|
CommandCTRL has been designed to allow integration with OpenAI’s API in order to leverage the large language models (LLMs) for more information on what is running on a machine in your environment. When looking at a list of processes on a machine, you can query OpenAI for the manufacturer and a description based on the process name. CommandCTRL currently utilizes OpenAI’s gpt-3.5-turbo-0125 or gpt-4 or gpt-4-turbo-preview , which is the LLM that ChatGPT was originally based upon.
You can integrate CommandCTRL with OpenAI gpt-3.5-turbo-0125 or gpt-4 or gpt-4-turbo-preview through its API (Application Programming Interface) Key and Organizational ID. This guides you through how to use OpenAI's API in CommandCTRL to provide you with information.
Note: Before you can use OpenAI's API, you will need to have an OpenAI account and CommandCTRL with OpenAI gpt-3.5-turbo-0125 or gpt-4 or gpt-4-turbo-preview. For more information, see Integrating CommandCTRL with OpenAI.
OpenAI charges a subscription and a fee for data (see OpenAI.com for their fee structure). Our integration was designed to help save you money by caching their information for 90 days. When anyone in your tenant clicks on the OpenAI 
icon, the information they provide is saved to cache and is available whenever anyone else in your tenant requests the same information within 90 days. After 90 days, it is assumed that the information needs to be refreshed since information is always changing.
When you are viewing the list of processes, this API integration allows you to get information about a specific process from OpenAI. This can be helpful when you are troubleshooting a user's machine and you want to know more about a process that might be causing their machine to have an issue.
There are multiple ways to get to the Processes screen. This procedure assumes you noticed an issue displaying in one or more of the widgets on the Dashboard for a machine and want to navigate to the Processes screen to investigate. To view all the processes and get information from OpenAI about a specific process, complete the following steps:
- In the side navigation panel, go to Dashboard.
- Locate the issue in any of the "Top 5" consumers widgets (e.g., CPU, RAM, Disk, and Net). For example, the Memory widget is triggering an alert so you want to view the processes in the Top 5 RAM Consumers widget.
- Click the Go To Processes
icon on any of the "Top 5" consumers widgets.
The Processes screen displays. The data being displayed is only for the machine you selected and the users on that machine. You are able to see each of the machine and user names under the User column. - On the Processes screen, scroll through the list or use the search field to locate the process in which you want information about that process.
-
Note: You can also click on any of the column headings (i.e., PID, CPU, Memory, Disk, Network, and User) to sort the columns in ascending or descending order. This can be useful when you are trying to find the process that is using the most resources.
- Click the (OpenAI) icon to the right side of the process.
A dialog box opens that contains the Process (e.g., smss.exe), Manufacturer of that process (e.g., Microsoft), and a description of the process shown in the Website field. - Click anywhere outside of the dialog box for it to close.
OpenAI Accuracy Statement
If you type "statement regarding accuracy of openai information" into ChatGPT's search field, you will receive the following statement:
"As an AI language model developed by OpenAI, my responses are generated based on patterns and associations learned from a vast amount of text data. While I strive to provide accurate and helpful information, my responses may not always be completely accurate or up-to-date, as the world is constantly changing and evolving. It is always a good idea to verify any information I provide with additional sources and consult with experts in the relevant field. Additionally, it is important to remember that OpenAI is a research organization, and the information provided by OpenAI, including by me as a language model, should not be used as a substitute for professional advice or guidance."
