§ feed · storyline
Understanding prompt injections: a frontier security challenge
OpenAI publishes guidance on prompt injection attacks, outlining how they work and the research, training, and safeguards the company is developing to address them.
Prompt injections are a frontier security challenge for AI systems. Learn how these attacks work and how OpenAI is advancing research, training models, and building safeguards for users.
§ sources1 publication · timeline below