OpenAI is set to unveil an open-source artificial intelligence (AI) model, marking the San Francisco-based company’s first foray into sharing an open-source model focused on reasoning. The anticipated release is slated for the upcoming months and follows the last open-source initiative by the firm, which was the GPT-2 model launched in November 2019. The organization is actively gathering input from the developer community to gain insights into their needs and expectations. A critical aspect of this model’s development will center around safety measures.
OpenAI Planning to Release an Open-Source AI Model
The open-source landscape for AI has seen a surge in competition, with community-centric players such as Meta, Mistral, and Alibaba, alongside tech titans like Google and Microsoft, all introducing various open models. In contrast, OpenAI has remained on the sidelines since the GPT-2 release, concentrating on proprietary software that is neither available for download nor modification for research or commercial use.
Earlier this year, during an ask-me-anything (AMA) session on Reddit, OpenAI CEO Sam Altman discussed the company’s stance on open-source initiatives. He acknowledged that OpenAI had been “on the wrong side of history” in this regard and expressed the need to devise a new open-source strategy, although he clarified that it was not the foremost priority for the company.
On Monday, Altman shared an “Open model feedback” form in a post on X (formerly Twitter), indicating a shift toward developing an open-weight AI model featuring advanced reasoning capabilities. For those unfamiliar, the weights refer to the numerical connections the AI utilizes to produce outputs.
While this aspect is crucial for the functionality of the AI model, the absence of publicly available datasets and training methodologies hinders the ability to replicate or construct the model from scratch for integration into other systems. Such proprietary AI models are often classified as partially open.
Altman emphasized that ensuring the safety of the model will remain a pivotal focus for OpenAI, especially since modifications will occur after its release. The company intends to utilize its Preparedness Framework to guide these efforts.
