This is a planned architecture change. It is not a feature you can turn on today. Here's what Microsoft has said, what it hasn't, and what it could mean for anyone who has watched a laptop crash because of a wireless driver.
What Microsoft actually said
Windows Latest reported the plan, citing documents it had seen. The outlet says Kucukgoz confirmed it in a live Ignite session, where he said Wi-Fi was "currently in the works" and "will be released in 2027."
Microsoft's Ignite Tech Hub lists a session called "Reliable by design: Driver Quality Initiative in Action" (THR892), presented by Kevin Sullivan and Mehmet Kucukgoz. It was an advanced-level (300) Windows theater session about driver quality, reducing kernel risk and Windows reliability. The listing confirms the session and its speakers. It doesn't include a transcript, so the exact 2027 quote comes from Windows Latest's account.
Microsoft's own blog supports the general direction but gives no date. When it introduced the Driver Quality Initiative (DQI) in May, Microsoft wrote that user-mode driver investments include performance updates to PCIe devices with DMA support as well as Wi-fi stack (coming soon). In short, "coming soon" was the May wording, and 2027 is the target Kucukgoz gave at Ignite.
In brief:
- Officially confirmed by Microsoft: user-mode support for the Wi-Fi stack is a DQI investment.
- 2027 target: from Kucukgoz's Ignite remarks, as reported by Windows Latest.
- Not yet published: supported adapters, which Windows release gets it, how it rolls out, and compatibility requirements.
DQI: the bigger program behind this
The Wi-Fi change is one part of a larger effort that started at WinHEC 2026 in Taipei. That was Microsoft's first Windows Hardware Engineering Conference since 2018. Microsoft described DQI as an ecosystem-wide push with OEM, silicon, independent hardware vendor (IHV) and ODM partners. It noted that thousands of partners maintain tens of thousands of active driver families across the Windows install base.
DQI has four pillars:
| Pillar | What Microsoft says it covers |
|---|---|
| Architecture | Hardening kernel mode drivers and enabling the third-party kernel mode driver transition to either user mode driver or Microsoft authored class drivers |
| Trust | Stronger verification, expanded automated analysis, and updated Windows Hardware Compatibility Program requirements |
| Lifecycle | Better Windows Update catalog hygiene, including deprecating outdated or low-quality drivers, advancing SBOM alignment and enabling faster issue analysis through driver symbols |
| Quality Measures | Measuring more than crashes, including stability, functionality, performance, and power and thermal impact |
At Ignite, Kucukgoz summed up the pillars this way: architecture modernizes the driver model, trust raises the bar for drivers and for who can submit them, lifecycle keeps drivers current and serviceable for their whole life, and measurement tracks quality the way customers experience it.
The architecture pillar isn't only about user mode. Microsoft is also expanding its own class drivers, including Soundwire Device Class for Audio (SDCA), the I3C class driver, and NCM USB ethernet class driver improvements.
DQI is already showing up in driver releases
There's early evidence that hardware makers are acting on this. Intel released Wi-Fi and Bluetooth driver version 24.50.0 on June 30, 2026. The Wi-Fi release notes say the drivers include enhancements aligned with Microsoft's Windows ecosystem quality initiative. TechRadar later noted that Intel appears to be ahead of the rest of the pack in implementing this.
Intel's changelog gives few technical details. Nothing suggests version 24.50.0 is a user-mode Wi-Fi driver. It's best read as a sign that DQI has moved from slides into shipping code, not as a preview of the 2027 change.
Why Wi-Fi drivers have lived in the kernel
Kernel-mode code runs with privileged access to core parts of the operating system. When a kernel driver breaks badly, the result is a stop error that takes the whole system down. That's why Microsoft has often put safeguard holds on feature updates for PCs with particular Wi-Fi or audio drivers: it didn't want a compatibility bug to crash machines in the middle of a rollout.
Running Wi-Fi drivers in the kernel wasn't laziness. Windows Latest lists several engineering reasons:
- Wireless drivers need to be fast to keep latency down.
- They need low-level access to hardware.
- They handle a constant flow of network traffic.
- They manage connections and process every incoming and outgoing packet.
Moving work out of the kernel also adds the cost of switching between kernel mode and user mode. Microsoft is upfront about this in its general UMDF documentation. Its FAQ says latency and CPU usage both increase somewhat, but that bus capacity is the main limiting factor for the device types UMDF supports. That's a statement about the framework in general. It isn't a measurement of the planned Wi-Fi design, and Microsoft hasn't published Wi-Fi benchmarks.
How user-mode drivers protect the rest of the system
The User-Mode Driver Framework (UMDF) isn't new. Windows Latest notes that the user-mode driver idea dates back to 2004. Microsoft's FAQ lists every Windows version from XP through Windows 11 as able to run UMDF drivers, and says UMDF 2.33 ships in Windows 11 version 21H2 and later. What's new is applying it to Wi-Fi.
