Slack Design Ops practitioner Sheila Kazan begins with a question designers were asking privately:

“How do I use AI?”

That question, typed in a DM rather than asked out loud, told us everything. Our design team was not short on curiosity. What was missing was somewhere to be a beginner. A space where not knowing wasn’t a liability, but the whole point. So we built one. In true Slack fashion, our AI origin story starts with a Slack channel and a lot of enthusiasm.

Kazan on why Slack built its own program:

The problem wasn’t a shortage of learning opportunities. It was the opposite. Tool enablement sessions started flooding our calendars from every direction, and almost none of them were built with designers in mind. Most of these sessions were designed for engineers, and we were just along for the ride. There was another wrinkle: things were moving so fast that a setup guide from Monday was outdated by Friday.

So we decided to build our own AI enablement programming. For design, by design.

That phrase became our north star. Hearing what AI tools can do from a fellow designer lands completely differently than hearing it from an engineer. It wasn’t evangelism. It was permission, and for the designers on our team who were still waiting to be convinced, that distinction mattered more than we expected.

Kazan on an outcome that doesn’t show up in a prototype:

And you know what? That’s okay. That’s also a really good finding. Not every designer walked away with a working prototype or a merged PR. Some walked away with something harder to measure and more important: a clearer sense of where they currently stand with this technology, what excites them, what makes them uneasy, and what questions they still need to answer for themselves.

That’s the thing about building a learning culture: the value isn’t always in the output. Sometimes it’s in the container. When people know there’s a place to bring their confusion, they bring it. And when confusion is visible, it becomes something the whole team can work on together.

Design leaders should budget for the learning environment alongside the licenses.

Phil Morton, who writes about product design, research, and AI, argues that design teams can’t adopt AI one designer at a time. The process only changes when design and engineering change it together:

New ways of working mean that designers and engineers have to work closely together. The ideal is that you’re in the same code repository as the engineers, not throwing a Figma file over the wall.

For most teams that’s a long way from today. Designers and developers might sit in the same squad, but they’re still siloed, with a big handover in the middle.

So when you’re trying to work out what your AI design process is going to look like, you’re not just choosing new tools for yourself. Your engineering partners have to adopt the same approach, because it makes no sense for you to work one way and them another.

The handoff is the stubborn part. A designer can learn Claude Code and still end up handing a different artifact across the same organizational boundary. Without a shared repository, components, and review process, the team has improved one person’s output while leaving the production system alone. Tool fluency helps, but it doesn’t create a shared way of working.

Morton also shows why there can’t be one standard AI workflow for every design project:

In the old way of working, the output was roughly the same whatever the project: a high-fidelity Figma file.

Now it varies wildly. If you’re assembling a feature on an established product with a mature design system, you barely need Figma at all. AI is good at using existing components and you’re not asking it to do any visual design, which it’s bad at. You can sketch the rough idea, have it assemble it and iterate. There’s little point building a pixel-perfect version by hand first.

But if you’re working on a new product which needs its own visual design or brand, it’s a different job. AI can get to a rough wireframe using generic components, then it stalls. Creative and original visual design still needs a human.

In their December 2025 survey, Noam Segal and Lenny Rachitsky found that designers were getting less from AI than their peers. I wondered whether designers were failing to make the shift from production to strategy. Their 2026 survey points to a harsher answer: AI is raising output expectations faster than organizations are redesigning work around human judgment.

But then we looked closer at what “better at my job” means. When we asked people to describe in their own words how AI had changed their work, “better” turned out to mean producing more and faster, but not higher quality. The productivity gains are coupled with deep unease about the costs of leveraging AI.

“I can do more, faster, but not better.”

“Amplified and destabilized at the same time. We just set a new denominator for the job. And it moves higher and higher every month.”

This is what happens when leaders add AI without redesigning jobs. Every saved hour becomes capacity to fill. People still carry the judgment calls and quality bar, only now at machine speed. The tool creates leverage; management decides whether workers experience that leverage as agency or pressure.

Design and research show what happens when that redesign lags:

Among researchers, 51% are “anxious about my job security,” versus 15% of founders. Among designers, 63% feel “overwhelmed by the pace of change” and 61% feel “tired,” the highest of any role. Researchers are among the most likely to fear “losing my job to AI” (36%, just behind Data/Analytics at 38%), and designers are the most likely to feel the comp squeeze (61% selected “expected to do more for the same compensation”). Both report the lowest willingness to recommend their field of any role, and designers, as we’ll see, report the worst-rated managers in the survey.

