OpenAI releases Lockdown Mode and Elevated Risk labels in ChatGPT

AI Impact Summary

OpenAI is releasing Lockdown Mode and Elevated Risk labels in ChatGPT to bolster defenses against prompt injection and data exfiltration attacks. These features provide organizations with greater control over AI interactions and help mitigate potential security risks. While not immediately critical, this capability represents a proactive step towards responsible AI adoption and aligns with growing concerns about AI safety.