No ChatGPT-5 yet, OpenAI's CEO Sam Altman confirms
Despite rumors, Sam Altman, the CEO of OpenAI, has confirmed that the company is not currently training GPT-5 and has no plans to do so anytime soon. This statement was made to clarify an earlier letter that claimed OpenAI was training GPT-5, which Altman referred to as "sort of silly."
Instead, the company is focusing on upgrades and updates for GPT-4 while addressing safety concerns related to the model. Altman emphasized the importance of taking the time to study the safety of the model, noting that OpenAI spent over six months training GPT-4 before its public release. As capabilities become more serious, the safety bar must increase, and moving with caution and rigor for safety issues is critical.
Altman also addressed concerns related to ChatGPT, stating that the open letter, allegedly signed by Elon Musk, lacks technical nuance about where the pause is needed. While OpenAI plans to continue with transparency, the company has not released any information on the AI bot's training data or its architecture, construction, or other true inner workings.
"Given both the competitive landscape and the safety implications of large-scale models like GPT-4, this report contains no further details about the architecture (including model size), hardware, training compute, dataset construction, training method, or similar," reads a report published by the company alongside the GPT-4 release.
