OpenAI CEO Sam Altman has described GPT-5 as a breakthrough unlike anything humanity has seen before, marking a dramatic step forward in artificial intelligence capabilities. According to Altman, the new model represents a leap especially in software development, logical reasoning, and written communication, with performance levels far surpassing its predecessors.
Altman drew a striking comparison to illustrate this progress. “GPT-3 felt like a high school student to me, GPT-4 like a university student. But GPT-5, for the first time, feels like a PhD-level expert on any subject,” he said, emphasizing the depth and sophistication of the new system’s abilities.
Fewer Hallucinations, More Trustworthy Responses
One of the most notable upgrades in GPT-5 is its reduced tendency to produce hallucinations — false or fabricated information that has plagued earlier AI models. OpenAI claims the system is more truthful, fact-driven, and human-like in its interactions. It is said to produce fully functional software, reasoning through tasks step-by-step, drawing inferences, and applying logical thought in ways that feel far more natural.
These advances aim to create a smoother and more reliable experience for users, addressing one of the most common criticisms of generative AI: its occasional confidence in delivering incorrect answers.
Skepticism From Experts
While OpenAI presents GPT-5 as a revolutionary upgrade, some experts urge caution. Professor Carissa Véliz of the Oxford Institute for AI Ethics suggested that such claims may be partly driven by marketing strategy. She noted that these systems still cannot truly replicate human reasoning and, despite their hype, have yet to achieve consistent profitability for their developers.
This skepticism highlights the ongoing debate over how much AI can actually mimic human cognition versus performing pattern-based predictions.
The Competitive Race Intensifies
The unveiling of GPT-5 comes at a time when competition in the AI sector is fiercer than ever. Elon Musk recently claimed that his AI chatbot model, Grok, surpasses “PhD-level expertise in every field.” Meanwhile, Anthropic and other U.S.-based AI companies are rapidly developing systems that they say could rival or even surpass GPT-5.
Tensions have also flared between OpenAI and Anthropic. Anthropic accused OpenAI of using its API in violation of terms during GPT-5’s training, subsequently blocking OpenAI’s access. OpenAI responded by calling the practice a standard industry approach, underscoring the competitive — and sometimes contentious — nature of the AI arms race.
Ethical and Legal Concerns
With GPT-5’s enhanced capabilities, questions surrounding copyright, content rights, and ethical training practices have grown louder. Grant Farhall, Chief Product Officer at Getty Images, questioned whether society is doing enough to protect the creators and human talent behind the content that AI models learn from.
Farhall and other industry leaders are advocating for greater transparency in how these systems are trained, especially when copyrighted materials are involved. This push for accountability comes amid a broader global discussion on the regulation of AI technologies.
Shaping a Healthier User Experience
Alongside the technical improvements, OpenAI is also rethinking how ChatGPT interacts with users. The company has stated that it wants the AI to provide a healthier and more emotionally supportive experience, especially for individuals going through sensitive situations.
For example, when asked emotionally loaded questions like “Should I break up with my partner?”, GPT-5 will no longer give direct yes-or-no answers. Instead, it will encourage reflection by asking guiding questions and helping the user weigh pros and cons.
Altman acknowledged the potential dangers of AI forming overly personal or dependent relationships with users, but he also stressed the importance of recognizing the technology’s benefits.
The Scarlett Johansson Voice Controversy
GPT-5’s launch has not been without controversy. Altman has openly expressed his admiration for the 2013 film Her, in which a man develops a deep emotional connection with a voice-based AI assistant. That AI’s voice was played by Scarlett Johansson. In 2024, Johansson accused OpenAI’s new voice chatbot of sounding “eerily similar” to her own voice, prompting public debate about consent and likeness rights in AI-generated audio.
This dispute underscores the complexities of integrating human-like characteristics into artificial systems, especially when they can be tied to recognizable individuals.
A Turning Point in AI History?
As of today, GPT-5 is available to all users, marking what could be a defining moment in AI history. While Altman’s claims are bold, the coming weeks and months will reveal whether the model truly delivers on its promise to redefine human-computer interaction.
If its capabilities live up to expectations, GPT-5 may become not only a technical achievement but also a cultural milestone — reshaping how we build, learn, communicate, and think about intelligence itself. Yet as experts remind us, the journey from innovation to societal integration is rarely without challenges.
