Gen Z’s growing fascination with journals, vinyl records, film cameras, knitting, sewing, physical books, and other tactile hobbies is often described as a rejection of technology, but that interpretation misses the most important part of the trend. Younger users are not necessarily abandoning computers; they are pushing back against ambient computing—the endless stream of notifications, algorithmic recommendations, synthetic content, and AI prompts that turns every idle moment into an engagement opportunity. That distinction could create an unexpected opening for Windows PCs, provided Microsoft can stop treating the operating system as an attention platform and rediscover the value of a computer that behaves like a quiet, dependable workbench.
The analog revival has been building for years, but it has gained new meaning as generative AI spreads across education, entertainment, creative work, search, social media, and productivity software. What once looked like nostalgia now resembles a broader attempt to reclaim agency over time, attention, and personal expression.
Michaels’ 2026 Creativity Trend Report offers a useful snapshot of that shift. The craft retailer reported that searches for analog hobbies—including knitting, crocheting, embroidery, painting, and journaling—rose by 136% over six months, while yarn accessory sales increased 40% year over year. Some narrower product categories recorded much larger spikes, illustrating how quickly tactile, handmade activities can move from niche communities into mainstream culture.
These figures should not be treated as a complete measurement of an entire generation. Michaels has an obvious commercial interest in promoting craft trends, and social-media-fueled aesthetics can fade as quickly as they appear. Even so, the pattern extends beyond one retailer: younger consumers are showing renewed interest in physical media, screen-free gatherings, disposable and film photography, traditional crafts, printed books, dedicated music players, and simpler communications devices.
The analog revival is potentially more durable because it changes the activity itself. Writing in a paper journal is not merely writing without Wi-Fi; it removes notifications, copy-and-paste behavior, infinite revision, automated suggestions, and the pressure to publish the result. Knitting, painting, baking, or developing film similarly makes slowness part of the experience rather than a technical inconvenience to overcome.
That helps explain why the revival is intersecting with AI fatigue. Generative systems promise faster output, but speed is not always the objective. For many people, the process of making something—and knowing that they made it—is precisely where the value lies.
That familiarity can produce skepticism rather than unquestioning enthusiasm. Younger users understand how quickly a short check of a social app can become an hour of scrolling, how online aesthetics are manufactured, and how algorithmic systems reward imitation. They also face a digital environment increasingly filled with AI-generated images, text, audio, advertisements, summaries, and comments whose origins are difficult to verify.
The attraction of analog activities is therefore not necessarily backward-looking nostalgia for a world Gen Z never experienced. It can be a search for boundaries that digital products have systematically removed.
Generative AI makes the problem more complicated because it is not confined to one application. It can appear in search results, operating-system interfaces, office suites, messaging tools, photo editors, customer-service portals, development environments, and educational platforms. Even users who appreciate AI for specific tasks may resent being repeatedly prompted to use it.
Human activities do not always fit that calculation. The person keeping a journal may not want a model to rewrite an entry. A photographer may enjoy selecting settings manually. A student may need to struggle with a paragraph to understand the subject, while a programmer may learn more from debugging code than from accepting a generated fix.
When AI removes effort, it can also remove practice, discovery, and ownership. That does not make automation inherently harmful, but it means the user should decide which parts of an activity deserve acceleration.
A knitted scarf may be less uniform than a factory-made equivalent, yet the unevenness becomes evidence of the maker’s time. A handwritten note may be less legible than a printed message, but it carries a stronger sense of intent. A photograph captured on film cannot be reviewed, regenerated, or instantly perfected, making each frame feel consequential.
This preference does not require hostility toward AI. It reflects a changing economy of meaning in which effort itself becomes part of the product.
The NPU was intended to handle sustained machine-learning workloads more efficiently than a CPU or power-hungry discrete GPU. Microsoft paired the hardware category with Windows experiences such as Recall, Click to Do, enhanced search, Live Captions translation, and Windows Studio Effects.
Technically, the strategy made sense. The PC industry needed a reason to encourage upgrades, chipmakers had begun integrating more capable NPUs, and local inference offered potential advantages in latency, privacy, offline availability, and cloud costs.
The problem was not simply the technology. It was the way Microsoft attempted to make AI the organizing narrative for Windows.
That naming created avoidable confusion. A customer could buy a computer with a physical Copilot key without necessarily understanding which capabilities ran locally, which required an internet connection, which needed a subscription, and which depended on organizational policies.
The branding also made every AI integration feel like part of one aggressive campaign. Even modest features inherited the controversy surrounding more sensitive capabilities such as Recall.
Microsoft subsequently delayed, redesigned, and hardened Recall, adding stronger controls and emphasizing opt-in deployment and Windows Hello authentication. Those changes addressed important technical concerns, but the initial announcement had already shaped public perception.