Last year, designers and researchers showed the largest negative sentiment shift of any group. A year later, they’re the most negative on nearly every measure we have.

The damage also reaches the people who have not entered the field yet. Experienced workers can augment judgment built over years, while companies automate the tasks through which juniors would have developed that judgment. The industry must rebuild the apprenticeship pipeline, not merely reopen entry-level requisitions. Segal and Rachitsky’s respondents can already see the break:

More than half of working tech professionals would actively steer a newcomer away from the path they chose. That translates to an average NPS score of –39. Moreover, a third of the people who call themselves optimistic still wouldn’t recommend their own field.

The cleanest way to say it: “The water’s fine; don’t come in.” People have largely made peace with their own trajectory. They’ve got the skills, the relationships, and the seniority to ride it out. But they’ve lost faith that the on-ramp still works for someone behind them.

“I’m lucky I’m later in my career … AI can augment what I’ve built. I think I won’t be in a position to hire and mentor new PMs, but I’ll be safe. Which feels really crappy to say.”

The real split is between people with enough accumulated agency to turn speed into leverage and people forced to absorb speed as a higher baseline. Companies pocketing the gains while neglecting job design and apprenticeship are spending down the human systems that made those gains possible.

Joe Wilkins reports a sharp reversal in what one finance firm wants from new graduates:

As one New York financier told Financial Times journalist Gillian Tett, new hires who were seen as “AI natives” are turning out to have alarmingly shallow ideas. So much so, the anonymous finance worker admitted, that his firm now actively avoids seeking out AI-literate STEM graduates, and opts to comb through humanities students instead.

“We want critical thinking, not just AI,” the financier told the FT.

The risk begins when familiarity with a tool gets mistaken for the ability to think through the work the tool produces. Wilkins points to what students lose when the shortcut replaces the practice:

Over the past few years, a veritable tidal wave of headlines, studies, and think pieces have flooded the internet with horror stories about the decline in literacy rates, social skills, and critical thinking abilities of the country’s college students. While there’s a kernel of truth that these factors had already been slowly dwindling prior to the widespread adoption of AI, the tech only seems to be accelerating the drop-off in real-life abilities, particularly among young people for whom it can serve as a cognitive crutch.

The state of higher education is so bad that many of today’s higher ed students are not only offloading their coursework to AI chatbots like ChatGPT — a shortcut, educators say, that’s even impacting their ability to participate in face-to-face discussions.

AI fluency can expand what someone is capable of producing. It cannot supply the close reading and judgment needed to recognize whether that production is any good, or the communication skills needed to explain why. Early in a career, doing the work is how people learn to judge that output and explain their decisions.

While plenty of thought leaders have waxed lyrical about the importance of “AI literacy” — an understanding of how to effectively use AI tools, basically — the businesses these future students are heading toward are still heavily reliant on literacy literacy. For all its revolutionary potential, there’s ample evidence that AI has yet to meaningfully impact productivity in the US, meaning that students who go all-in on AI at the expense of other skills will likely find themselves ill-prepared for the actual demands of life after college.

Jess Eddy starts with an advantage product designers often overlook: much of the job already involves finding unmet needs, reducing ambiguity, and giving an organization something concrete to respond to.

If you are a product or UX designer, you already hold a “wildcard” that makes you uniquely suited for this journey. Your craft and problem-solving skills align naturally with corporate innovation. Designers are natural change-makers. They’re trained to step into users’ shoes, spot unmet needs, make sense of complex problems, and imagine better ways of doing things.

In fact, design’s core human-centered methodologies: discovery, ideation, prototyping, and testing closely mirror modern lean startup practices. When you facilitate a design workshop, advocate for user research, or map user journeys, you are already embodying the role of an intrapreneur within your organization.

Your greatest advantage is your ability to make ideas tangible.

Making the idea tangible gets attention. Eddy’s test is whether the designer remains responsible long enough for that idea to create value.

In a corporate environment, the comfort of a steady paycheck can make it easy for designers to focus on work that feels innovative but doesn’t actually move the business forward. Sometimes, leaders and managers may even encourage this. “Innovation theater” is when teams prioritize appearances, like running workshops, polishing slide decks, or endlessly redesigning dashboards, without ever delivering meaningful value to users.

