OpenAI launches GPT-5, amid intensifying global AI competition

OpenAI has officially launched GPT-5, the latest iteration of its generative AI model, marking a major leap forward in artificial intelligence. The highly anticipated release, which debuted on August 7, promises to deliver significant advancements in coding, writing, health, and multimodal reasoning, setting the stage for even more transformative applications of AI in the real world. Available to 700 million ChatGPT users, GPT-5 is now the default model for signed-in users, replacing GPT-4 and previous versions. For the first time, it’s also accessible to free users, who will gradually gain access with lower usage limits. The model is designed to excel at both basic tasks and highly complex queries, positioning it as a powerful tool for businesses and professionals seeking more advanced capabilities.

GPT-5’s expanded capabilities

The improvements in GPT-5 are vast, making it a versatile solution for a wide range of applications. OpenAI CEO Sam Altman highlighted the model’s ability to serve as a “legitimate PhD-level expert” in nearly any field, particularly in coding, health-related advice, and writing. GPT-5 introduces features like vibe coding, where users can create entire software programs by simply providing written text prompts. This technology allows for rapid generation of complex software, making it a potential game-changer for industries looking to streamline development processes.

In healthcare, GPT-5 performs better on medical benchmarks, providing informed support without replacing professionals. In writing, GPT-5 handles more nuanced tasks, such as free verse poetry, and is far more proficient at drafting everyday content. Additionally, its ability to tackle advanced reasoning and solve complex problems in areas like maths and science sets it apart from its predecessors.

Big investments, bigger expectations

The release of GPT-5 comes at a critical moment for the AI industry, with the world’s largest tech companies doubling down on AI development. Microsoft, Meta, Amazon, Alphabet, and others have committed nearly $400 billion in AI investments this fiscal year, betting on the potential for artificial intelligence to redefine industries. OpenAI itself is in talks to increase its valuation to $500 billion, a substantial increase from its current $300 billion.

Despite the impressive advances of GPT-5, Altman cautioned that true artificial general intelligence (AGI)—AI that surpasses human intelligence in a broad range of activities—remains a distant goal. He emphasised that GPT-5 is not capable of continuous learning once deployed, a crucial element for AGI. However, Altman noted that the model marks a significant step toward achieving that long-term vision, with “orders of magnitude more gains” expected in the future.

Industry experts, including Meta’s Mark Zuckerberg, have expressed optimism that AI is on the brink of ushering in a new era, with some forecasting a near future dominated by superintelligent machines. However, some experts warned that while consumer spending on AI remains robust, enterprise adoption has been slower than expected, raising questions about the return on the substantial investments pouring into the sector.

A leap in AI, but not without criticism

While many have praised GPT-5 for its impressive performance, critics argue that the jump from GPT-4 to GPT-5 isn’t as groundbreaking as previous iterations. AI expert Simon Willison commented that GPT-5 exudes competence but lacks the dramatic improvements that defined earlier versions. Meanwhile, Elon Musk, who has launched his own competing AI model, Grok 4 Heavy, argued that his model outperforms GPT-5. Despite the competition, OpenAI remains confident that GPT-5 is a significant advancement, especially when it comes to safety. The new model has been trained to produce “safe completions”, minimising risks by offering answers that are not only accurate but also responsible and free of harmful content.

(With inputs from agencies)