HPE executive says AI could clear nearly all network trouble tickets within three years

HPE is extending AI-assisted network monitoring and automated fixes across its hybrid-cloud portfolio. Its networking chief says the longer-term aim is to resolve almost every network trouble ticket without human intervention, with hardware replacement the likely exception.

Watch Desk analysis

What happened

Rami Rahim, HPE’s president and general manager of networking, told The Register that he estimates 70 to 80 per cent of tickets already require no human intervention. He expects there to be “no issues that require humans” within two to three years, while allowing that hardware swaps may remain an exception.

The approach is to let operators enable automated actions feature by feature. Rahim offered rebooting a switch port as an example of a simple task to start with. He said HPE is extending technology that began at Juniper across its wider portfolio, and that network fabric management capabilities have already been integrated with OpsRamp, HPE’s hybrid-cloud observability software. Read The Register’s interview.

Why it matters

This is a concrete vision for how AI could change day-to-day network operations: diagnose what is behind a reported connection problem, then take selected actions to fix it. Rahim says the familiar complaint “the Wi-Fi sucks” can stem from faults elsewhere, including an application, wide-area network or cloud provider. Faster diagnosis and repair could matter as organisations add more connected devices and AI agents to their networks.

But the headline forecast is Rahim’s expectation, not a demonstrated timetable for every organisation. The useful detail is the proposed gradual switch-on of actions, rather than an overnight handover of the network to an agent.

Our read

There is a sensible middle ground between “let the machines handle it” and making an administrator approve every port reboot. Starting with narrow, reversible actions gives operators a way to test automation before trusting it with bigger fixes. The proof will be whether it resolves real faults reliably, and whether teams can see what it changed when something goes wrong.

What to watch

• Which automated actions HPE adds across its networking portfolio, and when.
• Whether HPE publishes results showing how often automated fixes resolve tickets without causing new problems.
• How operators can review, limit or reverse an automated action.

Discussion spark: Which network fixes would you let an AI agent handle without approval, and which should always require a person to sign off?

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.