Just How Much Is OpenAI’s o3 Disrupting The Coding Industry? Tech Titans Weigh In

India’s long-standing reputation as a global tech powerhouse was built on the back of its software engineers. But with artificial intelligence rapidly gaining ground, that edge may be slipping. A viral Instagram Reel by entrepreneur Vaibhav Sisinty has sparked a heated conversation online after he revealed that OpenAI’s latest model, O3, now codes like a top-tier engineer, one in the top 200 globally.

“This isn’t a headline. It’s a warning,” Sisinty declared, as he showcased how 65% of the code at his own startup is now generated by AI. The implications? For millions of Indian developers, the rise of machine-coded software could mean a fundamental shift in job roles — and job security.

Tech Titans Weigh In On AI’s Coding Capabilities

Sisinty’s insights aren’t isolated. Industry veterans like Zoho’s Sridhar Vembu and OpenAI’s Sam Altman have echoed similar concerns and predictions.

“AI will handle 90% of coding,” Vembu asserted, emphasising that the repetitive nature of most programming tasks makes them ripe for automation.

Altman, meanwhile, took a more measured but no less significant stance. “Each software engineer will just do much, much more for a while,” he explained during a conversation with tech analyst Ben Thompson. “And then at some point, yeah, maybe we do need fewer software engineers.”

He further emphasised the changing nature of competitive advantage in the tech industry. “The obvious tactical thing is just get really good at using AI tools,” Altman said, comparing today’s AI-first landscape with the early 2000s era of traditional coding prowess.

Codex 2.0: The Future of ‘Agentic Coding’

The latest iteration of OpenAI’s Codex marks a giant leap forward in AI-assisted development. Now fully integrated into ChatGPT, Codex isn’t just suggesting code snippets — it’s writing, testing, debugging, and even submitting pull requests. All from natural language prompts.

The upgraded Codex, powered by OpenAI’s specialised O3 model, operates in a secure sandbox environment and is currently available to Pro, Team, and Enterprise subscribers. It’s a glimpse into the future of what Altman calls “agentic coding,” where AI agents independently handle increasingly complex tasks.

“I think in many companies, it’s probably past 50% now,” Altman noted, referring to the amount of code already being written by AI. “The big thing I think will come with agentic coding, which no one’s doing for real yet.”

A Tectonic Shift In the Making

The growing reliance on AI to write code isn’t just a theoretical future — it’s already happening. Anthropic CEO Dario Amodei recently predicted that AI will write all software code within the next year. Meta’s Mark Zuckerberg chimed in earlier this year, telling Joe Rogan that AI is set to generate a substantial chunk of code used in their applications.

What was once a niche novelty has become the new normal. As Sisinty’s warning gains traction and tech leaders align in their forecasts, Indian developers — and the global tech workforce — face an urgent need to adapt. The new rule of thumb? Learn to work with AI, or risk being left behind.