As artificial intelligence technology accelerates beyond previous limits, even the minds behind its development are expressing alarm. OpenAI CEO Sam Altman, one of the most influential figures in AI, has admitted he feels “scared” about what’s coming next—ChatGPT-5.
Set to launch in August 2025, GPT-5 is anticipated to be exponentially faster and more capable than its predecessor, GPT-4. But this rapid leap in capabilities has left Altman and other experts questioning whether humanity is truly ready for what lies ahead.
In a candid interview on Theo Von’s This Past Weekend podcast, OpenAI CEO Sam Altman disclosed that testing GPT-5 gave him a profound sense of unease.
“It feels very fast,”
he said, describing not only the model’s extraordinary processing power but the unsettling pace of AI development itself. His remarks evoked parallels with the Manhattan Project—the top-secret World War II endeavor that developed the atomic bomb.
Altman reflected on the weight of that comparison:
“What have we done?”
he asked, pointing to the existential reckoning many scientists faced after unleashing a technology far beyond society’s control or comprehension.
Among the most telling moments in the podcast was Altman’s concern about governance in AI:
“It feels like there are no adults in the room.”
This blunt statement speaks to the lack of robust regulatory structures worldwide, as developers race ahead without adequate oversight.
Although OpenAI continues to advocate for the responsible development of AI systems, Altman’s comments hint that even industry leaders feel they may be moving too fast for safety protocols—and society itself—to keep up.
While the technical specifics of GPT-5 are still under wraps, insider reports suggest it will bring substantial improvements:
Enhanced multi-step reasoning
Longer memory capabilities
More advanced multimodal interactions
Near-instant response times with reduced model-switching delays
Altman himself has boldly remarked that GPT-4
“kind of sucks”
when compared to the capabilities of GPT-5, setting high expectations—and raising deeper concerns—about what’s to come.
Artificial General Intelligence (AGI)—AI systems that can perform any intellectual task that humans can—is OpenAI’s long-term vision. Sam Altman has previously claimed that AGI might arrive subtly, without major societal disruption. But his recent tone has shifted markedly. The release of GPT-5 could bring humanity closer than ever to AGI-like capabilities, potentially without a shared international plan to handle its implications.
Sources close to the company suggest that OpenAI is facing increasing pressure from investors to transition into a for-profit model before the end of 2025. Microsoft, which holds a $13.5 billion stake in the company, may play a significant role in what happens next. Some reports even claim that OpenAI could prematurely declare AGI-level success to trigger a clause that would end its contract with Microsoft, drastically shifting the balance of power in the AI industry.
Competitors like Google DeepMind, Anthropic, and Perplexity.ai are also rapidly advancing, making the GPT-5 rollout a defining moment not just for OpenAI but for the future of global AI leadership.
While some skeptics argue that Sam Altman’s dramatic comparisons to the Manhattan Project may be part of a publicity strategy, others believe his concerns are sincere. Altman isn’t afraid of AI turning evil—he’s afraid of humans losing their grip on a technology they created but no longer fully understand or control. In an industry where faster is often seen as better, Altman’s warnings may be the first genuine pause for reflection in a race that’s unfolding faster than the world can grasp.
As the world awaits the official release of ChatGPT-5 in August 2025, Sam Altman’s stark warnings offer a sobering perspective. While the excitement around the model’s revolutionary upgrades continues to build, Altman’s fear suggests that technological advancement without corresponding governance may lead humanity into uncharted and potentially perilous territory. If GPT-5 truly brings us closer to AGI, society will need not just technical breakthroughs, but collective wisdom, ethical foresight, and global cooperation. The future of AI is here—but are we ready for it?