Материал на языке оригинала: English.
Continuously hardening 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.
22 декабря 2025 г. · Источник: OpenAI news
Полный текст доступен у источника.