rss.livelink.threads-in-node

Miracast not working on Surface Hub 3 (Windows 11 IoT mode)

Andrea_Ragadini — Wed, 08 Apr 2026 09:16:29 GMT

Hi everyone,

we’re currently experiencing issues with Miracast on our Surface Hub 3 devices when running in Windows 11 (IoT) mode.

As far as we recall, Miracast used to work correctly in Windows 11 mode without any issues. We are also aware of the planned rollout for this functionality in MTR mode around May 2026. However, since this was already working for us in W11 mode, we’re wondering if something has recently changed that could have impacted this behavior.

So far, we’ve tried all the usual troubleshooting steps, including:

- Device reboots

- Policy/configuration checks and changes

- Full device reinstallation

Unfortunately, none of these actions resolved the issue.

Has anyone experienced something similar or is aware of recent changes that might explain this?

Thanks in advance,

Andrea

Microsoft Protection Plans now deliver more coverage

rawilson — Wed, 01 Apr 2026 18:33:46 GMT

Modern organizations rely on their Surface devices every day, across offices, job sites, classrooms, and hybrid work environments. Microsoft Protection Plans help reduce unexpected repair-related costs in accordance with the plan terms.

While every Microsoft Surface device comes with a minimum of 1-year Microsoft's Limited Hardware Warranty¹, you can protect your investment further with a Microsoft Protection Plan. Microsoft has recently updated its Protection Plans to provide broader coverage and reduce uncertainties in claim boundaries for customers. These enhancements are designed to better reflect how devices are used over time and to reduce uncertainty around repairs and long‑term ownership.

2026 Protection Plan Enhancements

Unlimited Mechanical Breakdown Coverage

Microsoft Protection Plans now include unlimited mechanical breakdown coverage²for the full duration of the plan. Previously, mechanical breakdowns and accidental damage claims were combined under a two‑claim limit. With this change, hardware failures caused by defects in materials or workmanship are no longer subject to a fixed claim count limit. Devices may be repaired or replaced as needed for mechanical issues throughout the coverage term.³

Accidental Damage Claims Are Now Separate

The existing two‑claim limit now applies only to Accidental Damagefrom Handling (ADH)⁴, such as drops, spills, or cracked screens.

By separating accidental damage from mechanical breakdown coverage, customers can more clearly understand how claims are applied and avoid using accidental damage claims for issues related to normal hardware failure.

New Battery Degradation Coverage

Microsoft Protection Plans⁵ now include battery degradation coverage.

Customers can file one dedicated battery claim⁶ if a device’s battery capacity drops below 70% of its original capacity during the plan term. This marks the first time Protection Plans address battery wear over time, not just battery failures caused by manufacturing defects.

Battery eligibility can be validated using tools such as the Surface Diagnostic Toolkit. Battery degradation claims are separate from accidental damage claims, so ADH limits remain unaffected.

Together, these updates help organizations extend device life, reduce unexpected repair costs, and plan more confidently for device ownership.

Customers can add a Microsoft Protection Plan when purchasing a new device, or within the eligible post-purchase window.⁷To purchase, contact your authorized partner or reseller, or engage the Microsoft Store virtual assistant.

[Consumer customers can purchase through the Surface app or your Microsoft account.]

These updates apply only to Microsoft Complete and Extended Hardware Service plans purchased on or after the applicable effective date (April 01, 2026) and are not retroactively modified. Coverage, whether for new or existing customers, is determined solely by the relevant Terms & Conditions for the specific plan purchased.

To learn more about Microsoft Protection Plans, visit Microsoft Surface Warranty & Protection Plans.

To see Warranty and Protection Plan Terms & Conditions, visit Support.microsoft.com.

Disclaimers

Without prejudice to any legal (statutory) rights to which you may be entitled under your local law, Microsoft Limited Hardware Warranty covers your device for one year from the date of original purchase from Microsoft or an authorized reseller. Restrictions apply. Please refer to Microsoft Limited Hardware Warranty & Agreement.
Additional extended coverage for mechanical breakdown and accidental damage from handling is available through the purchase of Microsoft protection plans. If the plan provides additional Mechanical Breakdown coverage, that coverage begins upon expiration of the manufacturer’s original warranty and continues for the remainder of the term shown on the Holder’s Proof of Purchase. Accidental damage from handling begins immediately upon purchase. Restrictions apply, for all Protection Plans, please reference the Terms and Conditions for the limit of liability and the applicable exclusions of the Protection Plan.
Subject to plan terms, conditions, and limit of liability.
Accidental damage from handling is included in the following plans: Commercial: Complete for Business, Complete for Business Plus, Accidental Damage Protection (EU), Accidental Damage Protection Plus (EU), and Consumer: Microsoft Complete. Extended Hardware Service (EHS) plans provide coverage for mechanical breakdown only and do not include Accidental Damage from Handling or battery degradation coverage.
Battery degradation is included in the following plans: Commercial: Complete for Business, Complete for Business Plus, Accidental Damage Protection (EU), Accidental Damage Protection Plus (EU), and Consumer: Microsoft Complete. Extended Hardware Service (EHS) plans provide coverage for mechanical breakdown only and do not include Accidental Damage from Handling or battery degradation coverage.
Customers with eligible plans are limited to one (1) battery degradation/replacement claim as set out in plan Terms and Conditions.
Availability and timing vary by market and plan type.

Dual Boot on Surface Hub 3

Stilldrey-MVP — Mon, 30 Mar 2026 22:41:08 GMT

Has anyone else shrunk the Surface Hub partition to support dual boot between MTR and Windows 11? I know it’s a niche case, but since the Surface Hub 3 has plenty of resources (memory & disk) to do more than just a meeting room, it might be interesting to explore!

I created a video demonstrating setting it up in this way for testing.

https://www.linkedin.com/posts/stilldrey_how-to-microsoft-surface-hub-3-dual-boot-ugcPost-7427475939858554880-l6pq?utm_source=share&utm_medium=member_ios&rcm=ACoAAADpC5gBtRjG2ffA5dQj2Vy6zWx7DvhK55A

#SurfaceMVP

LinkedIn video - dual boot Surface Hub 3

Surface Thunderbolt 4 Dock monitor(s) support

srazalir — Sun, 22 Mar 2026 15:03:27 GMT

Hi Everyone!

I am planning a setup with my Surface Thunderbolt 4 Dock and three monitors:

1x HP M24 23.8 inch webcam monitor (connected using USB-C)
2x HP 524sh 23.8 inch monitors (connected through DP ports DP-> HDMI cables)

I want to know if it's possible to run all three monitors simultaneously as I have shown above with my Microsoft Surface Pro 11.

Any tips, experiences, and limitations will be appreciated.

I am already running the 2x HP 524sh monitors and want to purchase the M24 monitor for my meetings. This is for a Home Office setup!! :)

Screen protector for surface pro 11 and slim pen

Kayla123 — Sun, 15 Mar 2026 18:38:03 GMT

Hi everyone.

I’ve had my surface pro 11 for a few weeks now and have been avoiding using the pen much because I haven’t been able to find a good screen protector. I hand write a lot for work and don’t want any scratches.

I got a tempered glass matte screen protector and I can barely write with the slim pen. It’s not great with responsiveness but it’s also really tough to write on. The pen barely moves and just gets stuck.

Any suggestions that will allow me to use my slim pen properly but also protect my screen?

Vibe Coding for the NPU

FrankBuchholz — Wed, 04 Mar 2026 21:53:07 GMT

If you just bought a Copilot+ PC and want to know what that NPU can actually do, this is for you. If you manage a fleet of them and need workload placement guidance, this is for you too.

Cloud doesn’t have to be the default anymore. The NPU (Neural Processing Unit) is now a practical and accessible development target, making it possible to run advanced AI workloads directly on your device. This means you can take advantage of your PC’s local hardware to get faster results, lower latency, and even work offline or in airplane mode. Getting started with NPU development is easier than you might expect.

Foundry Local serves models through an OpenAI-compatible endpoint on localhost, so if you’ve used the OpenAI SDK, you already know how to build for the NPU. Combine that with AI-augmented coding and you don’t need to be a pro developer... you just need to be specific about what you want.

