OpenAI has successfully canceled the discharge of o3, which was slated to be the corporate’s subsequent main AI mannequin launch, in favor of what CEO Sam Altman is looking a “simplified” product providing.
In a post on X on Wednesday, Altman mentioned that, within the coming months, OpenAI will launch a mannequin known as GPT-5 that “integrates quite a lot of [OpenAI’s] know-how,” together with o3, in its AI-powered chatbot platform ChatGPT and API. Because of that roadmap resolution, OpenAI now not plans to launch o3 as a standalone mannequin.
The corporate initially mentioned in December that it deliberate to launch o3 someday early this 12 months.
“We wish to do a greater job of sharing our supposed roadmap, and a a lot better job simplifying our product choices,” Altman wrote within the submit. “We wish AI to ‘simply work’ for you; we understand how sophisticated our mannequin and product choices have gotten. We hate the mannequin picker [in ChatGPT] as a lot as you do and wish to return to magic unified intelligence.”
Altman additionally introduced that OpenAI plans to supply limitless chat entry to GPT-5 on the “commonplace intelligence setting,” topic to “abuse thresholds,” as soon as the mannequin is mostly out there. (Altman declined to supply extra element on what this setting — and these abuse thresholds — entail.) Subscribers to ChatGPT Plus will be capable to run GPT-5 at a “greater degree of intelligence,” Altman mentioned, whereas ChatGPT Professional subscribers will be capable to run GPT-5 at an “even greater degree of intelligence.”
“These fashions will incorporate voice, canvas, search, deep analysis, and extra,” Altman mentioned, referring to a spread of options OpenAI has launched in ChatGPT over the previous few months. “[A] prime objective for us is to unify [our] fashions by creating techniques that may use all our instruments, know when to assume for a very long time or not, and usually be helpful for a really big selection of duties.”
Earlier than GPT-5 launches, OpenAI within the subsequent a number of weeks will launch GPT-4.5, code-named “Orion,” which Altman says would be the firm’s final “non-chain-of-thought mannequin.” Not like o3 and OpenAI’s different so-called reasoning fashions, non-chain-of-thought fashions are usually much less dependable in domains like math and physics.