The episode demonstrated a central problem with ambient AI: even if a feature runs locally, users may object to the computer observing more than is necessary. Local processing can improve privacy, but it does not automatically establish consent or trust.
Microsoft said it would reduce AI integrations where they did not add enough value, initially naming applications and surfaces such as Notepad, Photos, Snipping Tool, and Widgets. The company also promised broader work on responsiveness, memory use, interface consistency, and long-requested customization.
This was not an abandonment of AI. Microsoft continued investing in Windows AI Foundry, local models, agent infrastructure, Copilot services, and AI-capable hardware. The shift was instead about becoming more selective—and admitting, at least indirectly, that ubiquity is not the same as usefulness.
Windows AI APIs now account for a more heterogeneous hardware landscape. Depending on the model and workload, supported devices can use NPUs, GPUs, or CPUs, with newer NVIDIA and AMD graphics hardware gaining access to capabilities that once appeared closely associated with Copilot+ machines.
That evolution is practical. NPUs remain valuable for low-power, persistent workloads, especially on laptops, but discrete GPUs offer far greater raw throughput for many generative tasks. A desktop with a powerful graphics card should not become artificially ineligible for useful local AI merely because it lacks an NPU matching a marketing specification.
Modern Copilot+ laptops also deliver benefits that have little to do with whether a buyer uses generative AI. The latest Qualcomm, Intel, and AMD platforms have pushed Windows notebooks toward longer battery life, better standby behavior, quieter operation, and stronger performance per watt.
The category’s future may depend on whether Microsoft treats the NPU as a useful component rather than the computer’s entire identity. Buyers can appreciate efficient silicon without wanting every application to announce an AI assistant.
A traditional PC has a different physical and behavioral relationship with its owner. It usually requires a deliberate transition: opening a laptop, sitting at a desk, connecting peripherals, or entering a workspace. That small amount of friction can be beneficial because it marks the beginning and end of a computing session.
A well-configured PC can function similarly. It can be used to write a long document, organize photographs, design a pattern, record music, research a repair, edit video, manage finances, or play a game without remaining the user’s constant companion.
This does not mean PCs lack distractions. Windows contains notifications, advertisements, recommendations, browser prompts, news feeds, chat clients, game launchers, and attention-seeking third-party software. The difference is that these interruptions are not structurally inseparable from the device.
The keyboard is particularly important. Typing on a physical keyboard encourages sustained composition, while keyboard shortcuts enable experienced users to manipulate files and applications without navigating layers of touch-oriented menus. A stylus can similarly bridge analog and digital workflows for illustration, handwriting, annotation, and design.
The renewed interest in mechanical keyboards is part of the same cultural pattern. Buyers are choosing switches, keycaps, layouts, and sound profiles not because the fastest possible text entry demands such customization, but because tactile feedback makes interaction feel intentional.
Intentional computing would reverse that relationship. Instead of asking how Windows can occupy more of the user’s day, Microsoft could ask how the PC can help the user accomplish a chosen task with fewer interruptions.
A coherent intentional-computing mode could follow a clear sequence:
In each case, the user defines the objective and initiates the operation. The system processes a bounded set of data, shows what it is doing, and provides a result that can be accepted, edited, or rejected.
That model differs fundamentally from ambient AI that watches activity continuously, inserts buttons throughout the interface, or predicts that the user might want assistance. The difference is not the underlying model; it is the balance of power between the software and the person operating it.
A laptop can support analog hobbies without replacing them. It can store scanned journal pages, print sewing templates, catalog a vinyl collection, edit photographs after development, manage recipes, design labels, or provide long-form instruction before being closed and removed from the activity.
Photography provides another example. Film cameras create scarcity at capture time, but a PC remains useful for scanning, color correction, archival storage, printing, and sharing. The digital tool expands what the user can do after the tactile experience rather than erasing the experience itself.
Windows has an advantage because it supports an enormous range of printers, scanners, cutters, cameras, audio interfaces, drawing tablets, and specialist peripherals. Better device reliability and less intrusive software would make that ecosystem even more attractive.
Windows gaming nevertheless suffers from its own layers of friction: multiple launchers, overlays, account prompts, background updaters, driver utilities, storefront notifications, and performance-hungry services. Microsoft could apply the workbench philosophy here by improving full-screen focus, suppressing irrelevant processes, and making the Xbox app a cleaner organizer rather than another engagement surface.
The lesson is not that every deliberate activity is healthy by definition. It is that products should help users understand and control their attention rather than deliberately dissolving its boundaries.
AI can improve productivity, but uncontrolled integration can increase cognitive load. Workers may spend as much time evaluating generated material, correcting errors, checking sources, and managing prompts as they save through automation.
