There’s been a lot of hype building around OpenAI’s long-anticipated GPT-5 model over the past few years. This can mostly be attributed to the slim details about the AI model that have slipped through the cracks.
In 2024, OpenAI CEO Sam Altman promised “with a high degree of scientific certainty” that GPT-5 will be smarter than the “mildly embarrassing at best” GPT-4. The executive further indicated that ChatGPT would sport better performance and accuracy once the model ships.
According to Sam Altman:
“GPT-4 is the dumbest model any of you will ever have to use again by a lot.”
More recently, the company revealed its plans for GPT-5 during an AMA (ask me anything) session on Reddit. The session was primarily centered on OpenAI’s new AI coding assistant, Codex, designed to help developers navigate and automate common software engineering workflows.
While responding to a question about how GPT-5 will work once it ships alongside tools like Visual Code and whether it would be able to do things for humans on computers like creating documents and organizing files, OpenAI’s Vice President of Research, Jerry Tworek, indicated:
“GPT-5 is our next foundational model that is meant to just make everything our models can currently do better and with less model switching.”
The executive further indicated that OpenAI already offers a product that can help users automate several tasks on their PCs, dubbed “Operator”. He added that while the tool is still in preview, the ChatGPT maker plans to ship a significant upgrade for the product, which will make it more useful.
Interestingly, the executive indicated that the company would eventually develop a unified user experience, making Deep Research, Operator, Codex, and Memory accessible in one convenient location.
This news comes a little over a month after OpenAI announced its plans to retire GPT-4 from ChatGPT, effective April 20, 2025, replacing it with GPT-4o. However, it doesn’t come as a surprise, as Sam Altman hasn’t always been GPT-4’s biggest fan. He even admitted that the model “kind of sucks.”
Elsewhere, OpenAI CEO Sam Altman revealed that GPT-5 might work like a virtual brain with deeper thinking capabilities. Interestingly, he even suggested that the AI-powered model might be smarter than he is.