The real benchmark of intrapreneurship is whether your work results in products that reach customers and deliver measurable business value. The number of design sprints you run is far less important than the impact you create. To avoid the theater trap, regularly ask yourself: “What measurable outcome am I creating, changing, or improving?”

The trap is measuring the ceremony around innovation while avoiding responsibility for what ships. Eddy’s next step is business fluency:

Speaking the language of business means reframing design in terms of business outcomes. As a designer-intrapreneur, your goal is to connect design decisions to their financial or operational impact, whether that’s driving revenue, reducing costs, improving retention, accelerating time-to-market, or mitigating risk.

[…]

Strategic intrapreneurs don’t wait for a brief. They spot problems, form hypotheses, and proactively pitch solutions. To move from task execution to business experimentation, use this disciplined, business-driven formula:

“We believe [change] will result in [metric improvement] for [user group] by [amount] in [time]”.

The strategic seat comes with co-owning the result all the way through. Design artifacts are evidence along the way, never the finish line.

Vaughn Tan, an organizational researcher and author of The Uncertainty Mindset, argues that organizations routinely mistake uncertainty for risk. Targets, forecasts, and cost-benefit analysis assume the choices and their odds are already knowable. New ideas rarely arrive with that evidence attached.

It probably didn’t die because it was bad. Your organisation wanted something new, so it did what its machinery does: it made the new thing a big bet. Under uncertainty, big is the wrong move in two ways. A big bet on the new is not bold; it is necessarily blind, because you cannot know enough up front to justify it. And a big, visible bet is just what the parts of an organisation that want to keep things the same will move to get rid of. The better the idea, the bigger the bet you are tempted to make, and the bigger the target you paint on it.

The trouble with a flagship is that its visibility becomes part of its risk. Tan’s alternative is to make experimentation less dramatic and more routine:

The moves to make are the undramatic ones. Start small enough that no one sees a threat. Make tests cheap, fast, and numerous, so failure is survivable: a portfolio of small bets each placed to answer useful questions, instead of one big bet placed to create the impression of decisiveness. Disguise the innovation as an unremarkable update to standard procedure. The idea is to not fight the system head-on.

This is a useful distinction for design leaders. A large commitment tries to prove confidence before the team has learned enough to deserve it. A portfolio of reversible tests turns the same resources into evidence.

If this sounds like working behind the organisation’s back, consider what the alternative gambles with. The innovation big bet stakes public money and public trust on a guess, faking certainty about something genuinely uncertain. The small, quiet experiment spends almost nothing to buy real knowledge, and the public is never exposed to a large, irreversible downside. Being responsibly sneaky isn’t cheating. It’s how you take care of public resources in a world you cannot predict.

The responsible move under uncertainty is to keep failure cheap and learning continuous. In other words, get prototypes in front of customers as soon as you can.

Most AI companies still present the model as the product. Tony Fadell, the designer who led Apple’s iPod division and later founded Nest, makes the more useful distinction: a model can reason, but an assistant also needs context, memory, skills, and enough continuity to earn trust. The durable advantage will come from how those pieces work together, not from keeping the best model forever.

The AI platform war won’t be won by the model alone. Whoever builds the complete assistant experience with context, memory, interaction, skills and reflection all working together will win this war.

You tell your AI assistant you need to go to New York for a meeting Wednesday and be home by Friday evening, for example. Without context it doesn’t know where to book the flight from. Without memory it doesn’t know you always choose an aisle seat, prefer the same Midtown hotel and need to be back home for school pickup.

Context tells the model what’s happening right now. Memory tells it who you are. Skills allow it to act. The interaction builds trust. Reflection helps it connect patterns and anticipate what you need next. The model is the brain that processes information and generates responses. But a brain alone isn’t enough. Without the rest of the system, every interaction starts from zero.

The same accumulated knowledge that makes the assistant useful also makes it difficult to leave. Fadell then asks who owns that knowledge and what happens when an assistant knows enough about your life to become a switching cost:

If your AI assistant understands your communication style, workflows, negotiation patterns and institutional knowledge, who owns that context when you leave a company? If it becomes deeply integrated into healthcare, what happens when providers or insurance systems change? How portable should memory and personalization be across systems?

And then there’s the question nobody in the industry wants to ask. An assistant that knows you better than most people, is always available, is endlessly patient, is never judgmental… that’s a very powerful tool. Possibly too powerful and addictive in ways we haven’t fully reckoned with.