An enterprise-friendly Windows strategy should provide centralized control over:
Windows devices used in classrooms would benefit from clear learning modes that limit generative assistance during assessments, expose source provenance, and allow teachers to define acceptable tools. Such controls would be more useful than simply placing Copilot everywhere and leaving each school to invent policies after deployment.
Analog activities have pedagogical value here as well. Handwriting, physical experiments, drawing, reading printed material, and face-to-face discussion create different cognitive conditions from typing into a chatbot. A Windows PC should integrate with those methods, not assume that every educational experience improves when mediated by AI.
The analog revival gives Microsoft an opportunity because Windows does not depend on one vertically integrated consumer ecosystem. Its strength lies in variety: compact laptops, repairable notebooks, gaming towers, professional workstations, tablets, mini PCs, handhelds, and custom-built desktops can all serve different forms of deliberate computing.
Windows manufacturers cannot rely on a single design language or support model. However, they can compete through hardware diversity, broader peripheral compatibility, gaming, upgradeability, repair options, and price.
Microsoft’s role is to ensure Windows does not undermine those advantages with inconsistent interfaces and unsolicited additions. A beautifully engineered laptop cannot feel calm if the operating system repeatedly advertises services or changes default behavior after updates.
Windows does not need to imitate those platforms by hiding all complexity. It needs to make complexity optional. Novices should receive a clean and stable default experience, while enthusiasts retain access to advanced settings, local accounts, file management, scripting, virtualization, and hardware controls.
That balance would make Windows more compatible with both digital minimalism and the traditional PC enthusiast ethos.
The key is restraint. Every notification, taskbar icon, context-menu item, background service, recommendation, and assistant prompt consumes a small amount of attention. Individually, each addition may appear defensible; collectively, they can make the operating system feel adversarial.
Most people use default settings. They may not know which switch controls a particular behavior, and Windows settings have moved between interfaces over successive releases. If users must follow lengthy guides to make the operating system quiet, quietness is not truly a supported experience.
Microsoft should make the least intrusive configuration the default and allow users to opt into additional engagement. This would also reduce suspicion around genuinely helpful features because their appearance would reflect an affirmative choice.
Microsoft’s renewed focus on performance, reliability, and craftsmanship therefore aligns naturally with intentional computing. A dependable PC fades into the background, while an unreliable one constantly demands attention.
The analog revival is partly attractive because physical tools tend to make their state visible. A notebook does not update itself, change its layout, or require an account. Windows cannot become that simple, but it can become more predictable.
Microsoft’s treatment of defaults will be especially important. A feature that can technically be disabled but arrives enabled after an update still represents an attention-first design decision.
NPUs will retain an efficiency advantage for continuous, low-power processing. GPUs will remain preferable for heavier models and creative workloads. Windows needs a transparent scheduler and compatibility framework that selects the appropriate accelerator without forcing customers to understand every architectural detail.
Those qualities fit the analog revival better than a dedicated AI key. They make the machine feel like a reliable object rather than a constantly evolving service endpoint.
The most revealing signal may be how younger buyers describe their ideal computer. If they prioritize control, longevity, privacy, repairability, and focused creation over constant assistance, Microsoft and PC makers will have to reconsider what “modern” computing actually means.
Gen Z’s analog turn does not require Windows to imitate a notebook, a film camera, or a record player. It asks Microsoft to respect the qualities that make those objects appealing: clear purpose, visible boundaries, predictable behavior, personal ownership, and freedom from unsolicited intervention. The Windows PC can still run advanced local models, cloud services, professional applications, and demanding games, but its most valuable feature may be the ability to wait quietly until someone chooses to use it. If Microsoft can make Windows feel less like an engagement engine and more like a dependable tool, AI fatigue may not push younger users away from the PC at all—it may remind them why the PC remains worth keeping.
Overview
The analog revival has been building for years, but it has gained new meaning as generative AI spreads across education, entertainment, creative work, search, social media, and productivity software. What once looked like nostalgia now resembles a broader attempt to reclaim agency over time, attention, and personal expression.Michaels’ 2026 Creativity Trend Report offers a useful snapshot of that shift. The craft retailer reported that searches for analog hobbies—including knitting, crocheting, embroidery, painting, and journaling—rose by 136% over six months, while yarn accessory sales increased 40% year over year. Some narrower product categories recorded much larger spikes, illustrating how quickly tactile, handmade activities can move from niche communities into mainstream culture.
These figures should not be treated as a complete measurement of an entire generation. Michaels has an obvious commercial interest in promoting craft trends, and social-media-fueled aesthetics can fade as quickly as they appear. Even so, the pattern extends beyond one retailer: younger consumers are showing renewed interest in physical media, screen-free gatherings, disposable and film photography, traditional crafts, printed books, dedicated music players, and simpler communications devices.
