OpenAI is bringing the power of agentic AI to ChatGPT so that it can handle complex requests from users autonomously.
It leverages two of OpenAI’s existing capabilities: Operator, which can interact with websites, and deep research, which can synthesize information. According to OpenAI, these capabilities were best suited for different situations, with Operator struggling with complex analysis and deep research being unable to interact with websites to refine results or access content that required authentication.
“By integrating these complementary strengths in ChatGPT and introducing additional tools, we’ve unlocked entirely new capabilities within one model. It can now actively engage websites—clicking, filtering, and gathering more precise, efficient results. You can also naturally transition from a simple conversation to requesting actions directly within the same chat,” the company wrote in a blog post.
ChatGPT’s agent mode can handle requests such as “look at my calendar and brief me on upcoming client meetings based on recent news” or “plan and buy ingredients to make Japanese breakfast for four.”
After it is prompted, it will begin navigating websites, filtering results, running code, conducting analysis, and prompting a user to login when necessary so that it can access specific websites. It can also provide editable slideshows and spreadsheets summarizing its findings, OpenAI explained.
According to OpenAI, the agent will seamlessly switch between reasoning and action, and it requests permission from the user before taking action so that they can interrupt, take over, or stop tasks at any point in the process.
Since this is the first release of ChatGPT that allows it to take action on its own, there are added risks. To address these, OpenAI added new safeguards such as resistance against prompt injection; mitigations of model mistakes, including requiring explicit user confirmation, active supervision, and proactive risk mitigation (ie refusion high risk activity like bank transfers); and controls to limit data access, like privacy controls and not storing data, such as passwords, collected during browser takeover mode.
The new agent mode has begun rolling out to Pro, Plus, and Team users, and will roll out to Enterprise and Education users in the next few weeks. It can be activated through the tools dropdown menu at any time.
Pro users will get 400 messages/month, and other paid tiers will receive 40 messages/month. Additional usage can be purchased on a flexible credit-based system.
“While ChatGPT agent is already a powerful tool for handling complex tasks, today’s launch is just the beginning. We’ll continue to iteratively add significant improvements regularly, making it more capable and useful to more people over time,” OpenAI wrote.
