However, this new reasoning capability comes with a trade-off: speed. Unlike earlier models that focused on delivering responses quickly, the o1 model takes its time to think through complex tasks. OpenAI describes this as deliberate, much like how a human might slow down to reason through a difficult problem.
In tests, o1 has demonstrated impressive capabilities, particularly in science and mathematics. OpenAI reports that the model performs on par with PhD students in difficult tasks across disciplines such as physics, chemistry, and biology. For instance, it solved 83% of problems in the qualifying exam for the International Mathematics Olympiad. Despite its slower response time, the model’s ability to think through problems with accuracy is […]

Health ministry mulls legalizing marijuana for medical use
The health ministry’s expert panel starts discussing revising the Cannabis