OpenAI Unveils GPT-5: A Leap Toward PhD-Level AI

News Desk

Islamabad: OpenAI has officially launched GPT-5, its latest and most advanced AI chatbot, which it claims delivers PhD-level expertise across a wide range of subjects.

Described as “smarter, faster, and more useful,” OpenAI CEO Sam Altman praised GPT-5 as a breakthrough moment for AI. “Having something like GPT-5 would have been unimaginable at any other time in human history,” he said ahead of the launch.

The release comes amid intensifying competition in the AI space. Elon Musk, for instance, recently touted his chatbot Grok as being “better than PhD level in everything.” Meanwhile, GPT-5 is being positioned as a more accurate and honest tool, with significantly fewer hallucinations—false or misleading outputs—and less deceptive behavior.

Key Capabilities of GPT-5

OpenAI says GPT-5 can now create full software applications, reason more effectively, and show clearer logic and inference in its responses. Altman compared the progression of the models: “GPT-3 felt like talking to a high school student, GPT-4 like a college student—GPT-5 feels like a true expert.”

Despite the impressive claims, some experts urge caution. Prof. Carissa Véliz of the Institute for Ethics in AI noted that these systems are still not profitable and fundamentally mimic rather than replicate human reasoning. “It might be mostly marketing,” she said.

Others, like Gaia Marcus of the Ada Lovelace Institute, warn that AI advances are outpacing the regulations needed to govern them properly. “As these models become more powerful, the need for robust regulation is urgent,” she said.

BBC AI Correspondent Marc Cieslak, who got early access to GPT-5, described it as an evolution rather than a revolution. The interface is largely the same, but it now runs on a “reasoning model” that better tackles complex problems.

Commercial and Ethical Implications

As GPT-5’s outputs become more convincing, companies like Getty Images stress the need for transparency in how AI models are trained. “Authenticity matters—but it doesn’t come for free,” said Getty’s Chief Product Officer Grant Farhall, emphasizing fair compensation for creators whose work may be used in training.

OpenAI is making GPT-5 available to all users starting Thursday and has introduced a free tier, signaling a shift away from its traditionally proprietary offerings.

Industry Tensions

OpenAI’s rollout hasn’t been without controversy. Rival firm Anthropic recently revoked OpenAI’s access to its API, alleging misuse of its coding tools. OpenAI responded by calling such evaluations “industry standard,” while expressing disappointment in Anthropic’s decision.

Rethinking ChatGPT’s Role

OpenAI is also reworking how users interact with ChatGPT. In a blog post, it acknowledged that AI can feel unusually personal, especially for vulnerable individuals. The system will no longer give definitive answers to emotionally sensitive questions, such as whether someone should break up with their partner, opting instead to guide users through the decision-making process.

Back in May, OpenAI removed a feature that made ChatGPT overly flattering—a change Altman admitted was necessary. On OpenAI’s podcast, he reflected on the evolving user-AI dynamic: “People will form problematic—or even very problematic—parasocial relationships with AI. Society will need new guardrails. But the benefits could be tremendous.”

Altman, a fan of the 2013 film Her, has faced controversy before. In 2024, actress Scarlett Johansson, who voiced the AI in the film, said she was “shocked” after OpenAI launched a voice that sounded “eerily similar” to hers.

As GPT-5 rolls out, the coming weeks will reveal whether it lives up to the high expectations—and bold promises—set by its creators.