The anticipation surrounding the release of OpenAI’s open model has taken another turn, as CEO Sam Altman announced a delay in its launch. Initially scheduled for next week, the release has now been postponed indefinitely to allow for more comprehensive safety evaluations. This decision comes after a previous delay earlier this summer, highlighting the company’s commitment to ensuring the model’s reliability and safety.
In a recent update, Altman emphasized the importance of conducting thorough safety tests and examining potential high-risk areas before the model’s release. He stated, “We need time to run additional safety tests and review high-risk areas. We are not yet sure how long it will take us.” This cautious approach reflects OpenAI’s dedication to responsible AI development, acknowledging that once the model is released, it cannot be retracted.
The open model’s launch is one of the most awaited events in the AI community this summer, alongside the anticipated introduction of GPT-5. Unlike GPT-5, which will be a proprietary offering, the open model is designed to be freely accessible for developers to download and operate locally. This strategic move aims to reinforce OpenAI’s position as a leading AI research entity amidst growing competition from other tech giants and startups.
As developers await the release of this groundbreaking model, they will have to exercise patience. Reports suggest that the open model is expected to exhibit reasoning capabilities comparable to OpenAI’s previous series of models, with aspirations for it to be a top-tier option among open-source alternatives.
The competitive landscape for open AI models has intensified recently, with the launch of Kimi K2 by a Chinese AI startup, which boasts one trillion parameters and has reportedly outperformed OpenAI’s GPT-4.1 on various coding benchmarks. This development underscores the urgency for OpenAI to deliver a robust and innovative open model.
During the initial announcement of the delays, Altman hinted at achieving something remarkable, though he did not provide specific details. This sense of mystery adds to the excitement surrounding the eventual release.
Aidan Clark, OpenAI’s VP of research leading the open model initiative, expressed confidence in the model’s capabilities but acknowledged the high standards set for open-source releases. He stated, “Capability-wise, we think the model is phenomenal — but our bar for an open-source model is high, and we think we need some more time to ensure we’re releasing a model we’re proud of along every axis.” This commitment to excellence is crucial as the company navigates the complexities of AI development.
In related news, discussions have emerged regarding the potential for the open model to connect with cloud-hosted AI systems for more complex queries. However, it remains uncertain whether these features will be included in the final version of the open model.