OpenAI Rolls Out Locked Mode to All Personal ChatGPT Accounts

OpenAI is expanding its advanced security tool, Locked Mode, to a broader user base. First introduced in February for a small group of security-conscious users such as executives and security teams, the company now announces that the feature is officially available for all personal ChatGPT accounts—including Free, Go, Plus, and Pro tiers—as well as self-serve ChatGPT Business accounts. Users can enable it under the "Security" option in the settings.

Once Locked Mode is activated, ChatGPT will limit or disable several core features that connect to the internet or external services, including web browsing, plugins, code interpreter, and other components that may introduce external data or execute outside actions. This move is designed to offer users the most comprehensive protection against potential threats, particularly for individuals and businesses that need to use AI tools in sensitive scenarios. At the same time, OpenAI also announced that the Active Sessions feature is now available across all ChatGPT accounts and workspace types, making it easier to manage multiple login states.

As AI tools become more deeply integrated into both enterprise workflows and daily use, security and privacy boundaries have become a major focus for users. The full rollout of Locked Mode indicates that OpenAI is proactively reducing the risk of data leaks through technical measures, especially providing a more controllable interaction environment for industries with strict compliance requirements, such as finance and legal. In the future, similar granular security settings are likely to become standard across AI platforms, pushing the industry to shift from a "feature-first" to a "security-first" product design philosophy.