Microsoft's documentation describes how the isolation works:
- Driver host process. Vendor UMDF drivers load into a host process (Wudfhost), which provides their execution environment.
- Driver manager. A single Windows service per system launches and tracks every driver host process.
- Reflector. A kernel-mode component passes communication between applications and the user-mode driver stack, including Plug and Play and power requests.
According to the FAQ, the host process runs with LocalService account credentials. The design keeps vendor code out of the kernel. Kucukgoz put it simply at Ignite: when a user-mode driver faults, the kernel and the rest of the device keep running.
What would that mean in practice? Today, a Wi-Fi driver that breaks after a cumulative update can crash the whole PC. With a working user-mode design, the likelier outcome is that the driver host fails while Windows stays up. You'd lose Wi-Fi for a while rather than lose your unsaved work.
The limits matter, though. Microsoft hasn't promised that user mode will end all Wi-Fi crashes. Hardware faults, firmware bugs and any components that stay in the kernel can still cause trouble. Microsoft's FAQ also says some drivers can be split between user mode and kernel mode when they need kernel-only resources.
DMA is the key to performance
According to Kucukgoz's Ignite remarks, sensors, fingerprint biometrics, USB, Ethernet and HID are among the device categories where UMDF already works, and Microsoft is extending support to more of them. DMA (direct memory access) support in UMDF is what makes that possible. DMA lets devices move data to and from system memory without the CPU handling every transfer. Vendors have long pointed to performance like this as their reason for staying in the kernel.
Kucukgoz said DMA support should enable much broader hardware support "without giving up performance." That's Microsoft's stated goal, not a published result. Expect Wi-Fi chipmakers to want their own measurements before they commit.
The drivers that stay in the kernel
Microsoft admits that not every driver can move to user mode, mainly for performance reasons. For drivers that stay in the kernel, Kucukgoz described several additional protections:
- Security-focused compiler options to reduce exploitable code.
- Driver isolation, so kernel-mode drivers don't have free rein over the whole system.
- User-mode accessor APIs, safer dedicated functions for kernel-mode reads and writes of user-mode memory.
- DMA remapping, which keeps device memory access within set bounds to help block DMA attacks.
He described the end goal as "less third-party kernel mode, fewer crashes and attack paths, and safer servicing." These measures lower the risk from kernel drivers. They don't eliminate it.
A healthy dose of skepticism
Is this the end of driver-related crashes? Probably not, and Microsoft hasn't claimed it is. There are a few reasons to be cautious:
- User mode has been available for about two decades, and most driver vendors haven't switched. DQI's trust and lifecycle pillars, which include stricter certification and removing low-quality drivers from Windows Update, give Microsoft more leverage this time. How hard it will push is still unclear.
- Older hardware is uncertain. Nothing published says existing Wi-Fi adapters will automatically use a new driver model in 2027. Support will likely depend on vendors shipping new drivers.
- Watch for misleading claims. Some websites have published detailed DQI "phases" and benchmark numbers that Microsoft's own posts don't contain. Until Microsoft documents the Wi-Fi design, treat specific performance or timeline figures beyond the 2027 target with suspicion.
Still, the direction is sensible. Microsoft is also encouraging security vendors to use user-mode Windows APIs instead of injecting code into the kernel, so the Wi-Fi plan is part of a wider effort to keep third-party code out of the kernel.
What Windows users and IT admins should do now
Nothing has changed yet, and there's no setting or driver to install in preparation. Until more details arrive:
- Keep normal driver hygiene. Get Wi-Fi drivers from Windows Update or your OEM/chipset vendor, and keep notes so you can roll back if an update causes problems.
- Respect safeguard holds. They exist because kernel-mode driver bugs can still crash a whole PC today.
- IT teams should track vendor roadmaps. When Microsoft publishes requirements for user-mode Wi-Fi, it will matter for hardware refresh plans and driver-validation testing.
For more context on WindowsForum, read this alongside our coverage of Intel's 24.50.0 Wi-Fi and Bluetooth drivers, Windows 11 safeguard holds, and Microsoft's broader push to get security software out of the kernel.
Bottom line
Microsoft wants Wi-Fi driver failures to cost you your connection, not your whole PC. Microsoft's May WinHEC post officially commits to user-mode Wi-Fi support. The 2027 date comes from Kucukgoz's Ignite remarks. Performance, hardware coverage and vendor adoption are the open questions.
References
- I'm genuinely excited to see Microsoft's big driver quality overhaul is underway — this could be the start of a much less buggy Windows 11 techradar.com
- Reliable by design: Driver Quality Initiative in Action | THR892 - Tech Hub tech.hub.ms
- User-Mode Driver Framework Frequently Asked Questions - Windows drivers | Microsoft Learn learn.microsoft.com