Openai CEO Sam Altman has never overcome the risks of artificial intelligence, but his latest comments about GPT -5 have sent shockwaves through the technical world. Recently in the presence of podcast, Altman admitted that he was scared by testing the GPT-5, comparing its development to the Manhattan project-WWII-era initiative that created a atomic bomb.
Since OpenIE prepares to launch GPT -5 in early August 2025, ALTman’s concerns highlight the moral and existence dilemmas around AI’s rapid progress. But what is really worried about him? And should the rest of us also be worried?
Unprecedented capabilities of GPT-5
The GPT-5 is expected to have a monumental leap on its predecessor, GPT-4, with a significant improvement in logic, speed and multimodal processing (lessons, images and possibly videos). Altman revealed that during the internal test, GPT -5 answered a complex question that he could not understand.
“I fed it into the model—GPT-5—and it responded to the entire thing,” Altman said.
This moment underlined the ability of GPT-5 to improve humans in some intellectual works, raising questions about dependence on AI for decision making.
For more tech news Rozbaat
Why Is Altman Scared?
- Lack of Oversight
Altman warned that AI development is outpacing regulation, stating, “There are no adults in the room” when it comes to governance. Without proper safeguards, unchecked AI could lead to misuse, misinformation, or unintended consequences. - Manhattan Project Comparison
By likening GPT-5 to the Manhattan Project, Altman hinted at its transformative—and potentially destructive—potential. The analogy suggests that, like nuclear technology, AI could be a force for both progress and peril. - Ethical and Societal Risks
Altman voiced concern about society giving too much power to AI, stating, “Something is unsettling and risky about collectively choosing to live our lives according to what AI dictates.”
The Balancing Act: Innovation vs. Control
Despite his fears, OpenAI isn’t slowing down. GPT-5 is set to launch with three versions—standard, mini, and nano—to cater to different user needs. The model will also integrate OpenAI’s o3 reasoning engine, making it more versatile than ever.
However, Altman’s candid remarks suggest a growing tension:
- On one hand, GPT-5 could revolutionize industries, from healthcare to education.
- On the other, its power demands urgent regulatory frameworks to prevent misuse
Keywords:
- GPT-5 release date
- OpenAI CEO Sam Altman fears GPT-5
- GPT-5 capabilities
- Is GPT-5 dangerous?
- GPT-5 vs GPT-4
- Why is Sam Altman scared of GPT-5?
- GPT-5 ethical concerns
- OpenAI Manhattan Project comparison
- Will GPT-5 replace humans?
- GPT-5 regulation and risks
One thought on “OpenAI CEO Admits Fear Over GPT-5’s Potential”