OpenAI Season
As the AI industry gets more and more crowded — and its hype more diffuse — OpenAI CEO Sam Altman has quietly announced that the company won’t be releasing its super-advanced new reasoning model as a standalone product.
In a statement posted on his arch-nemesis Elon Musk’s social network, Altman wrote circuitously around the previously slated launch of o3, the latest awkwardly-named version of its frontier reasoning model that’s said to sometimes cost more than $1,000 worth of computing power per query.
“In both ChatGPT and our API, we will release GPT-5 as a system that integrates a lot of our technology, including o3,” the CEO said. “We will no longer ship o3 as a standalone model.”
That buried bit of the lengthy statement is, as TechCrunch flagged, a stark reversal from recent comments from OpenAI executives who claimed that o3’s full, standalone launch was slated for launch in February or March.
According to Altman, this product launch mixup is part of a greater drive to simplify what OpenAI is selling.
“We realize,” he wrote, “how complicated our model and product offerings have gotten.”
Be that as it may, scuttling o3 this close to its purported launch — and so soon after the Chinese AI company DeepSeek issued a ringing blow to OpenAI and every other mainstream AI company with its claims of a low-energy, high-powered chatbot — seems telling.
Since unleashing ChatGPT onto the world in November 2022, OpenAI has struggled to create as many waves as that first world-changing announcement. With each subsequent GPT large language model (LLM) launch being incorporated into the company’s suite of products, there has, of course, been ample splash — but with each new one, there’s been demonstrably less.
With its claims of self-fact-checking, the launch of o1 last fall indeed got some splash. But it faded quickly — and when’s the last time you heard about something actually exciting happening with it?
Were Altman et al looking to avoid a bellyflop if o3 didn’t turn out to be as much of a step forward as its predecessors? At this point, we’ll probably never know — but one thing’s for sure: a lot is going to be riding on whether GPT-5 can really break through.
More on OpenAI hype: Researchers Replicate OpenAI’s Hot New AI Tool in 24 Hours