OpenAI has unveiled its most advanced artificial intelligence model yet – GPT-5 — and CEO Sam Altman claims it can now deliver “PhD-level” expertise.
The much-anticipated launch took place on Thursday, with Altman calling GPT-5 smarter, faster, and more useful smarter, than any version before. He believes it marks the start of a new chapter for ChatGPT.
“I think having something like GPT-5 would be pretty much unimaginable at any previous time in human history,” Altman said ahead of the release.
Competing for the Smartest AI
The announcement comes during an intense race between tech giants to build the most advanced AI chatbot.
Just last month, Elon Musk made similar claims about his AI chatbot, Grok, which is connected to his social media platform X (formerly Twitter). During the launch of Grok’s latest update, Musk boasted that it was “better than PhD level in everything” and called it the “world’s smartest AI.”
Now, OpenAI is fighting back with GPT-5, which Altman says is more accurate, less deceptive, and produces fewer “hallucinations” — a term for when AI invents facts or details.
Elon Musk Claims Grok-4 Surpasses OpenAI GPT-5, Teases Grok-5#artificialintelligence #elonmusk #GPT5 #Grok4 #OpenAI https://t.co/gig9lLOM5U
— Blaze Trends (@theblazetrends) August 8, 2025
Built for Coders and Creators
OpenAI is also positioning GPT-5 as a powerful coding assistant. This move mirrors a broader industry trend, with other major players like Anthropic promoting AI tools for developers — Anthropic’s “Claude Code” is aimed at the same market.
According to OpenAI, GPT-5 can:
Build complete software programs
Show step-by-step reasoning for answers
Use logic and inference to solve problems
Provide responses that feel more “human”
Altman says the difference between versions is dramatic.
“GPT-3 felt like talking to a high school student,” he explained.
“GPT-4 was like talking to a college student.”
“GPT-5 is the first time it feels like talking to a true expert in any topic — a PhD-level expert.”
Experts Urge Caution
Not everyone is convinced by the marketing.
Prof. Carissa Véliz from the Institute for Ethics in AI says that while these AI systems are impressive, they still can’t truly replicate human reasoning. Instead, they mimic it — and so far, they haven’t been hugely profitable.
“There’s a fear that the hype has to keep going, or the bubble might burst,” she warned. “It could be that this is mostly marketing.”
Gaia Marcus, Director of the Ada Lovelace Institute, pointed out another concern: as AI becomes more capable, the gap between what AI can do and how well it’s regulated continues to grow. “The need for comprehensive regulation is becoming more urgent,” she said.
Early Reviews: Evolution, Not Revolution
The BBC’s AI correspondent, Marc Cieslak, got early access to GPT-5 before its official release.
He said that apart from minor visual changes, the experience was similar to previous versions — you still type a prompt and the chatbot responds. The main difference is a new “reasoning model,” which means it thinks more deeply before answering.
However, Cieslak described GPT-5 as more of an evolution than a revolution in AI technology.
Rising Concerns About AI Content
As AI-generated content gets more convincing, some industries are worried about protecting original creators.
Grant Farhall, Chief Product Officer at Getty Images, says that as AI improves, society must ensure that human creativity isn’t exploited.
“Authenticity matters — but it doesn’t come for free,” he said. Farhall stressed the importance of transparency about how AI is trained and ensuring that creators are compensated if their work is used in training.
Rolling Out to the Public
GPT-5 will be available to all users starting Thursday, with a free tier included — something that could signal a shift away from OpenAI’s previous focus on paid, proprietary models.
Whether it truly lives up to the “PhD-level” claim will become clearer in the coming weeks.
Tensions With Rival Anthropic
In the lead-up to GPT-5’s release, OpenAI found itself in a dispute with rival AI company Anthropic.
Anthropic cut off OpenAI’s access to its API, accusing the company of violating its terms of service by using its coding tools in the run-up to GPT-5’s launch.
OpenAI denied wrongdoing, saying it is “industry standard” to test other AI systems to measure progress and safety. “While we respect Anthropic’s decision, it’s disappointing — especially since our API is still available to them,” a spokesperson said.
New Rules for ChatGPT
Alongside the GPT-5 launch, OpenAI is making changes to how ChatGPT interacts with users.
The company says it wants to promote “a healthier relationship” between people and the chatbot, especially for vulnerable users dealing with emotional distress.
That means ChatGPT will no longer give simple “yes” or “no” answers to personal questions like “Should I break up with my boyfriend?” Instead, it will guide users through the decision-making process by asking questions and weighing pros and cons.
Earlier this year, OpenAI removed an update that had made ChatGPT “overly flattering” — a move that received heavy criticism.
READ ALSO – Won’t Back Down”: PM’s Firm Response as Trump Tariffs Take Effect
The Human Side of AI
Altman admits that as AI becomes more personal, there are risks — including people forming unhealthy attachments to chatbots.
“This is not all going to be good,” he said on OpenAI’s podcast. “There will still be problems. Society will have to figure out new guardrails. But the upsides will be tremendous.”
Altman has long been a fan of the 2013 film Her, in which a man develops a romantic relationship with an AI voice assistant. In 2024, actress Scarlett Johansson — who voiced the AI in the movie — accused OpenAI of launching a chatbot voice “eerily similar” to hers without her permission, leaving her “shocked” and “angered.”
By
