OpenAI's big spoiler: GPT-5 is about to be released, and the intelligent experience is about to take a leap forward!

OpenAI’s next big move has finally surfaced, and it affects every aspect of our future interactions with AI.

According to the latest news, OpenAI will continue to advance its grand blueprint - using  GPT-5  as the core to build a truly unified AI platform. What does this mean? Simply put, the future AI experience will be more seamless and efficient!

Jerry, vice president of research at OpenAI, recently revealed in an AMA (Ask Me Anything) session in the Reddit community that GPT-5 will be their next-generation basic model. Their goal is not disruptive innovation, but to strive for excellence, to take the capabilities of existing models to a higher level, and - highlight - to reduce switching between models !

Imagine that you no longer need to switch back and forth between code generation tools, research assistants, screen operation agents, and memory systems, but have an intelligent system that can work together and is seamless. Jerry emphasized that components such as Codex code assistants, Deep Research, Operators, and memory systems will be more closely integrated, and ultimately presented to users is a unified agent .

Of particular note is ChatGPT's "screen agent"  Operator , which is about to receive a major update! This tool, which is still in the research stage, has demonstrated preliminary browser control and other capabilities. Although it is not stable enough,  Jerry predicts that the upcoming update (expected to be "soon") is expected to make it a "very useful tool." Does this mean that future AI will be able to understand and operate our digital world more directly? It's exciting to think about it!

Of course, OpenAI's journey to perfect integration was not without its challenges. They admitted that full integration was more challenging than expected, so we saw the release of independent reasoning models such as o3 and o4-mini. But this did not shake their determination to build a unified platform.

Token demand continues to soar, and AI’s potential goes far beyond this

Jerry also talked about a topic that we may not pay much attention to, but is crucial - the use of tokens. With the deepening of AI applications, the demand for tokens is also growing. Some people imagine that in the future, AI assistants may need to continuously process a large number of tokens to perceive the environment and analyze information. Will this trend reach its ceiling?

Jerry’s answer is: No! He believes that the use of tokens is a balance between utility and cost, and both aspects are continuing to improve. Even if the model itself no longer makes leaps and bounds, AI can release huge value simply by expanding its scale. This explains why it is necessary to invest heavily in building infrastructure that can handle massive amounts of tokens.

At the same time, Jerry also gave a reassuring judgment: AI will not completely replace human labor. He firmly believes that there will always be some jobs that are unique to humans. And in the future, we may play the role of supervising AI systems to ensure that they always serve the best interests of mankind.

Say goodbye to benchmark-based thinking and embrace pragmatism

When asked about the comparison between GPT and other leading models such as Claude and Gemini, Jerry was particularly pragmatic. He believed that traditional benchmarks can no longer truly reflect the actual application effect of the model and are easily affected by targeted optimization.

OpenAI pays more attention to the performance of the model in real-world tasks . Only in this way can we truly evaluate whether a system can solve practical problems. Their long-term goal is that users no longer need to worry about which model to choose, because the system will automatically choose the one that best suits the current task.

As  Jerry said, “Different models and products have different advantages, but our goal is to solve this difficulty by providing the best model.”

Summary and Outlook

This series of actions by OpenAI undoubtedly indicates that AI technology will develop in a smarter and more unified direction. GPT-5 as the core engine, coupled with the evolving Operator and other tools, will greatly improve the efficiency and experience of our interaction with AI.

Are you also looking forward to the future development of OpenAI? How do you think a unified AI platform will change our work and life? Feel free to share your views in the comments section!