In a recent podcast, OpenAI CEO Sam Altman voiced significant apprehension about the company’s upcoming AI model, GPT-5. He compared its rapid development to the “Manhattan Project,” revealing a sense of unease and a lack of control over the new technology. Altman’s comments suggest that GPT-5 could be a pivotal—and potentially unsettling—step in the evolution of artificial intelligence.
The Alarming Pace of AI Development
Altman has previously been a vocal champion for his company’s work, famously calling GPT-4 “the dumbest model any of you will ever have to use again.” However, his recent remarks on the “This Past Weekend with Theo Von” podcast strike a different, more cautious tone. He stated that the new model’s development “feels very fast” and that he had moments of profound concern, similar to a scientist asking, “What have we done?”
His comment that “it feels like there are no adults in the room” suggests that the technology is advancing so quickly it’s outpacing the safety protocols and oversight designed to contain it. While he didn’t specify the exact capabilities of GPT-5 that caused this alarm, his words have fueled speculation about its potential to be a groundbreaking and potentially destabilizing force.
The Link to AGI and Corporate Stakes
The development of GPT-5 is not happening in a vacuum; it is deeply intertwined with OpenAI’s pursuit of Artificial General Intelligence (AGI). This pursuit is at the center of the company’s complex relationship with its primary backer, Microsoft.
The partnership agreement between the two tech giants dictates that Microsoft’s access to OpenAI’s technology could be limited once the AGI benchmark is achieved. This has led to high-stakes negotiations between the companies, with each side maneuvering to protect its interests. The possibility of OpenAI declaring AGI to sever ties with Microsoft before their 2030 partnership expires adds a layer of corporate intrigue to the release of GPT-5.
While there have been reports of Microsoft seeking to extend its partnership beyond 2030, this tension underscores the immense value and potential risks associated with the new AI models. GPT-5 is rumored to be a sophisticated, unified system with advanced reasoning capabilities and the potential for autonomous, multi-step task completion.
As the rumored August launch date for GPT-5 approaches, Altman’s cautionary words serve as a stark reminder of the ethical and societal implications of advancing AI technology, leaving many to wonder if the benefits will outweigh the risks.
