On December 31, 2024, OpenAI CEO Sam Altman posted on X (formerly Twitter) asking for suggestions to improve its AI platform. After reading the comments, he concluded that one of the most common topics (suggested by users) for improving OpenAI is to introduce AGI (Artificial General Intelligence). Since then, rumors began to arise that OpenAI is considering integrating AGI into its platforms. These rumors were further fueled by Altman’s recent blog post (titled Reflections), which claimed that the company is confident about creating AGI. But now, while extinguishing the fire with water, Altman has revealed that the company has not even created AGI yet.

He posted that the company is planning to release “cool stuff” next month, but AGI is not part of the plans. He also said that these rumors were based only on Twitter (now X) promotion and nothing more. Since the hype was so out of control, he urged users that even if OpenAI plans to launch interesting updates and features, they need to cut down their expectations 100 times.

But why is AGI such a distant thought for OpenAI? Artificial General Intelligence or AGI is defined as an AI system that can tackle any task just like a human. After AGI, we reach artificial superintelligence (ASI), which is AI that exceeds the capabilities of the human brain. Right now, we are working on AI chatbots like ChatGPT that uses natural language processing to create human-like conversational dialogues.

In the same blog post he said, “As we move closer to AGI, it seems like an important time to look at our company’s progress. There’s still a lot to understand, there’s still a lot we don’t know, And it’s still very early. But we know a lot more than when we started.” This shows that OpenAI still has a lot of work to do to achieve AGI-level. Now, the question is: If not AGI, what’s coming for OpenAI?

OpenAI may launch AI agent soon

Those who keep an eye on the headlines would think that GPT-5 should be the next step for OpenAI. However, some recent reports have suggested that the next flagship model is still in the pipeline, with an undetermined launch date. Rather, the company wants to work on an AI super-agent, and it could be released as early as January 30. There is also a new report from Axios that claims OpenAI is meeting with US government officials regarding the launch.

“We believe that, in 2025, we could see the first AI agents “joining the workforce” and materially changing companies’ output,” Altman said in the blog. We believe That putting great tools in people’s hands over and over again results in great, widely distributed results.”

This AI agent is a sophisticated software system that can perform complex tasks on its own with minimal human supervision. It can make decisions on your behalf, adapt and respond to different situations, analyze large amounts of data and multiple options. It is basically able to mimic human intelligence when you give it a goal-oriented task. And reports suggest that the AI ​​super agent that OpenAI is working on will be able to solve complex problems with the efficiency of a doctoral graduate. Read more about the upcoming OpenAI AI-Agent here.

Therefore, we can say that this AI-agent is one step closer to AGI. At this pace we can also guess that AGI is also not far away from becoming a reality.