More than a digital detox
A conventional digital detox is temporary. Someone switches off a phone for a weekend, deletes a social app, or takes a holiday without checking work email, then returns to the same digital environment afterward.The analog revival is potentially more durable because it changes the activity itself. Writing in a paper journal is not merely writing without Wi-Fi; it removes notifications, copy-and-paste behavior, infinite revision, automated suggestions, and the pressure to publish the result. Knitting, painting, baking, or developing film similarly makes slowness part of the experience rather than a technical inconvenience to overcome.
That helps explain why the revival is intersecting with AI fatigue. Generative systems promise faster output, but speed is not always the objective. For many people, the process of making something—and knowing that they made it—is precisely where the value lies.
Why Gen Z is central to the trend
Gen Z did not suddenly discover that screens exist. This is the first generation to move through childhood, education, relationships, and early employment under the influence of smartphones, recommendation algorithms, and persistent online identity.That familiarity can produce skepticism rather than unquestioning enthusiasm. Younger users understand how quickly a short check of a social app can become an hour of scrolling, how online aesthetics are manufactured, and how algorithmic systems reward imitation. They also face a digital environment increasingly filled with AI-generated images, text, audio, advertisements, summaries, and comments whose origins are difficult to verify.
The attraction of analog activities is therefore not necessarily backward-looking nostalgia for a world Gen Z never experienced. It can be a search for boundaries that digital products have systematically removed.
From Screen Fatigue to AI Fatigue
Screen fatigue and AI fatigue overlap, but they are not the same phenomenon. Screen fatigue concerns the physical and psychological burden of spending too much time looking at displays, while AI fatigue also involves trust, authorship, control, and the feeling that automation is being inserted into tasks without a clear invitation.Generative AI makes the problem more complicated because it is not confined to one application. It can appear in search results, operating-system interfaces, office suites, messaging tools, photo editors, customer-service portals, development environments, and educational platforms. Even users who appreciate AI for specific tasks may resent being repeatedly prompted to use it.
Automation can remove the satisfying part
Technology companies usually measure automation by time saved. If a feature turns a ten-minute task into a two-minute task, the product team can present an apparently straightforward productivity gain.Human activities do not always fit that calculation. The person keeping a journal may not want a model to rewrite an entry. A photographer may enjoy selecting settings manually. A student may need to struggle with a paragraph to understand the subject, while a programmer may learn more from debugging code than from accepting a generated fix.
When AI removes effort, it can also remove practice, discovery, and ownership. That does not make automation inherently harmful, but it means the user should decide which parts of an activity deserve acceleration.
Synthetic abundance changes the value of human work
Generative AI creates an effectively unlimited supply of competent-looking text and imagery. As the supply of synthetic content rises, handmade and visibly imperfect work can become more valuable because its scarcity and provenance are easier to understand.A knitted scarf may be less uniform than a factory-made equivalent, yet the unevenness becomes evidence of the maker’s time. A handwritten note may be less legible than a printed message, but it carries a stronger sense of intent. A photograph captured on film cannot be reviewed, regenerated, or instantly perfected, making each frame feel consequential.
This preference does not require hostility toward AI. It reflects a changing economy of meaning in which effort itself becomes part of the product.
Microsoft’s AI-First Windows Experiment
Microsoft introduced Copilot+ PCs in May 2024 as a new category of Windows hardware built around local AI processing. The original specification centered on a neural processing unit capable of at least 40 trillion operations per second, alongside minimum memory and storage requirements.The NPU was intended to handle sustained machine-learning workloads more efficiently than a CPU or power-hungry discrete GPU. Microsoft paired the hardware category with Windows experiences such as Recall, Click to Do, enhanced search, Live Captions translation, and Windows Studio Effects.
Technically, the strategy made sense. The PC industry needed a reason to encourage upgrades, chipmakers had begun integrating more capable NPUs, and local inference offered potential advantages in latency, privacy, offline availability, and cloud costs.
The problem was not simply the technology. It was the way Microsoft attempted to make AI the organizing narrative for Windows.
Copilot became both a product and a label
The Copilot name appeared across Windows, Microsoft 365, Edge, GitHub, security tools, enterprise services, and individual applications. Meanwhile, Copilot+ referred to a hardware category whose local Windows features were distinct from the cloud-based chatbot many users associated with the Copilot brand.That naming created avoidable confusion. A customer could buy a computer with a physical Copilot key without necessarily understanding which capabilities ran locally, which required an internet connection, which needed a subscription, and which depended on organizational policies.
The branding also made every AI integration feel like part of one aggressive campaign. Even modest features inherited the controversy surrounding more sensitive capabilities such as Recall.