I’m in Surface Marketing. I haven’t done daily development work since the Petzold “Programming Windows” days... back when writing a hello world app meant 90 lines of C and a WndProc callback. Early in my career I quickly figured out I wasn’t the most talented dev in the room, but I was good at integrating hardware solutions and telling the story of what they could do. Now, with vibe coding, the thing that wasn’t my vibe has become a superpower... “wait, could I vibe code that?” And the answer keeps being yes.

I just built a working on-device AI application with four tabs and five AI tasks by describing features to an AI coding assistant. This post is everything I learned along the way: the platform, the tools, the workflow, and the real considerations that only surface when you’re actually building on the hardware.

Ready to skip to the good stuff? Jump ahead:

Want a working app fast? → Your First NPU App in Minutes
Validating model choices? → AI Toolkit Model Catalog
Deploying to mixed fleets? → Cross-Platform Development + Operationalizing
Need vision capabilities? → Phi Silica
Want to see what we built? → The Surface NPU Demo App

Why does this matter? Three words: availability, economics, and data sovereignty. Details below.

Why Build for the NPU?

For years, the cloud has been powering our most demanding AI workloads. That makes sense for frontier reasoning, complex agent chains, the hard stuff. But the models that run locally now are good enough for the majority of what your fleet actually does. And by “locally” I mean the device in your user’s bag... their Surface on the train, in a customer lobby, on a job site. Not a server. Not a VM. The actual endpoint, with its own dedicated AI silicon.

Availability. An NPU-powered workflow runs in airplane mode, in a clean room, on a factory floor, in a field inspection truck in a dead zone. No connectivity required. The AI is on the device, and the device is wherever your user is.

Economics. NPU inference comes at no additional per-inference cost after the hardware purchase. No per-token API fees, no egress charges, no metered compute. I think of it as an 80/20 planning model: the majority of routine AI tasks can run locally at no incremental cost, reserving premium cloud inference for the tasks that genuinely need frontier reasoning.

Data sovereignty. When operating without cloud escalation, data is processed locally on the NPU, reducing your reliance on cross-border data transfers. As with any deployment, customers should assess their specific regulatory, legal, and operational requirements, including endpoint management, device access controls, and applicable local laws. For regulated environments, on-device AI can be a powerful part of a broader compliance and data governance strategy, but it does not replace the need for appropriate legal, contractual, and security controls.

What Is Vibe Coding?

The industry calls it AI-augmented development. The internet calls it vibe coding. Same thing: you describe what you want, an AI coding assistant writes the code, you run it, you tell it what broke, it fixes it. Repeat. You’re the architect and the QA engineer. The AI handles the implementation.

Examples include GitHub Copilot CLI, Cursor, and similar AI coding assistants. The specific tool matters less than the workflow: describe a feature, the assistant writes code, you run it, you report what worked and what broke, the assistant fixes it. The AI handles the boilerplate... Flask routes, CSS layout, regex, JavaScript event handlers. You handle the architecture decisions, the hardware testing, and the product judgment.

Why does this matter for NPU development specifically? Because AI coding assistants already know the OpenAI SDK patterns inside and out, and Foundry Local speaks a compatible API. “Compatible” means the standard SDK works for chat completions and most common inference patterns, though you may encounter minor differences in model IDs or streaming behavior. Microsoft’s documentation frames this as compatibility with “OpenAI-compatible SDKs and HTTP clients.” In practice, for the workloads covered in this post, the SDK works as-is. The barrier is significantly lower than it used to be... if you can describe a workflow, you can start building for the NPU.

How It Started: From Playground Curiosity to Working App

This is where the IT pro lesson starts: validate the model first, then write code. This didn’t start with a plan. It started with a click.

I opened VS Code, installed the AI Toolkit extension, and browsed the Model Catalog. There was a model already on the device: Phi Silica. I loaded it in the Playground, typed a message, and... it responded. On-device. No API key. No cloud endpoint. Just the NPU doing inference right there in VS Code.

Wait. I remembered from using LM Studio that local models get served on a port. If Foundry Local works the same way... could I write a web app that talks to it?

So I opened GitHub Copilot CLI and just started describing what I wanted:

“I have a local AI model running through Foundry Local on my Surface. It exposes a compatible API on a local port. Build me a Flask web app that connects to it and serves a chat interface.”

And we were off.

Copilot generated a working Flask app. I ran it. It connected to the local model. I could chat with an AI running entirely on the NPU through a web browser. No cloud, no API key, no subscription. Wild.

From there, the vibe coding loop took over. Each session I’d just describe the next thing I wanted: a sidebar with tabs, a daily briefing from local calendar data, a two-brain router for local-vs-cloud workload decisions, and finally a full field inspection workflow with voice, camera, pen annotation, and translation.

The real acceleration came when I moved from copy-paste iteration to GitHub Copilot CLI, which reads and edits the full codebase directly on the device. That’s when it went from a cool demo to a genuine multi-tab application with real architecture. Kevin Roose and Casey Newton on the Hard Fork podcast recently compared the leap in AI coding tools to the original ChatGPT moment... and having lived it, that tracks. (If you’re not listening to Hard Fork, you should be.)

Python, HTML, CSS, and JavaScript. Built primarily through conversation with AI coding assistants. By a marketing person who hadn’t done daily dev work since writing WndProc callbacks in C.

The Starting Point: AI Toolkit Model Catalog

Before writing any code, start where I started: the AI Toolkit for VS Code. Two things matter here: the Model Catalog and the Playground.

Open the Model Catalog and filter for models optimized for local NPU execution:

Model	Parameters	Strengths	NPU Support
Phi-4 Mini	3.8B	General text: summarization, extraction, generation, translation	Intel (OpenVINO), Qualcomm (QNN)
Phi Silica	N/A	On-device language and multimodal scenarios on Copilot+ PCs	Intel & Qualcomm (Windows AI APIs)
Qwen 2.5	7B	General text: larger context, stronger reasoning	Qualcomm (QNN), GPU fallback
Additional models	Varies	The catalog is growing. Check for new NPU-optimized variants regularly.	Varies by silicon

For the full local model landscape, see Ready-to-use local LLMs in Microsoft Foundry on Windows and the Windows AI overview decision tree.

The Playground is your validation sandbox. Select a model, load it, chat with it directly in VS Code. Test your prompts. Feel the latency. Hit the context limits. All before writing a line of application code. It’s also the most reliable method we’ve found to trigger the initial model download and readying for Phi Silica.

Platform Requirements

Hardware

Any Copilot+ PC with an NPU. Surface consumer Copilot+ PCs are currently Snapdragon X. Surface commercial (for Business) Copilot+ PCs are available with both Snapdragon X and Intel Core Ultra.

Silicon	NPU	Example Surface Devices
Intel Core Ultra (Lunar Lake)	Intel AI Boost, OpenVINO runtime	Surface Laptop for Business, Surface Pro for Business (Commercial)
Qualcomm Snapdragon X (Elite/Plus)	Hexagon NPU, QNN runtime	Surface Laptop, Surface Pro (Consumer), Surface Laptop for Business, Surface Pro for Business (Commercial)

Software Stack

Scope: Everything in this post is Windows-only. Foundry Local, AI Toolkit, and the NPU runtimes covered here require Windows 11 on Copilot+ PC hardware (Intel Core Ultra or Qualcomm Snapdragon X).

A note on Foundry Local: Foundry Local is currently in public preview. There is no SLA and no backward compatibility guarantee. Expect changes between releases and validate in staged rings before broad deployment. For the latest, see the Foundry Local documentation hub.

Component	Install	Learn More
Windows 11 24H2+	Windows Update	Windows AI APIs
Foundry Local (preview)	`winget install Microsoft.FoundryLocal`	Get started
Python 3.10+	`winget install Python.Python.3.11`
foundry-local-sdk	`pip install foundry-local-sdk`	SDK reference
OpenAI Python SDK	`pip install openai`	SDK integration guide
Flask	`pip install flask`
VS Code + AI Toolkit	VS Code Marketplace	AI Toolkit overview

⚠️ Package name warning: Install foundry-local-sdk (the official Microsoft SDK). Don’t confuse it with similarly named packages on PyPI. At time of writing, foundry-local (v0.0.1) is not the official SDK and may fail (or behave unexpectedly). Verify against Microsoft’s SDK documentation and PyPI.