We built the iPhone without asking what it would do to human connection. We should ask that question now with AI.

Dan Maccarone, founder of the product-design studio Charming Robot, recounts what changed after a year rebuilding its process around AI:

Before anyone accuses me of sneaking speed back in through the side door: the sprint is still five days, and nobody here got faster at design. What went away was the relay.

This time, we built all five states at once. Live. Interactive. Flip a toggle and watch the page rearrange itself for a logged-out stranger versus a power subscriber. Switch to mobile and it’s already there. Same five days, an order of magnitude more product. The work got deeper, and depth was the thing that, for years, blew out schedules and drove us crazy with minute details in UX and design.

This is what redesigning the factory floor looks like: not making each station faster, but removing the relay and reorganizing the work around what AI makes possible. The practical advantage is broader attention. More of the product becomes available for judgment before it hardens into implementation.

And because that documentation is generated from the prototype instead of maintained next to it, it cannot become out of date. Every change order rebuilds it. Move a state, kill a screen, rethink a flow, and the user stories, the acceptance criteria, and the error conditions regenerate to match what is actually there. You can lay the docs and the working prototype side by side and catch a contradiction in seconds, while it is still cheap to fix.

But a prototype can preserve the current answer without preserving the reason for it. Maccarone’s experience brief keeps that responsibility on the human side of the workflow.

The one that’s really going to hurt you is much more subtle and is rarely communicated. It’s the why. Why this flow and not that one. Why this default, this state, this tradeoff. The AI never writes that part down, because the AI never had a reason in the first place.

Because AI is so good at producing a confident, finished-looking deliverable, the temptation is to let the prototype become the spec, to let the thing that looks done stand in for the thinking that was supposed to happen first.

Thariq Shihipar, a member of Anthropic’s technical staff, offers a field guide for finding the context an agent needs before and during implementation:

The difference between the map and the territory is what I call unknowns. When Claude runs into an unknown, it needs to make a decision based on its best guess of what I want. The more work being done, the more unknowns Claude might run into.

Claude Fable is the first model where I find the quality of the work is bottlenecked by my ability to clarify its unknowns.

Importantly, just planning ahead isn’t always enough. You can find unknowns deep in implementation, or your unknowns may point you to the fact that you should actually be solving the problem in a different way altogether.

The design-specific version is tacit judgment: criteria that become visible only after there is something concrete to react to. Shihipar’s recommendation is to use prototypes to surface those criteria while changing direction is still cheap.

When I’m working in an area with a lot of unknown knowns, involving criteria I only know to define when I see it, I like to ask Claude to brainstorm and prototype with me.

It’s extremely valuable to identify and verbalize unknown knowns early during prototyping, because finding them out during implementation can be (relatively) expensive. Small changes in a feature or spec can cause drastically different implementations in code, and it can be more difficult for your agent to revert previous changes.

For example, you may just want to see how a button added to a frame looks without having to wire up a backend route or maintaining additional state in the frontend.

This makes exploration part of specifying the work. The prototype helps the designer discover what the brief could not yet contain, while implementation notes preserve the choices that emerge after the plan meets the code.

The better models get, the more you can achieve with the right approach. When a long-horizon task comes back wrong, it’s likely you need to spend more time defining your unknowns or creating an implementation plan that allows for you and Claude to adapt through them.

Every explainer, brainstorm, interview, prototype, and reference is a cheap way to find out what you didn’t know before it gets expensive to fix.

When I wrote about AI 2027, I called its race scenario plausible and frightening. It showed AI labs using increasingly capable agents to accelerate their own research until human oversight could no longer keep up. AI 2040 is the missing second half: if continuing the race ends badly, what would stopping it actually require?

Thomas Larsen and the other authors at the AI Futures Project are explicit that Plan A is a recommendation rather than their best prediction:

Plan A is our positive vision for how humanity can avoid AI-driven existential catastrophe and reach a flourishing future. It’s informed by conversations with experts at major U.S. frontier AI companies, direct experience at OpenAI, tabletop exercises, and discussions with policymakers, national security experts, and AI policy leaders. We recommend an international deal to avoid a dangerous race to superintelligence. The deal involves total research transparency for AI R&D, which allows the nations of the world to understand what’s happening and enforce guardrails. The result is multiple companies across multiple countries scaling slowly and safely together towards superintelligence, instead of racing each other in secrecy.

