Continuously hardening ChatGPT Atlas against prompt injection

Published: 3 days ago (December 21, 2025 at 07:00 PM EST)

1 min read

Source: OpenAI Blog

OpenAI strengthens ChatGPT Atlas against prompt injection

OpenAI is strengthening ChatGPT Atlas against prompt injection attacks using automated red teaming trained with reinforcement learning. This proactive discover-and-patch loop helps identify novel exploits early and harden the browser agent’s defenses as AI becomes more agentic.

Back to Blog

Syntax hacking: Researchers discover sentence structure can bypass AI safety rules

New research offers clues about why some prompt injection attacks may succeed....

Why AI safety should be enforced structurally, not trained in

Most current AI safety work assumes an unsafe system and tries to train better behavior into it. - We add more data. - We add more constraints. - We add more fi...

Securing Gmail AI Agents against Prompt Injection with Model Armor

markdown !Google Workspace Developers profile imagehttps://media2.dev.to/dynamic/image/width=50,height=50,fit=cover,gravity=auto,format=auto/https%3A%2F%2Fdev-t...

Updating our Model Spec with teen protections

OpenAI is updating its Model Spec with new Under-18 Principles that define how ChatGPT should support teens with safe, age-appropriate guidance grounded in deve...

OpenAI strengthens ChatGPT Atlas against prompt injection

Related posts

Syntax hacking: Researchers discover sentence structure can bypass AI safety rules

Why AI safety should be enforced structurally, not trained in

Securing Gmail AI Agents against Prompt Injection with Model Armor

Updating our Model Spec with teen protections