OpenAI has announced the release of GPT-5, the latest version of its language model that powers ChatGPT, expanding access to all 700 million users globally. This launch marks a significant step in OpenAI’s ongoing efforts to scale AI capabilities for both individual and enterprise-level users, emphasizing enhanced reasoning and broader infrastructure support.
At a briefing in San Francisco, OpenAI Chief Executive Sam Altman described GPT-5 as capable of providing responses that match the depth and quality of a PhD-level expert. Altman highlighted its improved performance across various fields including health, finance, and especially software development. One of the standout features of GPT-5 is its ability to generate complete and functional code from plain-language prompts, a functionality internally dubbed “vibe coding”.
The model builds on GPT-4’s strengths with refinements in science, mathematics, and advanced reasoning. While early users have acknowledged the improvements, many characterize the leap from GPT-4 to GPT-5 as more measured compared to previous generational jumps. Still, GPT-5 introduces new tools such as “test-time compute” which allows the model to use additional processing power in real time for more complex questions—this capability will now be accessible to users for the first time.
Altman emphasized that GPT-5 still cannot learn autonomously, a key threshold for achieving more human-like adaptability. Nonetheless, the model’s development signals OpenAI’s push to keep pace with rising expectations from users and enterprises as generative AI becomes increasingly central in global workflows.
The release of GPT-5 arrives as major tech firms including Alphabet, Meta, Amazon, and Microsoft—OpenAI’s primary backer—prepare to spend nearly US$400 billion on AI data centers this fiscal year. These infrastructure investments are aimed at supporting the exponential computational demands of models like GPT-5. Altman stated that expanding infrastructure globally is essential to ensure equitable access and to deliver on OpenAI’s mission to benefit all of humanity.
Despite strong consumer demand for tools like ChatGPT, industry experts caution that enterprise adoption remains slower. Economics writer Noah Smith pointed out that consumer usage alone will not justify the massive investments being poured into AI infrastructure, indicating a gap between technological advancement and commercial deployment at scale.
OpenAI is reportedly in early discussions to allow employees to sell their shares at a valuation of US$500 billion, significantly higher than its current US$300 billion estimate. At the same time, top AI researchers are being offered signing bonuses of up to US$100 million, underlining the intense competition in the AI talent market.
Nearly three years since ChatGPT’s debut triggered the generative AI wave, OpenAI now faces the challenge of pushing boundaries further amid growing complexities in model training and limited availability of high-quality training data.
