GPT-5 Launch: Sam Altman Hails India’s AI Innovation, Eyes It as Potential Biggest Market

OpenAI on Thursday released GPT-5, its first “unified” AI model that merges the deep reasoning of its o‑series with the rapid‑response capabilities of its GPT line.

Designed as the default engine in the next generation of ChatGPT, GPT-5 can not only answer questions but also execute complex multi‑step tasks, ranging from generating full software applications to managing calendar events and drafting research reports.

Beyond raw performance, GPT-5 introduces an on‑the‑fly routing system that dynamically chooses between fast replies and extended “thinking” routines. This removes the need for users to tweak settings, making advanced reasoning accessible to all free ChatGPT users for the first time.

“This is a significant step toward AI systems that act like agents rather than chatbots,” said CEO Sam Altman, calling GPT-5 “the best model in the world” and a milestone on the path to artificial general intelligence.

Benchmark Breakthroughs

In coding tests such as SWE‑bench Verified, GPT-5 scored 74.9 percent, edging out Anthropic’s Claude Opus 4.1 and far surpassing Google DeepMind’s Gemini 2.5 Pro. In scientific reasoning on the GPQA Diamond exam, GPT-5 Pro achieved an 89.4 percent first‑try success rate, outpacing its closest rivals.

The new model also slashes hallucinations, incorrect fabrications, down to 4.8 percent, compared with over 20 percent in prior iterations. Health‑focused benchmarks show hallucination rates as low as 1.6 percent, underscoring improvements in accuracy and safety.

OpenAI’s safety research lead, Alex Beutel, highlighted GPT-5’s reduced propensity for deceptive behaviour and its finer discrimination between benign and malicious queries. These enhancements aim to foster greater user trust by making the model more transparent and reliable in refusing unsafe requests while accommodating legitimate ones.

Free ChatGPT accounts will now default to GPT-5, while Plus subscribers receive higher usage limits. Pro tier users can tap an even more capable GPT-5 Pro variant. Organisations on Team, Edu and Enterprise plans will adopt GPT-5 next week.

For developers, OpenAI is offering three sizes, gpt-5, gpt-5-mini and gpt-5-nano, priced at $1.25 per million input tokens and $10 per million output tokens, with controls for response length and reasoning depth.

GPT-5’s launch follows OpenAI’s open-weight gpt-oss release, reflecting a dual strategy of proprietary frontier models and accessible AI tooling. As stakeholders from Silicon Valley to Washington monitor the impact of each generational leap, GPT-5’s real‑world adoption will test whether it truly raises the bar for AI agents and reshapes how businesses and consumers interact with intelligent systems.

Altman on India

Speaking at the GPT-5 launch briefing on Thursday, Altman said India is now the company’s second‑largest market after the United States and “may well become our largest market,” citing “incredibly fast‑growing” adoption of AI across the country. Altman also announced plans to visit India in September to deepen partnerships and tailor AI offerings to local needs.

“We’re especially focused on bringing products to India, working with local partners to make AI work great for India and make it more affordable for people across the country,” Altman told Economic Times reporters. OpenAI’s vice‑president of ChatGPT, Nick Turley, added that GPT-5 significantly improves multilingual understanding across more than a dozen Indian languages, an enhancement aimed squarely at broadening access for non‑English speakers.

Addressing concerns about AI displacing software engineers, Altman maintained there is “no evidence” of job losses attributable to generative AI’s coding prowess. He argued that “the world wants way more software,” predicting that increased AI‑driven productivity will unlock new demand, fuel economic growth and create related job opportunities rather than eliminate them.

Regulatory Engagement

On the subject of national AI regulations, Altman said OpenAI respects diverse legal frameworks and welcomes varied approaches. “All world leaders want to make sure that AI thrives in their country, and that the economic growth and societal benefits of that also come,” he noted, pointing to Australia’s regulatory experiments as one example among many.

With GPT-5 now available free to all ChatGPT users and delivering PhD‑level insights, Altman reiterated OpenAI’s mission to democratise advanced AI. His September visit to India is expected to include discussions with government officials, industry partners and academic institutions to build the infrastructure and collaborations needed for truly “local” AI deployment in the world’s fastest‑growing digital market.