- Operai has introduced a new tool called chatgpt agent to handle online tasks autonomously
- The chatgpt agent can navigate by websites, administer calendars, execute code and create slide presentations from a single natural language message
- The agent can execute tasks in applications and services, although it will seek user approval before some actions
Operai is implementing a new characteristic for Chatgpt designed to make the AI Chatbot a semi -autoomo digital assistant. The CEO Sam Altman and several other OpenAI luminaires demonstrated the new Chatgpt agent tool in a live broadcast, showing how it works similar to the OpenI operator tool, combined with the deep research function.
The chatgpt agent can complete the real world’s tasks on your computer from a warning and create long and complex reports about what you find online. It can also be linked to other programs and accounts, which means that it could administer its calendar, prepare a PowerPoint presentation and hunt through websites as a human would.
The deployment is currently only for subscribers of Chatgpt Pro, Plus or Team, so free users are not lucky at this time. You can activate the agent by choosing the “agent mode” in the ChatGPT tool menu.
This is more than a new feature to add to the chatgpt list. The chatgpt agent completely reorient the AI chatbot. In general, I would ask Chatgpt a question, I would receive an answer, open another application and perform a response -based action, etc. Now, you can say: “Help me plan a trip to Tokyo, find three hotels with less than $ 150 per night and put them on a table with pros and cons”, and the AI will take everything at once.
And although the chatgpt agent incorporates some of the elements of operators and Deep Research, it has the unique ability to integrate everything into a single task without problems. Let’s say you want to make reservations for dinner: not only will you extract options for places to eat, but you can also verify your calendar to see when you are free. You can ask you to compare products and get a personalized buyer guide with reviews, prices and availability.
You also have safeguards against going too far and will request approval before sending a reservation application or connect to your Gmail account. For the average person who plays with Chatgpt, this is the type of jump forward that makes AI feel less as an intelligent novelty and more as a useful tool that is worth paying, which, of course, is something that operates expects many people to feel about it.
AI AGENT
Attend
The live broadcast announcement included several examples of Altman and his chatgpt agent use team for practical tasks. He was asked to “plan and understand ingredients to prepare a Japanese breakfast for four” and “analyze three competitors and create a slide cover.” But that barely scratches the surface of what people could do with an AI agent.
I immediately imagined people who asked for help for everything, from long -term travel itineraries, specific gift purchase needs and even ideas to make emails less passive.
Keeping it embedded in Chatgpt is also an intelligent movement. Without installing anything new, it is much more likely that the average user really proves the agent. Of course, giving any AI that great agency raises obvious security questions, but Openai made a big problem about how much effort they have put in safety railings.
If your application even suggests remotely trying to design something shaded, the system marks it, redirizes it and, in some cases, turns it off completely. There are even tasks detection layers that sound innocent but can lead to harmful results. In other words, you cannot get to design and build a drug poster for you, regardless of how politely ask.
However, mainly, the chatgpt agent returns to work the chatbot AI in a more active partner to help with the small details of his life that need a hand. It offers action together with answers, at least if you are willing to pay the privilege.
