Openi Chat is introducing a new potential for GPT that is designed to help users handle complex, multi -faceted tasks. This feature, called Chat GPT Agent, is now available to Pro, Plus and Team users, and enables AI to complete assignments to finish from the beginning using Virtual Computer.
With agent mode activation, users can ask Chat GPT to perform such tasks like ::
- Review a calendar and summarize incoming meetings with relevant context
- Plan a prescription and purchase ingredients
- Analyze competitors and set a slide deck
Unlike the normal chat boot Interactions that include the text, the agent can navigate to the website, can be securely logged in with the user’s permission, operates the code, conducting research in the spreadsheets or slides, and providing output into edited forms.
Users maintain control throughout the process, request clear approval before submitting a form with Chat GPT or handling sensitive information.
What power does Chat GPT Agent
The new capacity prepares by combining on the previous tools of Openi:
- Operator – For web navigation and site interaction
- Deep research – In the source of information synthesis.
- Chat GPT’s primary model – Understanding the natural language and reasoning.
Simultaneously, these ingredients enable the AI to transfer between reasoning and process, and complete the tasks in a more independent and systematic manner, while indicating users whenever explanation or approval is required.
How to access the agent mode
Users can open any chat GPT conversation and select ‘Agent Form’ from Tolls Dropdown to enable the feature.
Once activated, the system can take a multi -step -by -step workflow that usually requires switching between apps, browser tabs or tools.
Pay attention to safety and user control
Open has emphasized the security systems made in this release. The Chat GPT agent is designed to avoid high -risk measures without user approval, such as sending email, purchasing or offering legal or financial advice.
It has been trained to identify and reject malicious or ambiguous instructions, and it informs consumers about any uncertainty or potentially sensitive measures.
To further reduce the risks, Open, Open, has implemented many safety measures, including always classifying, refusing to train and avoiding misuse, especially in biological or chemical risk areas.
These reservations reflect the largest preparation framework of the open, which directly prefers caution in the absence of direct evidence that the model can enable harmful activity.
The Open noted, “We do not have direct evidence that the model can help Noshika suffer severe biological or chemical damage,” but we are using caution. “
What is the next
According to the Open, this is an early step in enhancing the capabilities of AI. The company plans to add new features and improvements over time, which aims to make Chat GPT more useful for more versatile and wider sets of users.
Whether summarizing meetings, conducting research or developing presentations, the functionality of a new agent of Chat GPT indicates a change in hand -on task implementation. Fortunately, the monitoring and control remain in the hands of the user.
Process Tom’s leader on Google News Our latest news in our feeds, how, and get reviews. Make sure to click the follow button.