OpenAI on Thursday released its newest AI model, GPT-5.5, to its paid subscribers.
The release, coming just six weeks after the company debuted GPT-5.4, is an extremely fast turnaround that underscores how fiercely frontier AI labs are competing for enterprise customers, and how their models are increasingly evolving through continuous, incremental updates.
The company also said there are 4 million active Codex users and 9 million paying business users on ChatGPT. ChatGPT also has more than 900 million weekly active users and over 50 million subscribers. It no doubt hopes those figures will undercut a narrative that has been building across social media that OpenAI has lost traction among consumers and has fallen behind its arch-rival Anthropic in the race for enterprise customers.
OpenAI president Greg Brockman called the new model a “new class of intelligence” and “a big step towards more agentic and intuitive computing” during a press briefing. He acknowledged that “there are enough model releases that it’s probably getting hard to distinguish one from another,” but said GPT-5.5 can do more with less guidance and is “way more intuitive to use—it can look at an unclear problem and figure out what needs to happen next.”
Those capabilities, he added, “feel like they’re setting the foundation for how we’re going to do computer work going forward, or how agent computing at scale will work,” particularly in areas like scientific research that are “very intelligence bottlenecked.” He pointed to an example of a math professor who used GPT-5.5 and Codex to build an algebraic geometry app from a single prompt in 11 minutes.
Brockman also emphasized efficiency gains. “It’s a faster, sharper thinker for fewer tokens compared to something like 5.4,” he said, adding that this means “more frontier AI available for businesses and for consumers.” Tokens are the basic units of data that large language models process and the basis on which AI companies charge many business customers. In text, a token is about equivalent to a word and a half.
The Bank of New York has been testing GPT-5.5 in recent weeks, alongside early access to models from rivals like Anthropic. CIO Leigh-Ann Russell said the improvements are meaningful.
“What we’re actually seeing from 5.5, that I think is really important for a highly regulated institution, is the response quality—but also a really impressive hallucination resistance,” she said. “A bank needs to have very high accuracy, so this becomes critical, and we are seeing a step change with this model.”
Anything in the bank that has better accuracy will help with scaling the company’s over 220 AI use cases, she added. “We see resiliency as being commercial. So the more quickly we can just establish that accuracy, the faster we can scale our models to have completely redesigned financial services.”