5OpenAI Release Notes·
OpenAI rolls out Lockdown Mode to all ChatGPT logged-in users as opt-in security setting
OpenAI has made Lockdown Mode available to all logged-in ChatGPT users across account types and workspaces. The optional setting restricts network-enabled capabilities—including web browsing, deep research, agent mode, and file downloads—to reduce the risk of data exfiltration via prompt injection attacks. Personal users can enable it via Settings > Security; workspace admins can configure it through role-based access controls.
Related guides (3)
Related events (8)
Introducing Lockdown Mode and Elevated Risk Labels in ChatGPT
OpenAI is introducing two new enterprise security features in ChatGPT: Lockdown Mode, designed to help organizations defend against prompt injection attacks, and Elevated Risk labels to flag AI-driven data exfiltration attempts. These features target organizational deployments where adversarial manipulation of AI systems poses operational security risks. The announcement signals growing attention to agentic and enterprise threat models within ChatGPT's product surface.
OpenAI updates deep research in ChatGPT with source control and mid-run steering
OpenAI is rolling out improvements to the deep research feature in ChatGPT, including the ability to restrict research to specific websites and connected apps as trusted sources. Users can now create and edit a research plan before execution and adjust direction mid-run. UI changes include a redesigned sidebar and fullscreen report view. The update is rolling out to Plus and Pro users first, with Free and Go users to follow.
OpenAI updates deep research in ChatGPT with source control and mid-run steering
OpenAI has shipped improvements to the deep research feature in ChatGPT, adding the ability to restrict research to specific websites and connected apps as trusted sources. Users can now create and edit a research plan before execution and adjust direction mid-run. A redesigned UI with a fullscreen report view and sidebar entry point accompanies the update, with no changes to RBAC for Enterprise and Edu tiers.
OpenAI launches ChatGPT for Academic Researchers program with free Pro-level workspace access
OpenAI is offering faculty and postdoctoral researchers 12 months of complimentary ChatGPT Pro-level workspace access through a new program called ChatGPT for Academic Researchers. Each workspace supports up to five verified members from the same institution and includes business data protections. Applicants verify affiliation via SheerID and must submit a qualifying paper and research description. The program covers ChatGPT access only, not API credits.
OpenAI makes Workspace Agents generally available in ChatGPT Business, Enterprise, and Edu
OpenAI has moved Workspace Agents to general availability for ChatGPT Business, Enterprise, and Edu tiers, enabling teams to deploy agents that own entire workflows, follow team processes, and be shared across organizations. New admin controls allow agent builders to set per-app action safeguards, and admins gain activity and usage visibility in the admin console. The free period has been extended to July 6, 2026, after which credit-based pricing begins.
Building more helpful ChatGPT experiences for everyone
OpenAI is announcing a set of ChatGPT safety and helpfulness improvements including new parental controls for teen users, routing of sensitive conversations to reasoning models, and partnerships with external experts. The update reflects OpenAI's ongoing effort to balance accessibility with safeguards across different user demographics. Routing sensitive queries to reasoning models is a notable architectural/policy decision that may affect response quality and safety outcomes.
OpenAI begins testing ads in ChatGPT for Free and Go tier users in the U.S.
OpenAI is rolling out an ad test in ChatGPT for logged-in adult users on Free and Go tiers in the United States, with paid plans (Plus, Pro, Business, Enterprise, Education) remaining ad-free. Ads are clearly labeled, visually separated from answers, and personalized using in-session data that is not shared with advertisers. The system excludes ads from sensitive topics (health, mental health, politics) and users predicted to be under 18, with user controls for dismissal and data deletion.
OpenAI launches ChatGPT Work, an agentic product for long-horizon tasks with scheduling and multi-app integration
OpenAI has released ChatGPT Work, an agent-oriented product tier within ChatGPT designed for extended, multi-step tasks including research, document creation, and cross-app workflows. It supports human-in-the-loop oversight, scheduled and triggered task execution, and integrates with connected apps and files. The rollout targets paid plans (Pro, Enterprise, Edu first; Plus and Business following), with a two-week admin-controlled preview for enterprise workspaces. Alongside this, OpenAI is replacing its App Directory with a Plugin Directory that bundles skills, apps, and templates for specific workflows.