According to The Verge, OpenAI is set to launch its next-generation model, codenamed Orion, by December. Unlike previous releases, this powerful successor to GPT-4 will be available exclusively to select corporate partners at first, rather than being widely accessible through the ChatGPT platform, The Verge said, citing a source familiar with the plan.
The publication also notes that the strategic rollout appears to be taking shape behind closed doors, with Microsoft engineers already laying groundwork to host Orion on Azure as early as November. This privileged access for Microsoft, OpenAI’s primary deployment partner, underscores the increasingly intertwined relationship between the two tech giants.
However, the path to Orion’s launch has been marked by conflicting narratives. Following initial reports, OpenAI CEO Sam Altman dismissed the news as “fake,” while company spokesperson Niko Felix offered a more nuanced response, denying plans for an “Orion-coded” release while confirming upcoming “great technology.” This carefully worded statement has only intensified speculation about the model’s true nature and timeline.
What makes Orion particularly intriguing is its reported capabilities, with one OpenAI executive suggesting it could be up to 100 times more powerful than GPT-4. The model stands distinct from OpenAI’s recently released o1 reasoning system, though both appear to be pieces in the company’s broader puzzle of achieving artificial general intelligence (AGI).
Technical details have begun to emerge, with reports indicating that OpenAI utilized its o1 model (codenamed Strawberry) to generate synthetic training data for Orion. The completion of the model’s training phase was apparently celebrated by OpenAI researchers in September, coinciding with Altman’s cryptic social media post about “winter constellations.”
i love being home in the midwest.
the night sky is so beautiful.
excited for the winter constellations to rise soon; they are so great.
— Sam Altman (@sama) September 14, 2024
This developmental milestone arrives at a crucial juncture for OpenAI. The company recently secured a landmark $6.6 billion funding round, necessitating its restructuring as a for-profit entity. Simultaneously, the organization faces significant leadership changes, with the departures of CTO Mira Murati, Chief Research Officer Bob McGrew, and VP of Post Training Barret Zoph.
The timing and selective release strategy of Orion suggest a careful balancing act between maintaining OpenAI’s technological edge and managing the responsible deployment of increasingly powerful AI systems. As winter approaches, all eyes in the tech world will be watching to see if this new constellation in the AI firmament lives up to its celestial namesake’s brightness.
Leave a Reply