Your First NPU App in Minutes

For the official quickstart, see Get started with Foundry Local. What follows adds the vibe coding workflow on top.

Step 1: Install the Runtime

winget install Microsoft.FoundryLocal
pip install foundry-local-sdk openai flask

Step 2: Prompt Your AI Coding Assistant

“Build me a Flask app that serves a chat interface on localhost:5000. The backend should use the OpenAI Python SDK pointed at a local Foundry Local runtime. Use the foundry-local-sdk to get the endpoint URL dynamically (don’t hardcode the port). The model alias is ‘phi-4-mini’. Make it a single-file app with the HTML inline.”

The assistant will generate something close to this:

from flask import Flask, request, jsonify
from openai import OpenAI
from foundry_local import FoundryLocalManager

# Start Foundry Local and discover the endpoint dynamically
manager = FoundryLocalManager("phi-4-mini")
client = OpenAI(base_url=manager.endpoint, api_key=manager.api_key)
model_id = manager.get_model_info("phi-4-mini").id

app = Flask(__name__)

@app.route("/chat", methods=["POST"])
def chat():
    user_msg = request.json["message"]
    response = client.chat.completions.create(
        model=model_id,
        messages=[
            {"role": "system", "content": "You are a helpful assistant."},
            {"role": "user", "content": user_msg}
        ],
        max_tokens=512
    )
    return jsonify({"reply": response.choices[0].message.content})

if __name__ == "__main__":
    app.run(host="127.0.0.1", port=5000, debug=True)

The key line is base_url=manager.endpoint. Don’t hardcode localhost:5272 or any other port. Foundry Local assigns a dynamic port each time the service starts. Always use the SDK’s endpoint discovery, or use foundry-local-sdk to resolve the active endpoint programmatically. For details, see the Foundry Local SDK reference.

Step 3: Run It

python app.py
# First run downloads the model (~3 GB). Subsequent starts are near-instant.
# Open http://localhost:5000

First model download requires network connectivity. After that, the model is cached locally and launches work fully offline. In practice, runtime or model updates and repair flows can occasionally re-trigger downloads. For cache management details, see the Foundry Local CLI reference.

Step 4: Iterate

Keep describing features. The AI coding assistant handles the implementation. You handle the testing on the actual device and the “this doesn’t work on the NPU” feedback that no AI assistant can discover on its own. Hardware-in-the-loop is the key. Don’t write a full spec. Write one feature at a time, test it, then describe the next.

Cross-Platform Development: Intel and Qualcomm

Foundry Local handles model selection across silicon families automatically through its alias system. For most developers and single-device use, you write one codebase and it works. If you’re deploying to a mixed enterprise fleet, here are the platform-specific optimizations we’ve found.

Architecture Detection Under Emulation

On Windows-on-ARM, Python x64 reports AMD64 via platform.machine(). Both it and PROCESSOR_ARCHITECTURE are wrong under emulation. Use WMI:

result = subprocess.run(
    ["powershell", "-NoProfile", "-Command",
     "(Get-CimInstance Win32_Processor).Name"],
    capture_output=True, text=True, timeout=5,
)
cpu = result.stdout.strip().lower()
if "qualcomm" in cpu or "snapdragon" in cpu:
    return "qualcomm"

Run native ARM64 Python where possible on Snapdragon devices for better performance and fewer emulation quirks.

Rule: On Snapdragon, prefer ARM64 Python. Avoid x64 emulation when you can.

Model Compatibility Varies by Silicon

Observed behavior during testing on preview runtimes (February 2026). Results may change with future Foundry Local or driver updates.

Model	Intel (OpenVINO NPU)	Qualcomm (QNN NPU)	GPU Fallback
Phi-4 Mini 3.8B	✅ Stable	⏳ NPU variant in development	✅ Both
Phi-3.5 Mini	✅ Stable	✅ NPU (Foundry 0.8.119+)	✅ Both
Qwen 2.5 7B	✅ Stable	✅ Stable	✅ Both
Phi Silica (text + vision)	✅ NPU	✅ NPU	N/A (Windows AI)

Foundry Local handles variant selection by alias. You request phi-4-mini, it pulls the best available build for your hardware. Model availability across NPU execution providers is expanding with each Foundry Local release. For tool-calling workloads on Qualcomm today, Qwen 2.5 7B is fully NPU-accelerated and production-ready. Your AI coding assistant handles the detection and model routing automatically.

Runtime Lifecycle Differs

if SILICON == "qualcomm":
    # Skip warmup - QNN is unstable with rapid reconnection attempts
    print("Skipping warmup on Qualcomm (first request will load model)...")
else:
    # Intel: warmup + keepalive for consistent latency
    warmup_model()
    start_keepalive_thread(interval=180)

Intel benefits from warmup and keepalive pings every 3 minutes. Qualcomm is the opposite: warmup destabilizes the QNN runtime. Let the first real request trigger model load. Use aggressive auto-reconnection on both platforms. These are observations from preview runtimes and may evolve.

NPU-to-GPU Fallback Chain

try:
    manager = FoundryLocalManager("phi-4-mini")  # NPU first
except Exception:
    try:
        from foundry_local.api import DeviceType
        manager = FoundryLocalManager("phi-4-mini", device=DeviceType.GPU)
    except Exception:
        # ⚠️ NOT RECOMMENDED FOR PRODUCTION - last resort only.
        # The port is dynamic; this may break if the service restarts.
        client = OpenAI(
            base_url="http://localhost:5272/v1",
            api_key="not-needed"
        )

For additional troubleshooting, see Foundry Local best practices.

Designing for the Token Budget

This is the constraint that shapes everything. Context limits vary by model. Phi Silica has a ~4K window; Phi-4 Mini supports much larger contexts, but practical token budgets still matter for latency and cost. Observed ranges: structured field extraction ~1,884 tokens, document classification ~500 tokens, morning briefing ~2,200 tokens.

Design principle: build single-shot endpoints that do one thing well. Instead of “one chatbot that does everything,” build endpoints like /extract-fields, /classify-doc, /summarize. One call, one job, one clean response.

And honestly? This constraint produces better architecture. Not every AI task needs a frontier model. Most of them are casual... summarization, extraction, classification, one-shot and done. It maps directly to the 80/20 planning model: the NPU handles the 80% that doesn’t need GPT-4.

Phi Silica: On-Device Vision

If you don’t need vision capabilities for your first app, skip this section. Start with Foundry Local + Phi-4 Mini. Come back when you need on-device image classification.

Phi Silica requires a few more setup steps than Foundry Local, but it unlocks on-device vision capabilities that no cloud-dependent workflow can match. Phi Silica is Microsoft’s on-device model accessed through the Windows AI APIs, supporting on-device language and multimodal scenarios available on Copilot+ PCs. For the full walkthrough, see Get started with Phi Silica and the Phi Silica tutorial.

Unlike Foundry Local (standard REST API), Phi Silica is a Windows API requiring: MSIX (Microsoft’s modern app packaging format) packaging with the systemAIModels restricted capability, a LAF (Limited Access Feature) token tied to your Package Family Name, and model provisioning on-device. Development builds may relax LAF enforcement; production deployments should always assume a valid token is required. See Windows AI API troubleshooting for LAF details.

Minimum Viable Checklist

☐ MSIX-packaged app with systemAIModels capability
☐ Phi Silica model readied on-device (AI Toolkit Playground or AI Dev Gallery)
☐ App checks GetReadyState() before calling vision APIs
☐ LAF token from Microsoft
☐ Health endpoint for fallback detection

The Sidecar Pattern

Encapsulate the Windows API complexity in a small C# ASP.NET Core service rather than MSIX-packaging your entire app:

Flask app (localhost:5000)  →  Vision Service (localhost:5100)  →  Phi Silica on NPU
                                      ↑
                               MSIX packaged, LAF token

The Vision Service exposes /health, /classify, /describe, /extract-text. Your primary app stays a standard Python web application.

