Race for the future ChatGPT-5 and the contest over superintelligent AI

A significant step forward but not a leap over the finish line. That was how Sam Altman, the chief executive of OpenAI, described the latest upgrade to ChatGPT this week.

- Dan Milmo Dara Kerr

The race Altman was referring to was artificial general intelligence (AGI), a theoretical state of AI where, by OpenAI's definition, a highly autonomous system is able to do a human's job.

Describing the new GPT-5 model that will power ChatGPT as a "significant step on the path to AGI", Altman added a hefty caveat. "[It is] missing something quite important; many things quite important," he said, such as the model's inability to "continuously learn" after its launch. In other words, these systems are impressive but have yet to crack the autonomy that would allow them to do a full-time job.

OpenAI's competitors are straining for the tape too. Last month, Mark Zuckerberg, the chief executive of Facebook's parent company, Meta, said development of superintelligence - another theoretical state where an AI far exceeds human cognitive abilities - was "now in sight".

Google's AI unit outlined its next step to AGI on Tuesday by announcing an unreleased model that trains AIs to interact with a convincing simulation of the real world, while Anthropic, another company making significant advances, announced an upgrade to its Claude Opus 4 model.

So where does this leave the race to AGI and superintelligence?