Recall changed the public conversation
Recall was designed to make past PC activity searchable by periodically capturing representations of what appeared on the screen. Microsoft presented it as a way to recover previously viewed information, but critics immediately raised questions about sensitive data, retention, access, malware, shared devices, and the sheer scope of recording required to make the experience useful.Microsoft subsequently delayed, redesigned, and hardened Recall, adding stronger controls and emphasizing opt-in deployment and Windows Hello authentication. Those changes addressed important technical concerns, but the initial announcement had already shaped public perception.
The episode demonstrated a central problem with ambient AI: even if a feature runs locally, users may object to the computer observing more than is necessary. Local processing can improve privacy, but it does not automatically establish consent or trust.
The 2026 Windows Course Correction
By early 2026, Microsoft’s Windows messaging had begun to change. Windows and Devices chief Pavan Davuluri emphasized performance, reliability, and craftsmanship while acknowledging user frustration with unnecessary Copilot entry points.Microsoft said it would reduce AI integrations where they did not add enough value, initially naming applications and surfaces such as Notepad, Photos, Snipping Tool, and Widgets. The company also promised broader work on responsiveness, memory use, interface consistency, and long-requested customization.
This was not an abandonment of AI. Microsoft continued investing in Windows AI Foundry, local models, agent infrastructure, Copilot services, and AI-capable hardware. The shift was instead about becoming more selective—and admitting, at least indirectly, that ubiquity is not the same as usefulness.
Build 2026 revealed a broader platform strategy
Copilot+ PC branding was less dominant at Build 2026 than it had been during the category’s launch. Microsoft’s technical direction increasingly emphasized Windows as a platform on which developers could run models across different accelerators, rather than presenting one NPU specification as the single gateway to local AI.Windows AI APIs now account for a more heterogeneous hardware landscape. Depending on the model and workload, supported devices can use NPUs, GPUs, or CPUs, with newer NVIDIA and AMD graphics hardware gaining access to capabilities that once appeared closely associated with Copilot+ machines.
That evolution is practical. NPUs remain valuable for low-power, persistent workloads, especially on laptops, but discrete GPUs offer far greater raw throughput for many generative tasks. A desktop with a powerful graphics card should not become artificially ineligible for useful local AI merely because it lacks an NPU matching a marketing specification.
The hardware category is not disappearing
It would be premature to declare Copilot+ PCs dead. Microsoft still promotes the category, PC manufacturers continue to release qualifying systems, and the 40-plus-TOPS NPU remains part of the official consumer specification.Modern Copilot+ laptops also deliver benefits that have little to do with whether a buyer uses generative AI. The latest Qualcomm, Intel, and AMD platforms have pushed Windows notebooks toward longer battery life, better standby behavior, quieter operation, and stronger performance per watt.
The category’s future may depend on whether Microsoft treats the NPU as a useful component rather than the computer’s entire identity. Buyers can appreciate efficient silicon without wanting every application to announce an AI assistant.
Why the PC Is Different From the Smartphone
A smartphone is a general-purpose computer, but its design encourages persistent availability. It travels everywhere, remains connected, receives notifications throughout the day, and provides immediate access to social feeds optimized around continuous engagement.A traditional PC has a different physical and behavioral relationship with its owner. It usually requires a deliberate transition: opening a laptop, sitting at a desk, connecting peripherals, or entering a workspace. That small amount of friction can be beneficial because it marks the beginning and end of a computing session.
The PC as a workbench
The workbench analogy captures what Windows can offer in an analog-oriented lifestyle. A person approaches a workbench to perform a task, arranges the necessary tools, completes the work, and then steps away.A well-configured PC can function similarly. It can be used to write a long document, organize photographs, design a pattern, record music, research a repair, edit video, manage finances, or play a game without remaining the user’s constant companion.
This does not mean PCs lack distractions. Windows contains notifications, advertisements, recommendations, browser prompts, news feeds, chat clients, game launchers, and attention-seeking third-party software. The difference is that these interruptions are not structurally inseparable from the device.
Physical input still matters
Keyboards, mice, pens, drawing tablets, control surfaces, and large displays offer forms of precision that touch-first mobile interfaces often sacrifice. Those peripherals also make the user an active operator rather than a passive recipient of a feed.The keyboard is particularly important. Typing on a physical keyboard encourages sustained composition, while keyboard shortcuts enable experienced users to manipulate files and applications without navigating layers of touch-oriented menus. A stylus can similarly bridge analog and digital workflows for illustration, handwriting, annotation, and design.
The renewed interest in mechanical keyboards is part of the same cultural pattern. Buyers are choosing switches, keycaps, layouts, and sound profiles not because the fastest possible text entry demands such customization, but because tactile feedback makes interaction feel intentional.