The Three-Tier Fallback Pattern

If there’s one architecture principle to take from this entire post, it’s this: never let the app fail silently. Models will hang. Drivers will update. The NPU will occasionally just... not cooperate.

Tier	Strategy	Example: Photo Classification
Tier 1	Full AI pipeline (preferred)	Phi Silica Vision analyzes the actual image
Tier 2	Simpler AI approach (degraded but functional)	Phi-4 Mini infers from the filename
Tier 3	Hardcoded safe default (always works)	Pre-baked classification for known scenarios

Build these tiers from the start. They cost almost nothing to implement and they save everything when it matters.

Operationalizing NPU Apps in Enterprise

Foundry Local is in preview, so treat everything here as a framework that will evolve. That said, if you’re thinking about fleet deployment, these are the things that will bite you if you don’t plan for them.

Model acquisition and offline readiness. Models download on first use and cache locally (typically %LOCALAPPDATA%\.foundry\cache\models). Cache is per-user context, so plan pre-provisioning accordingly.

Update cadence. No SLA, no backward compatibility guarantee. Test in rings (dev → pilot → broad). Pin runtime versions. Revalidate after every Foundry Local or Windows update.

Logging. Track inference latency, token counts, fallback tier used, model load times. Do not log raw prompts or responses containing user data or PII (personally identifiable information).

Cloud escalation policy. Make cloud routing an explicit, auditable decision: local-by-default, cloud-only-when-invoked, with a log entry every time data leaves the device. Provide an admin toggle to disable cloud escalation entirely for sensitive environments.

Security. Foundry Local binds to localhost. Only local processes can reach the model endpoint. Don’t proxy to external interfaces without explicit security controls.

Packaging. Runtime installs via winget install Microsoft.FoundryLocal (Intune-compatible). (use --scope machine for machine-wide deployment; Intune-compatible). Stage three components: runtime, pre-cached model(s), and your application. The demo repo’s setup.ps1 validates each.

Fleet heterogeneity. If your fleet spans Intel and Qualcomm, the WMI detection pattern and three-tier fallback architecture are required, not optional. Test on both silicon families before broad deployment.

What We Built: The Surface NPU Demo App

Four tabs. Five AI tasks. Zero cloud calls in our demo. This is what a vibe-coded application looks like when it grows up.

I’ve demoed this to partner sellers, enterprise customers, and internal leadership. The moment that changes the conversation every time? Turning on airplane mode and watching it keep running. (That, and showing the tokenomics dashboard: $0.00 in cloud costs.)

Feature Details

AI Agent: governed local chat assistant powered by Phi-4 Mini through Foundry Local. Natural language queries, tool-calling, structured responses. This is where most people start.

My Day: takes local calendar, email, and task data and generates a structured morning briefing entirely on-device.

Two-Brain Router: evaluates each request and decides whether the local NPU can handle it or if it needs to escalate to a frontier cloud model. Shows the decision logic in real time, asks for explicit user consent before any data leaves the device.

Field Inspection Copilot: the multimodal showcase. Five NPU capabilities in a single workflow:

Voice: speak inspection findings, NPU transcribes and extracts structured fields (location, issue type, severity)
Camera: photograph the issue, Phi Silica classifies it on-device (water damage, structural crack, mold, equipment fault)
Pen: annotate photos with the Surface Pen, with local handwriting recognition
Report generation: NPU synthesizes voice, photos, and annotations into a formatted inspection report
Translation: one tap to translate the full report into Spanish entirely on-device

Target industries: construction, insurance claims, utilities, manufacturing QA, property management, OSHA compliance. The offline capability is load-bearing because these environments often have poor or zero connectivity.

The dashboard closes every session with the demo numbers: 5 local AI tasks, 0 cloud calls, ~520 tokens consumed, $0.00 in cloud inference costs, 0 bytes transmitted off-device.

The Code: Fork It, Build On It

github.com/frankcx1/surface-npu-demo

Clone it. Run setup.ps1. Open http://localhost:5000. It auto-detects your silicon, selects the right model, and you’re running.

Fork it and build your own use case on top of it. Add a tab for your industry workflow. Swap in a different model from the catalog. If you build something cool, submit a PR... I’ll merge it. This is a living project.

Lessons Learned

After building this across both Intel and Qualcomm devices over several months, here’s what I wish someone had told me on day one.

For IT Pros

The NPU is a production-capable inference target for supported workloads. Sustained low-watt inference for tasks that can cost pennies per call in the cloud (based on published Azure OpenAI pricing). Think about what that means for your fleet at scale.
Start with the AI Toolkit Model Catalog. Browse, test in the Playground, understand limits before committing.
Model constraints shape your architecture. Phi Silica is ~4K; Phi-4 Mini supports much larger contexts but token budget still matters for latency and cost. Design focused, single-task endpoints, not chatbots.
Cross-platform isn’t free. Intel and Qualcomm NPUs behave differently. Test on both. Use WMI. Build fallback chains.
Foundry Local is preview. No SLA. Stage updates through rings.
Airplane mode is the proof point. Turn off Wi-Fi. Kill the 5G. Run the app. That’s the demo that changes the conversation every single time.

For Vibe Coders

Start with Foundry Local + the OpenAI SDK. Fastest path to NPU inference.
Always use SDK endpoint discovery. base_url=manager.endpoint, not a hardcoded port.
Test on hardware early and often. Model hangs, context overflows, and driver quirks only surface on the actual device.
Hardcode fallbacks for everything. Not laziness. Professionalism.

What On-Device NPU Apps Are Not For

Just as important as knowing what to build... knowing what not to build. I learned some of these the hard way.

Not for model training. NPUs are inference accelerators.
Not for long-running agent loops. Small models lose coherence. Design for single-shot endpoints.
Not for unbounded conversation history. Manage state explicitly within the ~4K context window.
Not a cloud replacement. On-device handles the routine majority so your cloud budget goes to the tasks that need it.

Get Started Today

# 1. Install the runtime
winget install Microsoft.FoundryLocal

# 2. Install Python dependencies
pip install foundry-local-sdk openai flask

# 3. Clone the demo
git clone https://github.com/frankcx1/surface-npu-demo.git
cd surface-npu-demo
.\setup.ps1

# 4. Run it
python npu_demo_flask.py
# Open http://localhost:5000

# 5. Or start from scratch with your AI coding assistant:
# "Build me a Flask app that uses Foundry Local to serve
#  Phi-4 Mini on the NPU with an OpenAI-compatible API.
#  Use the SDK for endpoint discovery. Single file, HTML inline,
#  chat interface on localhost:5000."

Pro tip for vibe coders: Copy this entire post into your AI coding assistant as context. It knows the SDK patterns, the gotchas, the fallback chains. That’s the whole point.

Pick one workload this week. PII scanning. Document classification. Intake form extraction. Contract triage. Run it on the NPU, in airplane mode. Measure latency. Measure what you’d have paid in cloud inference. Build the business case from real numbers on your hardware.

I started at Silicon Graphics (SGI) and watched the introduction of the GPU up close... from rendering wireframes to reshaping entire industries. It’s wild to see how far that arc has come. Hardware has its own rise and fall and rise again, and dedicated AI silicon feels like the next big chapter. I’ve had the privilege of shipping things like Surface Hub along the way. The tools change. The builder instinct doesn’t. The difference now is that the tools meet you where you are... you don’t need to be an engineer to build something real.

Start building. Good luck out there.

Microsoft Learn References

Topic	Link
Foundry Local overview	What is Foundry Local?
Foundry Local quickstart	Get started with Foundry Local
SDK integration (Python, C#, JS)	Integrate with inference SDKs
Foundry Local architecture	Architecture and components
SDK reference	Foundry Local SDK reference
CLI reference	Foundry Local CLI reference
Windows AI APIs	What are Windows AI APIs?
Phi Silica	Get started with Phi Silica
Phi Silica tutorial	Phi Silica walkthrough
Windows AI troubleshooting	API troubleshooting
AI Toolkit for VS Code	AI Toolkit overview
Windows AI decision tree	Use local AI on Windows
Ready-to-use local LLMs	Local LLMs on Windows
Foundry Local GitHub	microsoft/Foundry-Local

Built on Surface Copilot+ PCs. Repo: github.com/frankcx1/surface-npu-demo

A pen for Surface Pro 7 and 4.

mr_K — Mon, 23 Feb 2026 20:33:00 GMT

Hi,

Is there a pen for Surface Pro 7 or Pro 4 that doesn't need charging?

