Recently, OpenAI, a leading company in the field of artificial intelligence, announced significant news: the upcoming release of a new open language model, inviting global developers to participate in providing feedback. This announcement was officially made on March 31, marking a new step for OpenAI in the large language model (LLM) domain.
This will be the second open language model since OpenAI's GPT-2 transitioned to a proprietary model. OpenAI's CEO, Sam Altman, revealed in a post on the X platform that the model is expected to be officially launched in the coming months.
The characteristic of an open-weight model is that users can modify it according to their needs. In the generative AI field, "weights" refer to the connections between digital neurons, and by adjusting these weights, one can alter the concepts associated with the model or the importance given to these associations. This newly launched open model from OpenAI will offer users more autonomy and flexibility.
Altman mentioned in his post that the model under development will be a "reasoning" model similar to OpenAI o1. He emphasized, "We have been contemplating this for a long time, but we have always been occupied by other priorities. Now, with continuous technological advancements and a maturing market, we believe launching such a model is very important."
To gather extensive opinions and suggestions, OpenAI is seeking feedback on this open model from global developers, researchers, and the broader community. On OpenAI’s official website, a feedback application form has been launched. Interested developers need to provide links to their GitHub profiles along with social or professional profile links. The application form also asks applicants about which open models they have used before, what features they hope to see in OpenAI's open-weight model, and how they plan to use the model.
In addition, OpenAI plans to host several events in San Francisco, Europe, and the Asia-Pacific region to collect more developer feedback and let them experience early prototypes. Altman revealed that the event in San Francisco will take place in the coming weeks, with other regional events to follow.
Before the release, OpenAI will evaluate the model according to a prepared framework to ensure its stability and usability. Considering that the model might be extensively modified and used after its release, OpenAI will also undertake additional work to address potential challenges and issues.
It is worth noting that the definition of "open source" in the artificial intelligence field has always been somewhat controversial and ambiguous. An organization named the Open Source Initiative defines open-source AI as a model that can be used for any purpose, modified in any way, studied to observe its output methods, and shared with others. Such a definition helps us better understand the essence and characteristics of open-source AI and provides clearer positioning and explanation for OpenAI's newly launched open model.
With the continuous development and popularization of artificial intelligence technology, open language models will become one of the important trends in future technological advancement. OpenAI's newly launched open model will not only provide developers with more autonomy and flexibility but also inject new vitality and momentum into the innovation and development of artificial intelligence technology. We look forward to OpenAI continuing to lead the development trend in the artificial intelligence field and making greater contributions to global technological innovation and progress.