Intentional Computing Could Be Windows’ Best Feature
Microsoft has historically presented Windows as the platform that can do almost anything. That flexibility remains a competitive advantage, but it becomes less compelling when the operating system constantly recommends additional services, applications, subscriptions, and AI features.Intentional computing would reverse that relationship. Instead of asking how Windows can occupy more of the user’s day, Microsoft could ask how the PC can help the user accomplish a chosen task with fewer interruptions.
A practical intentional-computing mode
Windows already includes Focus sessions, Do Not Disturb controls, notification settings, virtual desktops, app permissions, and power-management options. The problem is fragmentation: users must configure multiple systems, and application developers can still design around attention capture.A coherent intentional-computing mode could follow a clear sequence:
- The user selects an activity, such as writing, studying, editing, gaming, or reading.
- Windows opens the relevant workspace, including chosen applications, files, browser tabs, and virtual desktops.
- Nonessential notifications are suppressed, while calls or messages from approved contacts remain available.
- Recommendations and promotional content disappear for the duration of the session.
- AI features remain dormant unless summoned, rather than presenting proactive prompts.
- The session ends with a simple summary, after which Windows closes or saves the workspace without trying to extend engagement.
AI should become a tool, not an atmosphere
The strongest role for AI in intentional computing is on demand. A user might request transcription, remove noise from a recording, search a local document collection, translate a video, identify duplicate files, or summarize a meeting.In each case, the user defines the objective and initiates the operation. The system processes a bounded set of data, shows what it is doing, and provides a result that can be accepted, edited, or rejected.
That model differs fundamentally from ambient AI that watches activity continuously, inserts buttons throughout the interface, or predicts that the user might want assistance. The difference is not the underlying model; it is the balance of power between the software and the person operating it.
Consumer Implications
For consumers, the analog revival could strengthen the laptop’s role as a purposeful household device. Smartphones have absorbed casual photography, messaging, navigation, entertainment, banking, and browsing, but their convenience also makes them difficult to put down.A laptop can support analog hobbies without replacing them. It can store scanned journal pages, print sewing templates, catalog a vinyl collection, edit photographs after development, manage recipes, design labels, or provide long-form instruction before being closed and removed from the activity.
Windows can complement physical creativity
The most promising workflows move naturally between physical and digital spaces. A crafter might sketch an idea on paper, refine dimensions in a Windows application, print a template, and then perform the actual construction by hand. A musician might record an analog instrument into a digital audio workstation without using AI to generate the composition.Photography provides another example. Film cameras create scarcity at capture time, but a PC remains useful for scanning, color correction, archival storage, printing, and sharing. The digital tool expands what the user can do after the tactile experience rather than erasing the experience itself.
Windows has an advantage because it supports an enormous range of printers, scanners, cutters, cameras, audio interfaces, drawing tablets, and specialist peripherals. Better device reliability and less intrusive software would make that ecosystem even more attractive.
PC gaming also fits the intentional model
Gaming may not look analog, but a dedicated gaming session differs from endless mobile scrolling. A player chooses a title, enters a structured experience, develops skills, and usually recognizes when the session begins and ends.Windows gaming nevertheless suffers from its own layers of friction: multiple launchers, overlays, account prompts, background updaters, driver utilities, storefront notifications, and performance-hungry services. Microsoft could apply the workbench philosophy here by improving full-screen focus, suppressing irrelevant processes, and making the Xbox app a cleaner organizer rather than another engagement surface.
The lesson is not that every deliberate activity is healthy by definition. It is that products should help users understand and control their attention rather than deliberately dissolving its boundaries.
Enterprise and Education Impact
The argument for intentional computing becomes even stronger in business and education. Employees and students often use PCs because they must complete complex tasks, yet modern software surrounds those tasks with chats, alerts, dashboards, assistant prompts, and collaboration notifications.AI can improve productivity, but uncontrolled integration can increase cognitive load. Workers may spend as much time evaluating generated material, correcting errors, checking sources, and managing prompts as they save through automation.
Enterprises need predictable AI boundaries
Corporate IT departments care about data governance, auditability, identity, application compatibility, support costs, and policy enforcement. They cannot treat every new AI button as a harmless convenience.An enterprise-friendly Windows strategy should provide centralized control over:
- Which AI models can run and where processing occurs.
- Which applications can access organizational data for inference.
- Whether prompts and outputs are retained, logged, or used for training.
- Which users may enable screen-aware or agentic features.
- How administrators disable unsupported AI entry points.
- What happens when a model, provider, or hardware accelerator changes.
