• The Marvis chatbot from Juniper, which debuted in 2018, assists administrators in resolving technical problems affecting their company’s corporate network.
  • A new software service called Mist Access Assurance is being introduced at the same time as the improved version of Marvis.

A recent update to Juniper Networks Inc.’s Marvis chatbot uses OpenAI LP’s ChatGPT model to simplify network troubleshooting.

A new Wi-Fi endpoint and several other software updates are being released at the same time as the ChatGPT integration. According to Juniper, the latter device has a special sensor for identifying network interference. All of the brand-new products the company has added to its lineup use artificial intelligence models to some degree.

Marvis Gets More Integrations

The Marvis chatbot from Juniper, which debuted in 2018, assists administrators in resolving technical problems affecting their company’s corporate network. Wi-Fi access points and the wide area network that connects a company’s office locations are just two examples of the infrastructure problems it can locate the root cause of. According to Juniper, 90% of support tickets relating to this infrastructure can be automatically resolved by the software.

The company claims that the recently-debuted ChatGPT integration will allow its chatbot to automate even more tasks. Marvis uses the model created by OpenAI to assist administrators in locating information in Juniper’s product documentation. For instance, an administrator can ask the chatbot how to set up a Wi-Fi access point or how to interpret the router’s warning signals.

Bob Friday, Juniper’s Chief AI Officer, recently said, “AI is the next step in automating tasks that typically require a human IT domain expert, improving how IT teams operate the network with AI-driven tools like Marvis and its conversational interface. With these latest LLM enhancements, Marvis will provide even more actionable knowledge and be an even more valuable member of the IT team.”

Marvis will be able to troubleshoot Zoom calls’ technical problems thanks to a second integration that is simultaneously rolling out. According to Juniper, the integration enables Zoom Video Communications Inc.’s cloud backend to provide diagnostics data to Juniper’s chatbot. To resolve technical problems, Marvis compares the data with network logs from a company’s internal network.

An administrator can quickly identify technical problems like excessively high processor utilization or connectivity problems in the user’s Wi-Fi network by using the integration to track down Zoom errors. In addition, Juniper claims Marvis can identify mistakes that keep happening over time. The chatbot’s detection of error patterns enables administrators to foresee future malfunctions and address the underlying technical problem.

Cloud-based Access Control 

A new software solution called Mist Access Assurance is being released at the same time as the improved version of Marvis. It’s a tool known as NAC or network access control. Administrators use these tools to control how and which devices can connect to the corporate network.

In the past, NAC products required businesses to install specialized hardware in their data centers. According to Juniper, that requirement is eliminated by Mist Access Assurance. The business claims administrators can specify which users can access which applications and under what circumstances by using a cloud-based management console.

Mist Access Assurance provides support for microsegmentation as part of its feature set. Administrators can therefore permit a device to log in to specific systems while barring connections to the remainder of the network. Microsegmentation reduces the risk posed by compromised user accounts and devices by limiting the number of systems that each employee may access.

Mist Access Assurance uses AI to streamline the process of adding new endpoints to the network. Thanks to its microservices architecture, the platform can support tens of thousands of endpoints and users, claims Juniper. A company’s cybersecurity tools can also be integrated with it using an application programming interface.

Greater Selection of Hardware

The AP24, a new Wi-Fi access point made for use in places like offices, was introduced by Juniper alongside the latest additions to its software lineup. Support for the Wi-Fi 6E standard is its key selling point.

By encoding data into radio signals, wireless routers transmit data. Many times, various frequencies are used to transmit those signals. The cause is the possibility of interference between two Wi-Fi connections operating on the same frequency, which lowers download speeds and could result in outages.

In order to lessen network interference, Wi-Fi 6E, the version of the Wi-Fi standard that Juniper’s new AP24 endpoint supports, was developed. The standard expands the range of frequencies that wireless endpoints are capable of using for data transmission. As a result, the AP24 can disperse Wi-Fi connections over a larger portion of the radio spectrum, preventing interference between them.

According to Juniper, the AP24 also lowers the possibility of connection errors in other ways. It employs artificial intelligence to proactively look for network inference brought on by Wi-Fi connections that use the same frequency. The AP24 has the ability to switch frequencies when this interference is found automatically.

Three radios are housed in a seven-square-inch rectangular chassis for the AP24. According to Juniper, the onboard AI uses one of those radios as a network interference sensor. Wi-Fi connections can receive up to 3.6 gigabits of throughput per second from each AP24.