Beautifully invisible: The engineering intelligence behind Surface for Business displays

Jussi_Ropo — Mon, 02 Mar 2026 20:03:51 GMT

Open a Surface Copilot+ PC, and the display just feels right. Whites look clean, text is sharp, and colors stay true and consistent in varied lighting conditions. Move from a bright office to a dim café and the screen keeps pace, adjusting as the light around you changes. That sense of ease is deliberate. It comes from years of engineering designed to make the display feel natural in any setting, with work that is meant to be invisible rather than attention-grabbing. In this blog post, we’ll look at some of the innovations behind the scenes that go into delivering a visual experience that just works.

Calibrated for consistency

Every Surface display goes through individual factory calibration before it leaves production. This is not a batch process. Each panel is measured and corrected on its own. Manufacturing introduces measurable differences between panels, and those differences—if left uncorrected—show up as variations in color, gamma, or luminance. Over the lifetime of a product, those variations erode consistency.

Figure 1. Distribution of display white points before and after factory calibration, shown in CIE 1931 xy color coordinates. Blue dots represent uncorrected panels; green dots show values after calibration, and the red point marks the target. Calibration reduces wide variation into a tight cluster centered on the reference point.

Figure 2. Visualization of white point consistency across panels. Before calibration (left), color shifts appear across different panels. After calibration (right), those variations are reduced, producing a uniform gray appearance.

Figure 3. Luminance error before (blue) and after (green) calibration shows how calibration lowers the variation and makes luminance difference between the units non-noticeable.

Surface calibration configures each panel to a defined target so that all units of the same model render content identically. A spreadsheet viewed in one conference room looks the same on another device halfway across the globe. Designers, video editors, and anyone working in color-critical applications rely on that assurance. For everyone else, the value is more subtle but no less critical. Photos display with reliable skin tones. Presentations render the same across the Surface product portfolio.

The calibration data is stored directly in the hardware, so it is present from the first boot. Users never have to set up or tune the display themselves. What feels natural reflects hundreds of hours of engineering and testing, with the goal of making accuracy something people never need to notice.

Responsive to its environment

Lighting conditions vary constantly. A user may start the day in a brightly lit office, move to a client site with warm incandescent light, and finish a call from a café with mixed natural light. In each of those settings, the display must maintain visual balance.

Figure 4. Comparison of Adaptive Color (left screen in images) versus fixed D65 white point (right screen in images). In warm ambient light (left image) and daylight (right image), Adaptive Color adjusts the display white point to better match the viewing environment, while the fixed white point remains constant.

Surface devices include Adaptive Display features that utilize ambient light sensors to track the environment's light intensity and color. If lighting in a room shifts from cool LED to warm halogen, the display adapts white balance (Adaptive Color), contrast (Adaptive Contrast), and brightness (Adaptive Brightness) accordingly. These adjustments are tuned to occur gradually, so there is no visible flicker or sudden change. Transitions are invisible, but the result is a screen that always feels immersive and natural.

This responsiveness requires the entire system to be tuned to work in harmony. Sensors capture data, firmware interprets it, and Windows algorithms apply adjustments in real time. Each layer must act with precision. If the response is too slow, the user notices lag; if it is too aggressive, the shift feels distracting. Because the display is calibrated to a known standard on Surface devices, the system can deliver adjustments that remain accurate across environments. The result is a screen that feels consistent, natural, and reliable no matter where it is used.

Figure 5. Anti-reflective technology comparison. The left display demonstrates how anti-reflective coating maintains visibility and readability even under glare. The right display shows distracting reflections and reduced clarity.

Anti-reflective coating technology further supports readability in dynamic ambient light conditions. It reduces reflections from overhead lights, windows, outdoor glare, and other variables that can interfere with visibility.

Context-aware experiences

Displays also need to respond to how people work. A static chart on a slide requires different behavior than fast scrolling in a lengthy document. A video clip requires different rendering than a CAD model or a digital painting.

Surface devices offer the option of using Auto Color Management (ACM) to detect the type of content and apply proper color management automatically. With support for high bit depth—more than a billion (10-bit) colors instead of 16.7 million (8-bit) colors—ACM provides greater range and accuracy across scenarios. High Dynamic Range (HDR) extends beyond video playback to include gaming, still images, and apps using Windows Advanced Color, with results that can deliver higher color saturation and expanded luminance range. Graphics built for the sRGB pipeline are displayed faithfully without oversaturation, and mixed workflows that include images with embedded profiles are interpreted correctly. Users do not need to switch modes or adjust settings—the system applies the proper treatment under the hood.

Motion designed to match the moment

When objects move across a screen, your eyes follow them smoothly. If the pixels that represent those objects don’t update at the same pace, your brain perceives that mismatch as motion blur—a faint smear that makes edges look soft and movement feel less immediate.

Two technical factors determine how a display handles motion: refresh rate (how often the image is updated) and response time (how quickly each pixel changes from one state to another). Higher refresh rates reduce positional lag between your eyes and the screen; faster response times minimize trailing or ghosting during transitions. Both are essential for clarity in motion, but both also consume more power.

Surface devices use Dynamic Refresh Rate to balance those parameters automatically. The display can run at up to 120 Hz for inking, scrolling, or animation—keeping motion sharp and continuous—then step down to lower rates when the image is still. Combined with tuned pixel response times, this adaptive system maintains precision without unnecessary energy use.

The result is motion that feels lifelike and immediate, whether you’re sketching with a pen, moving between slides, or gaming after hours. It’s a small part of the engineering work that makes the Surface display feel effortlessly responsive.

Figure 6. Motion Blur matrix for varying refresh rates (X axis) and response times (Y axis).

Building these systems requires deep integration across hardware, software, and display panel technology. The work involves years of alignment between design teams, Windows engineers, and manufacturing partners. The foundation comes from color science and human vision research, which guides how features are designed to match the way people perceive light and color. For the user, it simply feels seamless. The complexity stays hidden, and the display feels like it is paying attention to both context and activity.

Designed for the long run

Devices live through years of use, and sometimes through repair. A display may need to be replaced after damage or failure. For many products, a repair can mean differences in color or brightness that don’t match the original. Surface engineering avoids this by storing calibration data directly in the display hardware.

When a panel is replaced, the system retrieves the embedded profile and applies it automatically. The new panel inherits the same calibration standard as the original, and the experience remains consistent. Users do not need to recalibrate or adjust. Repairs become part of the lifecycle without breaking the quality of the display.

This attention to repair reflects a philosophy of long-term value. Devices are expected to stay reliable over years of service, and the display is central to that experience. Ensuring that a repaired device feels the same as a new one is part of treating quality as a continuous commitment rather than a one-time achievement.

Innovation that extends across Windows

Surface frequently pioneers new display technologies that later become available across the Windows ecosystem. Features such as Adaptive Brightness, Adaptive Color, HDR pipelines, Auto Color Management, and dynamic refresh rates have been refined on Surface before extending to other devices.

Partnerships with display manufacturers and technology providers make this scaling possible. For example, when Dolby Vision IQ support was introduced, Surface engineers worked with Dolby to tune the entire Windows pipeline so the experience could adapt to ambient light. That work benefited all Windows devices capable of supporting the feature, not just Surface models.

Because Surface devices often include a superset of display capabilities, engineering effort goes into making sure these features work seamlessly together rather than only in isolation. By solving complex problems once and sharing the solutions, the broader ecosystem improves.

This role as a reference platform means that innovation on Surface often shapes the standard for how displays behave across Windows. The work that begins as precision engineering for one device ends up defining expectations for many.

Engineering that gets out of the way

The Surface display team sees itself in the reproduction business. Every design choice—from panel selection to firmware algorithms—is guided by one principle: content should appear as its creator intended, no matter where it is viewed. The goal isn’t to make images look “better,” but to make them true—faithful to the artistic or functional intent behind them.