Schools face a deeper authorship problem
Educational institutions must decide when AI assistance supports learning and when it substitutes for learning. A generated answer can help explain a difficult concept, but it can also bypass the intellectual struggle through which understanding develops.Windows devices used in classrooms would benefit from clear learning modes that limit generative assistance during assessments, expose source provenance, and allow teachers to define acceptable tools. Such controls would be more useful than simply placing Copilot everywhere and leaving each school to invent policies after deployment.
Analog activities have pedagogical value here as well. Handwriting, physical experiments, drawing, reading printed material, and face-to-face discussion create different cognitive conditions from typing into a chatbot. A Windows PC should integrate with those methods, not assume that every educational experience improves when mediated by AI.
Competitive Implications for the PC Industry
Microsoft is not the only company confronting digital fatigue. Apple, Google, Samsung, Meta, and numerous application developers all compete for attention while simultaneously offering tools intended to manage the resulting overload.The analog revival gives Microsoft an opportunity because Windows does not depend on one vertically integrated consumer ecosystem. Its strength lies in variety: compact laptops, repairable notebooks, gaming towers, professional workstations, tablets, mini PCs, handhelds, and custom-built desktops can all serve different forms of deliberate computing.
Apple has the stronger simplicity narrative
Apple has long marketed the Mac around focus, craftsmanship, battery life, and a controlled relationship between hardware and software. Apple is also investing heavily in AI, but its brand gives it more room to present machine learning as an invisible capability rather than the reason the computer exists.Windows manufacturers cannot rely on a single design language or support model. However, they can compete through hardware diversity, broader peripheral compatibility, gaming, upgradeability, repair options, and price.
Microsoft’s role is to ensure Windows does not undermine those advantages with inconsistent interfaces and unsolicited additions. A beautifully engineered laptop cannot feel calm if the operating system repeatedly advertises services or changes default behavior after updates.
ChromeOS and mobile platforms remain a warning
ChromeOS demonstrated that many users value a simple, low-maintenance computer even when it offers less flexibility than Windows. Tablets similarly attract people who want fewer visible layers of system management.Windows does not need to imitate those platforms by hiding all complexity. It needs to make complexity optional. Novices should receive a clean and stable default experience, while enthusiasts retain access to advanced settings, local accounts, file management, scripting, virtualization, and hardware controls.
That balance would make Windows more compatible with both digital minimalism and the traditional PC enthusiast ethos.
Designing a Quieter Windows
A quieter Windows would not be technologically primitive. It could contain sophisticated security, accessibility, search, synchronization, and local AI systems while presenting them only when they serve a clear purpose.The key is restraint. Every notification, taskbar icon, context-menu item, background service, recommendation, and assistant prompt consumes a small amount of attention. Individually, each addition may appear defensible; collectively, they can make the operating system feel adversarial.
Defaults determine the real experience
Microsoft often points out that users can disable notifications, remove widgets, change privacy settings, uninstall some applications, hide search elements, and turn off recommendations. That defense places too much responsibility on the customer.Most people use default settings. They may not know which switch controls a particular behavior, and Windows settings have moved between interfaces over successive releases. If users must follow lengthy guides to make the operating system quiet, quietness is not truly a supported experience.
Microsoft should make the least intrusive configuration the default and allow users to opt into additional engagement. This would also reduce suspicion around genuinely helpful features because their appearance would reflect an affirmative choice.
Reliability is part of digital well-being
Digital well-being is often discussed only in terms of screen time, but unreliable software creates its own cognitive burden. Failed updates, unexplained restarts, driver conflicts, inconsistent menus, sluggish interfaces, and unwanted default changes force users to think about the operating system when they should be thinking about their work.Microsoft’s renewed focus on performance, reliability, and craftsmanship therefore aligns naturally with intentional computing. A dependable PC fades into the background, while an unreliable one constantly demands attention.
The analog revival is partly attractive because physical tools tend to make their state visible. A notebook does not update itself, change its layout, or require an account. Windows cannot become that simple, but it can become more predictable.
Strengths and Opportunities
Microsoft and its hardware partners have several concrete advantages if they choose to embrace deliberate, user-controlled computing.- Windows can support analog hobbies without absorbing them. Its broad peripheral ecosystem makes it useful for printing, scanning, archiving, audio recording, photography, design, and fabrication.
- The PC already has a natural boundary. Closing a laptop or leaving a desk creates a clearer end to a session than putting a connected smartphone back into a pocket.
- Modern Windows hardware is more efficient. Newer processors and NPUs can improve battery life, thermal behavior, video processing, accessibility, and background workloads even when generative AI is not the primary selling point.
- Local AI can remain available without becoming compulsory. On-device models can provide transcription, translation, search, and image processing while reducing cloud dependency.
- Microsoft can turn focus into a platform capability. A unified session system could coordinate notifications, applications, virtual desktops, power policies, and permitted AI behavior.
