Sam Altman Reveals: OpenAI to Launch o3-mini

To maintain its leading position in the AI technology race, OpenAI is set to launch its newest artificial intelligence model, the o3-mini. In a blog post, OpenAI's CEO Sam Altman confirmed that the initial version of the new reasoning AI model, o3-mini, has been completed and will be officially released within weeks after external safety researchers have conducted tests.

Notably, back in December 2024, the developers of ChatGPT formally announced the development and testing of advanced reasoning models o3 and o3-mini. These models are designed to tackle complex tasks from various fields such as science, mathematics, and programming, offering significantly enhanced reasoning capabilities compared to the existing o1 model.

With Altman confirming the completion of the final version of the o3-mini model, it is expected to be released by the end of January, followed by the more extensive o3 model. The previous o1 reasoning AI model was introduced in September 2024. This upcoming o3 model (not named o2 due to trademark issues) represents a substantial upgrade, with OpenAI hinting at achieving a certain level of Artificial General Intelligence (AGI) under specific conditions, although the company remains cautious about labeling it as an AGI-ready model.

Unlike the o1 model which focuses on solving harder queries by increasing processing time, these new large language models are engineered to handle more challenging tasks. Additionally, Sam Altman revealed in his blog post that OpenAI plans to release APIs and ChatGPT concurrently with the new models.

In the realm of advanced reasoning AI models, OpenAI is not alone. Google also unveiled a new reasoning AI model called "Gemini 2.0 Flash Thinking Experimental" in December last year, indicating an intensifying competition in the AI field.

As part of its monetization strategy, OpenAI continues to roll out new features. Earlier this week, the company launched a beta feature called Tasks for ChatGPT, similar to AI virtual assistants like Apple's Siri and Amazon's Alexa, expanding ChatGPT's application scenarios.

Prior to this, OpenAI also confirmed it was testing a mobile registration-only feature for ChatGPT users in the US and India. Furthermore, reports suggest that the company has developed a new AI model named "GPT-4b Micro," aimed at biological research. This AI model is trained to optimize protein factors to enhance their functions, showcasing OpenAI's strong expertise in protein engineering. Additionally, the company has developed an AI language model capable of transforming ordinary cells into stem cells, advancing the field of bioscience.

In summary, through the introduction of models like o3-mini, OpenAI not only solidifies its leadership in the AI sector but also lays a robust foundation for future technological competitions.