OpenAI CTO Srinivas Narayanan, speaking at the Hindustan Times Leadership Summit 2025, stressed that building strong guardrails is central to how OpenAI designs and deploys its AI models. He said the company thinks about safety at every stage, rigorously evaluates potential risks before launch, and maintains a very high bar for what is allowed into production.