Most Surface display technologies, including factory calibration, adaptive color, contrast, and brightness, are built to maintain that fidelity across environments. Whether you’re in bright daylight or under warm indoor lighting, the display adapts so what you see remains accurate to the source. Only a few deliberate exceptions, such as the optional Enhanced and Vivid color profiles, are designed to offer a more interpretive, saturated aesthetic when desired.

This philosophy defines Surface display engineering. Every layer—from optics and electronics to software integration—is tuned to make the experience predictable, trustworthy, and human-centric. For the user, all that complexity disappears. The screen simply looks right. It adapts when the environment changes, responds when tasks shift, and stays consistent over time.

That is the measure of success for the engineers who build it: technology that works so well it disappears into the background. The Surface display is designed to be a window to the real world—a transparent, accurate medium through which people see work, ideas, and creativity exactly as they were meant to be seen.

Discover it for yourself. Explore more at www.microsoft.com/surface/business.

If you’re interested in learning how to set up a Surface display for evaluation, see Set up Surface devices for SDR & HDR display measurements on Microsoft Learn.

Is Microsoft planning firmware updates for Surface Laptop 3 devices for 2011 Secure Boot key?

Auld_Michael_B — Fri, 13 Feb 2026 00:57:05 GMT

As many people know, the UEFI Secure Boot Key provided as part of the Windows devices firmware is expiring in June of 2026. Microsoft have announced plans to update these keys for most Surface devices, but have not made any such announcement for Surface Laptop 3 devices. Will the company issue updated firmware for these devices before June? Or, does the company recommend that users replace these devices before the 2011 key expiration date?

The information publicly available is sparse, but it does appear that Microsoft no longer intends to update firmware for Surface 3 Laptops - which is fine, except that the after the key expiration, these devices will be substantially unusable, with impacts to UEFI Secure Boot, BitLocker and more... This is a Windows problem, and I will personally be very disappointed if Microsoft chooses to abandon users of these machines...

Miracast by end of 2025?

ChrstnMchl — Wed, 28 Jan 2026 15:48:07 GMT

It is now 2026. MS said early last year that Miracast was coming to SurfaceHubs by the end of 2025 and I haven't seen an update since. Has anyone seen something maybe I haven't? Why aren't there more posts about this exact topic? It seems like this should be core functionality...

Enhance AI Productivity with Designed for Surface Accessories

Jenn_Marescalco — Tue, 27 Jan 2026 22:29:20 GMT

In today’s workplaces, productivity thrives when technology helps people work smarter and with less friction. AI tools designed to support the way people think, create, and solve problems can help users move through tasks with greater confidence and clarity. But productivity and great AI experiences often benefit from more than compute power.

Designed for Surface (DfS) accessories provide meaningful extensibility to the Surface portfolio, enhancing work in real-world applications and quietly elevating the day‑to‑day experience in environments where devices need protection while remaining accessible and comfortable to use.

Below are a few scenarios that highlight some of the ways DfS accessories can achieve this.

Education

Pictured, left to right: JCPal Verskin Silicone Keyboard Protector (Spanish EU), STM Dux Rugged case for Surface Laptop 7th Edition (13.8"), JCPal Verskin Inclusive Keyboard Protector

Scenario:

A teacher walks between desks as Copilot summarizes next week’s lesson plan.
Students join a hybrid class, relying on inking, real‑time captions in Teams, and translation to keep up.
A principal catches a moment before dismissal to summarize a full day of meetings using AI.

All of this can happen when the devices in the room are charged and ready. Shared Surface Pros stay charged and organized in an AVerCharge C36i+ cart, so no one loses valuable class time hunting for power.

When backpacks get tossed in lockers or onto floors, STM’s Dux Rugged cases available for both the Surface Laptop 13” and Surface Laptop 13.8” absorb the impact and keep learning on track. And when students type in multiple languages, or need clearer visibility, JCPal’s VerSkin multi-lingual and accessibility‑focused keyboard covers help them participate more fully in class through AI‑supported writing and reading.

Teachers can easily attach and detach Kensington MagPro privacy screens to promote confidential testing environments.

These represent just a few of the intentionally designed DfS accessories to support learning, minimize interruptions, and increase inclusivity.

Frontline Retail

Pictured, left to right: MobileDemand Rugged xCase for Surface Pro with Magtek iDynamo, The Joy Factory Elevate II Countertop Kiosk for Surface Pro (White)

Scenario:

A manager uses Copilot to analyze weekly sales, prioritize restocking, and plan future orders based on seasonal shopping trends.
An associate uses a Surface Pro to pull up product details and show customers side‑by‑side options, helping them make quick decisions during busy moments.

Retail spaces are fast-paced and dynamic. Store employees can improve customer experience and reduce the time needed to address customer questions by tapping into Copilot to find the right product details, inventory availability, and more—within seconds.

During a rush, a retail associate can direct customers to self-checkout kiosks on a Surface Pro, securely mounted in The Joy Factory Elevate II enclosures, which can be placed anywhere with a variety of mounting options.

A manager can use Copilot to quickly identify inventory trends on their Surface Pro before moving to provide floor support, answering customer questions and taking payments to clear a long queue with their MobileDemand Rugged xCase in hand.

On a busy day, a Copilot+ PC—driven by powerful processors and all-day battery life and supported by DfS accessories—can help provide dependability in an ecosystem built for motion.

Information Workers & Power Users

Pictured, left to right: PanzerGlass Screen Protector for Surface Pro (Privacy), UAG Plyo Series Case for Surface Laptop (13")

Scenario:

A knowledge worker spends the morning reviewing documents with AI-generated summaries, then heads to a client site and continues refining content on the go.
Another teammate catches up on Copilot-summarized action items from a meeting, then jumps into a deep focus sprint.

Workflows feel natural when the setup supports them. A Kensington Elevated Stand raises the screen to a comfortable height for long hours of work. Sensitive content stays private in open offices or airport lounges thanks to PanzerGlass privacy screens. When work spans multiple locations, the UAG Plyo cases available for both Surface Pro 12” and Surface Laptop 13” keep devices protected without adding bulk, making it easy to carry Surface-powered productivity anywhere.

When the physical experience is comfortable, secure, and uninterrupted, AI becomes something people lean on effortlessly — not a feature they have to think about.

Your Everyday Support

Today’s workflow is becoming more streamlined and efficient. AI can accelerate work, sharpen focus, and free people from repetitive tasks, but only when the devices they depend on stay powered, protected, and ready for action.

The right Designed for Surface accessories make Surface more accessible and reliable in the real world, unlocking new functionalities. Together, they help people get more done, wherever and however their work happens.

Explore the full catalog of 200+ Designed for Surface accessories and see how they transform productivity by visiting DesignedforSurface.com.

surface pro x for business needs constant graphics driver reset, how to fix?

E_M_A_ — Tue, 27 Jan 2026 01:14:04 GMT

I have a surface pro x for business that is up to date and has no company or IT restrictions, it is my device for home use and I have control of everything. I am noticing that whenever my device wakes up from sleep, restarts, or turn on after shutting down, I have to do ctrl+shift+windows key+B to get all the content to fit into the screen. For instance, if I have the microsoft store page open, and my windows are maximized (I even tried this on several browsers after minimizing and maximizing again) the bottom of the page at any given position does not fit the screen. Another example is ctrl+f, I cannot see that bar that searches and highlights the result or on word I cannot see the bootm row that has word count and language until I reset the graphics driver. I saw an optional update and it temporarily fixed it but the issue has returned. Any advice on how to fix this on windows 11 (I have the i7)?

Surface Laptop Power Issues?

Mike Curtis — Thu, 22 Jan 2026 01:17:00 GMT

Hi,

I have a Surface Laptop that’s 5? 6? years old. The other day it acted like it was no longer connected to AC power and shutdown when the battery got low. It was plugged in the whole time. The power adapter lights up when connected to the laptop like there is no issue. I took it to Best Buy to try one of their adapters and they let it sit, plugged in, for 5 minutes but it does not look to be charging. Pressing the power causes it to act like it turns on, but displays a battery on the screen after the Microsoft logo. My only recourse seems to be to ship it to uBreakIFix but I would like to avoid that route. Any ideas or suggestions?

