On December 5, 2024, OpenAI launched a new series of daily livestreams called “12 Days of OpenAI,” showcasing various AI innovations. Among the highlights was the introduction of the video AI Soraund and SearchGPT, making these tools accessible to the public. OpenAI also announced that ChatGPT can now be used via phone or WhatsApp.
OpenAI began this series with the introduction of a new AI model, GPT-o1, designed to handle complex thinking tasks. Users who subscribe to the ChatGPT Pro plan for $200 per month can activate the Pro mode for GPT-o1, which enhances the chatbot’s ability to solve complex mathematical and programming problems.
Some users might want to wait before subscribing, as on the twelfth and final day of the “12 Days of OpenAI,” the company unveiled even more powerful “thinking” AI models, o3 and o3 Mini. These models are currently in the internal testing phase. OpenAI has invited safety and research experts to apply for early access to test these models before their public release.
According to OpenAI, the o3 model surpasses its predecessor, o1, with a 22.8% improvement in optimized coding performance. It also performs significantly better in mathematics and science compared to previous models.
In the ARC-AGI test, which evaluates how efficiently an AI system can acquire new skills beyond its trained data, o3 exceeded “human level” performance. This achievement has led OpenAI to promote o3 as a step towards Artificial General Intelligence (AGI).
An interesting detail is that o3 directly follows o1. OpenAI skipped the expected model number o2 due to potential trademark issues with the telecommunications company Telefonica.
OpenAI’s new developments highlight the rapid advancements in AI technology, pushing the boundaries of what AI systems can achieve. The introduction of these models reflects OpenAI’s commitment to advancing AI capabilities and exploring the potential of AGI.
OpenAI continues to innovate and expand the reach of AI technologies, making them more accessible and efficient for users. The introduction of these new models and features demonstrates the company’s dedication to enhancing the functionality and performance of AI systems.
As OpenAI tests these new models, the AI community eagerly anticipates their public release and the potential they hold for various applications. The advancements in AI technology promise to bring significant improvements in areas such as coding, mathematics, and science, further solidifying AI’s role in solving complex problems.
OpenAI’s focus on safety and research during the testing phase underscores the importance of responsible AI development. By involving experts and researchers, OpenAI aims to ensure that these new models are safe and effective before they are widely available.
With these developments, OpenAI is poised to lead the way in AI innovation, driving progress towards more advanced and capable AI systems that can benefit a wide range of industries and users.