AI-system security
Adversarial inputs and vulnerable integrations can undermine an AI application’s intended behavior.
READING ROOM ↗A historical model record tied to its original research announcement.
OpenAI announced GPT-4 on 14 March 2023.
March 2023 release documentation; this is not a live model-availability listing.
OpenAI described a multimodal model with text and image-input capabilities.
The announcement discusses safety work while acknowledging remaining limitations.
Developer-documented concerns include hallucinations, social biases and adversarial prompts.
Links provide context. Read the stated relationship; inclusion does not imply misconduct, endorsement or a legal obligation.
OpenAI · Primary source · Developer documentation
Published 2023-03-14 · Accessed 2026-10-07
Review dates track editorial checks. Updates may reflect corrections or added context; they are not new incidents. Private draft revisions are omitted.
Initial sourced historical collection. Review date describes this record, not the date of the event.