Thank you,

Mike

Elevate teaching and learning with AI-powered experiences on Surface Copilot+ PCs

Jeanine_Park — Wed, 21 Jan 2026 19:25:55 GMT

Surface is a premium endpoint, designed and built by Microsoft to run Microsoft technology. When it comes to the classroom, Surface Copilot+ PCs bring the best of Microsoft—hardware, Windows, Microsoft 365¹, and Microsoft 365 Copilot¹—into one teaching device, delivering intelligent experiences more securely on‑device and in the cloud. For educators, this means every lesson and interaction is powered by a device purpose-built for teaching and learning in the digital era.

Why Surface for Education?

Surface Copilot+ PCs, combined with Windows 11, give teachers a powerful platform designed to simplify teaching and elevate learning. With lightning-fast performance, educators can create engaging lessons, generate content, and personalize instruction. Windows 11 features like Snap Layouts, Click to Do², and Copilot Voice streamline multitasking and lesson prep, while intuitive touch, pen³, and voice input make teaching feel natural. Together, Surface and Windows 11 help deliver a more secure, AI-supported solution that can save time, support creativity, and help teachers focus on what matters most—students.

Engage students with AI-enhanced learning

Picture a classroom where every student is actively engaged, their curiosity sparked by lesson plans and quizzes thoughtfully designed by teachers with the help of AI. Microsoft Learning Zone, included with all Microsoft Education licenses at no extra cost⁴, is an AI-powered learning app for Windows 11 designed to help educators create engaging lessons. The app’s AI powered lesson creation feature is designed specifically for Copilot+ PCs. Thanks to the Copilot+ PC’s built-in Neural Processing Unit (NPU), these devices offer fast, reliable performance by running AI models directly on the device and combining them with cloud-based capabilities when needed. This hybrid approach helps Microsoft Learning Zone generate lessons quickly and keeps the experience smooth, secure, and ready for classroom use. It also helps educators streamline lesson planning with interactive activities, instant feedback, and personalized learning pathways. For example, teachers can use Microsoft Learning Zone to develop Kahoot! quizzes for a whole class or to prepare individualized learning experiences like personalized practice questions based on recent student performance, while keeping student data more secure and private.

Surface Copilot+ PCs can also help teachers use AI in context to streamline lesson prep and administrative tasks, so they can spend less time on administration and more time inspiring students. For example, teachers can engage with Copilot in a single click using the Copilot key⁵ on a Surface keyboard or by saying “Hey Copilot” out loud. And using pen, touch, and voice commands in conjunction with Teach in Microsoft 365 Copilot⁶ running on Surface, teachers have a central hub for generating lesson plans, quizzes, rubrics, flashcards, feedback, and more. They can transform ideas and research into engaging lectures in moments, tailor instruction to meet the needs of every learner, and connect with colleagues to share best practices.

Recently, we also announced the Microsoft Elevate for Educators skilling program along with more AI-powered experiences purpose-built for education, including the Study and Learn agents in Microsoft 365 Copilot, and Study Guide. Microsoft Elevate helps equip educators with the AI skills they need for the classroom of the future. Delivered through platforms like Microsoft Learn and Minecraft Education, these flexible learning paths ensure that educators can build AI fluency at their own pace, whether they're just beginning their journey or advancing to specialized applications. Read more about Microsoft Elevate and other AI tools for education here.

In classrooms where both teachers and students are using Surface Copilot+ PCs, Live Captions⁷ with on-device automatic translation can help make spoken content accessible to all students, including students with hearing impairments. The NPU transcribes and translates audio in real time, supporting 40+ languages into English, all processed on-device versus the cloud.

Ideally with Surface Copilot+ PCs and Microsoft’s AI-powered tools, technology fades into the background in the classroom, helping teacher creativity and connection take center stage and enabling student learning to be more dynamic, inclusive, and impactful.

Empower the classroom of the future, today

Surface Copilot+ PCs, purpose-built by Microsoft, are designed to be the foundation for today’s classrooms and the launchpad for tomorrow’s AI innovations. With Windows evolving as the canvas for intelligent AI and agents, Surface devices and Windows together form an essential, AI-assisted platform for educators.

Now, every teacher can activate Copilot agents⁸ directly on their Surface device to act as a digital teaching partner. These agents can adapt to each teacher’s unique style, streamline daily routines, and unlock new possibilities for student learning. Building agents is simple: teachers or IT can use ready-made templates or create custom agents using natural language, all within the familiar Surface and Windows environment.

Surface’s intuitive hardware—touchscreen, pen, voice, and the dedicated Copilot key—makes accessing AI support effortless. Teachers can get immediate answers to classroom questions, troubleshoot tech issues, or navigate school resources using natural input methods. By combining Copilot’s intelligent capabilities with Surface’s secure hardware, educators can gain a more personalized, efficient, and protected teaching experience ready for the future of learning.

In addition, Surface Copilot+ PCs help support the full range of learning needs by delivering the performance and experience required for the education tools students and educators depend on every day. Surface devices are designed to work with common education apps like TestNav for assessments, Minecraft Education for STEM, Adobe Express for creativity, and assistive technologies such as JAWS.

And, going forward, the built-in NPU on Copilot+ PCs like Surface enables Microsoft and other educational software providers to develop innovative new AI experiences that can run on the device, in the cloud, or both.

Boost productivity and collaboration

In the classroom, Surface Copilot+ PCs can become the teaching command center: always ready, always responsive. With a simple voice command, Copilot Voice and improved Windows Search⁹ instantly pull up lesson plans, student materials, or answers to unexpected questions, freeing instructors from frantic searches and giving back precious prep time.

Collaborating on Surface is intuitive and efficient. Teachers can quickly save lesson materials in Teams or OneDrive¹ and share them with students for interactive feedback. Whether teachers are leading a lively discussion in person, connecting with students remotely, or conferencing with colleagues via video, Windows Studio Effects on Copilot+ PCs ensure they’re always seen and heard clearly. Subtle features like background blur, eye contact, and automatic framing help maintain a professional presence, so the focus stays on interacting with students or other teachers—not on the tech.

And Surface Copilot+ PCs are designed to empower mobility in and outside the classroom. With extended battery life and lightweight devices, teachers are no longer tethered to a desk or a charger. They can move about freely, interact with students, project and present seamlessly, and focus on teaching. For example, on a Surface Pro with Surface Pen inking, Dual Studio Mics, and natural language prompts in Copilot, teachers can annotate readings, capture ideas by voice, and generate lesson materials on the fly—without breaking the flow of instruction.

Throughout the day, Surface Copilot+ PCs can also help teachers anticipate what’s next. Context-aware Windows Search doesn’t just find files—it suggests smart next steps, like opening a document in Word or sharing it with a colleague, streamlining workflows. When inspiration for a lesson strikes, Click to Do lets teachers quickly summarize, explain, or create new content on the fly, helping them build engaging lectures and materials in real time. Surface’s vibrant touchscreen and Snap Layouts can help keep resources organized and visible, supporting a productive work experience.

And, most importantly, Surface devices help safeguard faculty, staff, students, and sensitive school data with advanced security and remote management features¹. As innovations advance in AI, Microsoft and Surface provide built-in protection at every layer—hardware, firmware, operating system, cloud, software applications, and identity.

Surface Copilot+ PCs are more than just devices—they’re partners in teaching, learning, and innovation. By combining Microsoft’s advanced hardware, intuitive software, and powerful AI, Surface empowers educators to engage students, boost productivity, and modernize their classrooms, all while keeping data more secure and private.

Visit Surface.com/Business to learn more.

