ChatGPT agent System Card
OpenAI just shipped agentic ChatGPT with browser automation and code tools. Here's what the safety guardrails actually look like.
Why it matters
OpenAI is moving from chat interface to autonomous agents—deploying browser automation, code execution, and research tools under a formal safety framework. This signals the shift from user-directed prompting to independent task execution at scale.
The key facts
5 to knowChatGPT agent capabilities: browser automation, code execution, research tools
Safety framework: OpenAI Preparedness Framework applied to agentic model
Product status: Shipped/live (System Card published indicates production readiness)
Integration model: Research + browser + code tools unified in single agent
Governance: Formal safeguards documented in System Card (transparency signal)
Go to the source
OpenAI Blogopenai.com
Publisher excerpt: ChatGPT agent System Card: OpenAI’s agentic model unites research, browser automation, and code tools with safeguards under the Preparedness Framework.