- PC makers can differentiate around tactility and repairability. Better keyboards, matte displays, physical privacy controls, replaceable storage, accessible ports, and repair-friendly construction fit the desire for understandable tools.
- Windows can serve creators who value authorship. The operating system supports professional applications that help users edit and refine original work rather than merely generate finished output from a prompt.
- Enterprise management is a competitive advantage. Microsoft can give organizations precise policy control over AI features in ways that consumer-first platforms may struggle to match.
Risks and Concerns
The opportunity is real, but it should not be romanticized. The analog trend contains contradictions, and Microsoft could easily misunderstand or exploit it.- A retail trend is not necessarily a permanent generational shift. Search spikes and social-media aesthetics may reflect temporary fashion rather than lasting changes in technology use.
- Analog hobbies can become content-production pipelines. A person may start journaling or knitting for relaxation, then feel pressure to photograph, post, monetize, and optimize the activity for social media.
- Microsoft may convert “focus” into another subscription. Packaging basic distraction controls as a premium AI service would undermine the entire concept.
- Reduced AI promotion does not guarantee reduced data collection. Users need clear disclosures and enforceable controls, not merely fewer visible Copilot buttons.
- Local AI can still create security risks. Screen-aware tools, autonomous agents, and searchable activity histories require strong isolation, authentication, retention rules, and malware resistance.
- Hardware fragmentation may confuse buyers. NPU, GPU, CPU, memory, model compatibility, and Copilot+ requirements remain difficult to explain consistently.
- Aggressive simplification could alienate enthusiasts. A calmer default should not remove advanced controls, local software support, customization, or access to the underlying file system.
- The PC industry may continue selling AI before proving its value. Stickers and TOPS figures are poor substitutes for reliable applications that solve recognizable problems.
What to Watch Next
The next phase will reveal whether Microsoft’s 2026 course correction changes Windows at a fundamental level or merely reduces the most unpopular Copilot placements. Users should watch both what Microsoft adds and what it chooses not to add.Windows preview builds
Preview releases will show whether promises about performance, memory use, taskbar flexibility, interface consistency, and reduced AI clutter produce measurable improvements. The crucial test is whether those changes survive broad deployment without being offset by new recommendations or promotional surfaces elsewhere.Microsoft’s treatment of defaults will be especially important. A feature that can technically be disabled but arrives enabled after an update still represents an attention-first design decision.
The evolution of Copilot+ requirements
Microsoft continues to define Copilot+ PCs around capable NPUs, but Windows AI tooling is expanding toward GPUs and, for some workloads, CPUs. That could turn Copilot+ from a rigid dividing line into one tier within a broader local-AI ecosystem.NPUs will retain an efficiency advantage for continuous, low-power processing. GPUs will remain preferable for heavier models and creative workloads. Windows needs a transparent scheduler and compatibility framework that selects the appropriate accelerator without forcing customers to understand every architectural detail.
PC design beyond the AI sticker
Manufacturers may begin emphasizing features that users can immediately appreciate: battery longevity, quiet cooling, excellent keyboards, readable displays, repairability, privacy shutters, durable hinges, accessible ports, and predictable sleep behavior.Those qualities fit the analog revival better than a dedicated AI key. They make the machine feel like a reliable object rather than a constantly evolving service endpoint.
Evidence of lasting behavioral change
Retail reports provide clues, but stronger evidence will come from multi-year purchasing patterns and product usage. Growth in physical media, dedicated cameras, craft supplies, printed books, simpler phones, local-first software, and distraction-blocking tools would suggest that intentional technology use is becoming more than a passing style.The most revealing signal may be how younger buyers describe their ideal computer. If they prioritize control, longevity, privacy, repairability, and focused creation over constant assistance, Microsoft and PC makers will have to reconsider what “modern” computing actually means.
Gen Z’s analog turn does not require Windows to imitate a notebook, a film camera, or a record player. It asks Microsoft to respect the qualities that make those objects appealing: clear purpose, visible boundaries, predictable behavior, personal ownership, and freedom from unsolicited intervention. The Windows PC can still run advanced local models, cloud services, professional applications, and demanding games, but its most valuable feature may be the ability to wait quietly until someone chooses to use it. If Microsoft can make Windows feel less like an engagement engine and more like a dependable tool, AI fatigue may not push younger users away from the PC at all—it may remind them why the PC remains worth keeping.
References
- Primary source: Windows Central
Published: 2026-07-21T16:30:43+00:00
Analog hobbies are booming as AI burnout hits Gen Z hardest — and the Windows PC could be the unexpected winner | Windows Central
PCs aren't phones, and that's a win for Windows. Will Microsoft lean into the trend?www.windowscentral.com