ChatGPT has crossed 400 million weekly energetic customers (WAU), with over 2 million enterprise customers leveraging the device for work, in keeping with OpenAI COO Brad Lightcap.
“We really feel very lucky to serve 5% of the world each week,” Lightcap mentioned in a latest put up on X. He additionally famous a fivefold improve in reasoning mannequin API utilization for the reason that launch of OpenAI’s o3 mini mannequin.
“2M+ enterprise customers now use Chatgpt at work, and reasoning mannequin API use is up 5x for the reason that o3 mini launch,” he mentioned.
OpenAI is making ready to introduce GPT-4.5 and GPT-5 to ChatGPT and the API. Lightcap confirmed that free-tier customers may have limitless entry to GPT-5, whereas Plus customers may have the choice to function at “even larger intelligence.”
“GPT-5 will unify our GPT and o-series fashions right into a single highly effective mannequin — we predict even low-taste testers will like it,” he added.
OpenAI additionally plans to develop ChatGPT’s agent capabilities this yr, alongside different updates but to be disclosed.
OpenAI’s development comes because it faces new competitors from China’s DeepSeek, which unsettled tech markets in January over issues about its impression on U.S. AI corporations’ profitability and dominance.
In the meantime, OpenAI chief Sam Altman is contemplating whether or not OpenaI ought to open-source its subsequent AI challenge. Altman requested customers on X whether or not they desire an o3-mini-level mannequin or a phone-sized model.
“For our subsequent open-source challenge, would it not be extra helpful to create an o3-mini-level mannequin, which is comparatively small however nonetheless requires GPUs, or the perfect phone-sized mannequin we are able to develop?” Altman requested in a ballot on X.
OpenAI is going through stiff competitors from its opponents. xAI not too long ago launched Grok-3 and made it freely obtainable to all of the customers. Grok is at the moment the No. 1 app on the app retailer, surpassing ChatGPT and DeepSeek. Unbiased benchmarks confirmed that Grok 3 outperformed Google Gemini 2 Professional, DeepSeek V3, Claude 3.5 Sonnet, and GPT-4 in checks reminiscent of AIME, GPQA, and LCB.
In the meantime, Anthropic is making ready to launch its subsequent reasoning mannequin, a hybrid AI that can allocate extra computational energy to advanced queries whereas effectively dealing with easier duties.
The put up ChatGPT Surpasses 400 Million Weekly Energetic Customers, OpenAI Plans GPT-5 Launch appeared first on Analytics India Journal.