Students, parents and educators can save up to 10% on select Surface devices and more at the Microsoft Store.¹⁰

Disclaimers:

Sold separately. Software license required for some features.
Click to Do. Copilot+ PC feature. Image actions now available across devices; other actions vary by device, region, language, and character sets. Subscription required for some actions. Learn more.
Surface Pen sold separately. Surface Slim Pen (2nd Edition) experiences and compatibility vary by which Surface device you are using it with. Visit Surface Slim Pen Compatibility to learn more.
Microsoft Learning Zone requires a Microsoft 365 Education Essentials, Core (A3), or Advanced (A5) license. Microsoft Education licenses.
Copilot key feature availability varies by market, see aka.ms/keysupport.
This feature is only available to Faculty/Staff with a Microsoft 365 for Education license. Teach in the Microsoft 365 Copilot App.
Copilot+ PC feature. Live Captions translates video and audio subtitles into English from 40+ languages and from 25+ languages into Chinese (Simplified). See Copilot+ PC FAQs.
Copilot with commercial data protection is available at no additional cost for users with an Entra ID with an enabled, eligible Microsoft 365 license. Copilot for Microsoft 365 sold separately and requires a qualifying volume license or subscription - Microsoft Copilot for Microsoft 365 | Microsoft 365. Minimum age limits apply to use of Copilot and certain AI features. Details.
Copilot+ PC feature. Improved Windows search works with specific text, image, and document formats only; optimized for select languages (English, Chinese (Simplified), French, German, Japanese, and Spanish). See Copilot+ PC FAQs .
Microsoft Store Education discount is available to K-12 and higher education students, faculty and parents. Education discount only valid on select products, and may not be combinable with other offers. See terms and conditions at Education & Student Discounts on Laptops, Microsoft 365, Windows, Surface | Microsoft Store

ICYMI: Enabling AI Fluency in 2026 with Surface Copilot+ PCs

Chauncey_Larsen — Wed, 14 Jan 2026 01:06:02 GMT

Welcome to 2026.

In case you missed it, Sandra Andrews, Global GM of Surface Marketing, recently posted an article detailing ways businesses and Frontier Firms can enable AI fluency using Surface Copilot+ PCs. Head over to LinkedIn to read the full post:

Enabling AI Fluency in 2026: How Surface helps organizations become Frontier Firms

Also, be sure to follow Microsoft Surface on LinkedIn for more updates: Microsoft Surface | LinkedIn

Stay tuned and keep exploring what's possible!

Meeting Control, Now Within Reach: Surface Hub 3 + MAXHUB Universal Console

FrankBuchholz — Tue, 13 Jan 2026 21:09:08 GMT

Modern meeting spaces require flexibility, simplicity, and consistency for IT teams managing collaboration environments. Announced at Microsoft Ignite 2024, center-of-table console support for Surface Hub 3 extends Microsoft Teams Rooms on Windows by enabling certified third-party touch consoles for in-room meeting control.

Center-of-Table Console Support

Surface Hub 3 now supports Teams Rooms Certified Universal Consoles, introducing an additional option to control meetings beyond the interactive display. This includes the MAXHUB TCP33T Universal Console, a certified hardware solution in this ecosystem. By connecting a touch console directly to Surface Hub 3, users can join meetings, manage controls, and share wired content from their PCs while seated at the table.

Benefits of Certified Universal Consoles

Teams Rooms Certified Universal Consoles can help provide IT admins with:

Microsoft Teams Rooms integration through certified accessories
Flexible room design options including front-of-room deployments
Consistent user experience across meeting spaces
Simplified meeting controls for table-based interaction

MAXHUB TCP33T Universal Console Overview

The MAXHUB TCP33T is a Teams Rooms Certified Universal Console designed to work with Windows-based touch board devices such as Surface Hub 3. It delivers touchscreen control for Microsoft Teams Rooms meetings and serves as a dedicated entry point for wired content sharing.

Common Deployment Scenarios

Large conference rooms requiring table-based control
Standardized Teams Rooms deployments across buildings
Collaborative meetings requiring easy content sharing

Requirements

Deployments require Surface Hub 3 running Microsoft Teams Rooms on Windows and a Teams Rooms Certified Universal Console such as the MAXHUB TCP33T.

Looking Ahead

This new Universal Console ecosystem reflects Microsoft's commitment to flexible, scalable collaboration experiences. By pairing Surface Hub 3 with certified accessories like the MAXHUB TCP33T, organizations can design meeting spaces that meet modern collaboration needs.

Live Webinar Wednesday the 14th of January... Sign up now!

Join us for an exclusive webinar with Microsoft and MAXHUB as we explore how the MAXHUB Universal Console enhances the Microsoft Surface Hub experience.

Hear from industry experts and learn how to:
✔ Reduce friction in the room
✔ Streamline collaboration workflows
✔ Centralize meeting controls — without leaving your seat!

Featured Speakers:
• Frank Buchholz, Director, Product Marketing — Microsoft
• Jean Barrett, Technology Solutions Professional — Microsoft
• Alan Nicholls, Microsoft Teams Professional — MAXHUB

Register for a session and secure your spot👇

January 14 @9AM PST session: Register here
January 14 @5PM PST session: Register here

Resources

Will the 26H1 be available to Snapdragon-powered Surface Laptop 7?

hud — Tue, 13 Jan 2026 04:14:27 GMT

The news mentioned that the next Windows 11 update is specifically for Snapdragon-powered X2 ARM64 PCs. Since Surface Laptop 7 is also using the same ARM architecture, would it be great if these laptops receive the same update so that the performance will be optimized too?

Surface Pro 8 – Hardware-Liquid Damage

augustusleo — Fri, 09 Jan 2026 22:48:44 GMT

Device: Microsoft Surface Pro 8

Pre-incident condition: Fully functional, no known hardware faults

Incident Details

A small Coca-Cola spill occurred around the Type Cover (keyboard) connector area only
Area exposed cleaned immediately
No liquid exposure to USB-C ports, Surface Connect charging port, vents, or display
Type Cover remained attached following the spill
Device was left connected to the charger for approximately 3 hours after the spill, with the keyboard still attached

Observed Behaviour (Following Morning)

Device will not power on unless connected to AC power
Investigated- Found liquid around type cover (nothing visible on the ports)
When powered on:

Boots normally
Reaches Windows login
After login and loading the desktop, the system shuts down abruptly without warning

Shutdown occurs consistently after OS load
No shutdown occurs during UEFI/POST

Security / Firmware Event

Device subsequently displayed a BitLocker recovery prompt
Correct recovery key was entered
System successfully decrypted and booted into Windows
Device again shut down abruptly after reaching the desktop

Current Status

Device is currently powered off
No factory reset, OS reinstall, or internal cleaning attempted
Type Cover and charger have been disconnected
No visible external corrosion or physical damage

Assessment Context

SSD, display, and CPU appear functional (successful OS load)
Failure appears load-dependent and occurs post-login
Behaviour suggests possible power delivery, embedded controller, or internal board-level instability
No user-replaceable component has been identified as a definitive failure point

Please advise on:

Viability of the device for reliable continued use
Whether the observed behaviour is consistent with recoverable hardware condition or non-economical repair
Whether Microsoft recommends replacement vs. repair given the described failure pattern

Surface Hub 2s with MTR-W no Miracast support!

GdWEijkelkamp — Tue, 06 Jan 2026 14:18:28 GMT

We have 3 Surface Hubs 2S, which are recently have been updated to MTR-W (W11) (What a bummer...)
What we desperately need is Miracast support!

We've tried to set this up with the admin account, and once you are in admin mode you are able to configure 'Projecting to this PC' but back in normal mode, you can't use Miracast (Windows-K) anymore. I've read Miracast will become available for Surface Hub 3 but there are no plans to support the Surface Hub 2s.

Anyone out there who is/was facing the same problem? Did you manage to solve this? If so, how did you solve this?

Thank you,

Gerrit

Wifi issues - need to configure wifi for Machine based authentication

Michael_77 — Tue, 06 Jan 2026 11:49:49 GMT

All our devices now do not connect to the government available wifi called GovWifi.

This is due to the fact the way the devices sign in using Windows 11 with the Skype account having no security assigned to it as it auto signs in.

Is there a way to create an xml script or setting to allow this wifi to be set as machine based authentication please.

These Windows 11 MTR builds have been very troublesome and we've experienced issue after issue since implementing these builds, so suggestions would be welcomed.

Thanks

Michael