OpenAI launches Lockdown Mode to block prompt injection in ChatGPT

OpenAI introduced Lockdown Mode in ChatGPT to prevent prompt injection attacks by disabling live web browsing, image retrieval, deep research, and agent mode. Users can only access cached content; sensitive data exfiltration risks are reduced but not eliminated. The feature is available to self-serve ChatGPT Business and eligible personal accounts. OpenAI states prompt injections can still occur in cached content or uploaded files.