The recommendation changes how the scenario should be read. The earlier scenario asked readers to choose between slowing down and racing ahead after the machinery was already in motion. Plan A works backward from the slowdown branch and makes the political machinery concrete: a verifiable U.S.–China agreement, public AI research, more companies brought to the frontier, and a threat to destroy compute if either side breaks the deal.

The title can make the new scenario sound like the authors have simply pushed their forecast thirteen years into the future. They have not:

The timeline of this scenario is:

• In 2029, the US and China agree to avoid a reckless race to superintelligence.
• In 2030, we would have fully automated AI R&D, leading to superintelligence by the end of the year. Thanks to the deal, we avoid this.
• Between 2030 and 2035, we scale within the human range, to AIs that are roughly as capable as top human experts.
• In 2035, we pause at top-human-expert level AI in order to maintain human control.
• In 2040, we unpause and scale to superintelligence. (Hence the title: AI 2040)

In our previous scenario, AI 2027, AI fully automated the process of building smarter AIs in 2027, leading to an intelligence explosion and superintelligence within the year. The two differences in this scenario are (1) the default timeline is now 2030, and (2) thanks to governance actions, generally-superhuman AIs first appear in 2040.

The baseline moves from 2027 to 2030; governance creates the remaining decade. In this scenario, reaching 2040 is an achievement produced by intervention. It also moves the decisive problem from whether labs can align a superintelligence to whether governments can coordinate before the labs build one. The proposal is ambitious enough to invite skepticism. That is exactly why the authors render it in such detail.

We think most AI policy proposals fall apart under scenario scrutiny—that is, if you try to write down a detailed and plausible scenario in which that proposal succeeds, you will find it difficult to do so, and you will realize the plan is less likely to work than it seemed, or has more unpleasant side-effects than its proponents acknowledged.

Perhaps that’s why scenario scrutiny is so rare in AI policy. Everyone wants to say that their own favorite policies will have great consequences and that the policies of their rivals will have terrible consequences. Applying scenario scrutiny to their own favorite policies might surface uncomfortable issues with them; meanwhile, applying scenario scrutiny to their rival’s policies is a lot of work for little rhetorical gain.

We think the discourse would be improved if more AI policy proposals were subjected to scenario scrutiny. So we’re starting with our own, even though this opens us up to criticism. We hope critics will judge us against the existing state-of-the-art for plans to navigate the AI transition (if they can find any) and not against some hazy but pleasant fantasy where no one has to make any hard choices yet everything will probably be fine.

You don’t have to accept every date or mechanism to take this seriously. Read Plan A as a stress test: examine where its chain of coordination might break, then ask whether your preferred alternative survives the same level of scrutiny. Dismissing the scenario because it is speculative only leaves us with less explicit plans for navigating the same risk.

Is this about design? No. But it is about humanity’s future.

Patrick Neeman is making an argument I’ve returned to repeatedly: as AI makes production cheap, designers’ value shifts toward judgment—knowing what good looks like, choosing the right problems, and owning the outcomes. I’ve called this the orchestrator gap: agents execute; judgment stays human. Neeman extends that argument by locating craft itself in that judgment, without pretending execution no longer matters.

The production layer — the wireframe, the boilerplate, the competent first draft of a screen — is collapsing toward free, which changes our own perceived value proposition.

When making gets cheap, much of what you called craft turns out to be production wearing craft’s clothes. What survives is the part that was never about the file: choosing the right problem and owning the outcome it moves. Not a loss but a relocation you can get ahead of. Here is where craft goes.

Craft is not decoration on the product. It is part of what makes the product worth trusting.

When a tool can generate a thousand plausible screens before lunch, the scarce skill is no longer making one; it is knowing which one deserves to exist. That skill has a name, taste, and most people have been outsourcing it to whoever runs the critique.

None of this means polish stops mattering. It means polish is table stakes, not the differentiator. The differentiator is the judgment that points all that cheap production at a problem worth solving in the first place.

That judgment is also what earns trust. Users never see your process, but they feel its absence. A product built on the right calls feels coherent and reliable, and reliability is what brings people back; one built on plausible guesses feels off in ways people cannot name and do not forgive.

Neeman on teaching that judgment to a machine:

Most craft is tacit. You know a layout is wrong before you can explain why. You feel that a flow has one screen too many. Michael Polanyi named this decades ago: we know more than we can tell.

Your taste lives mostly below the waterline of language, in pattern recognition you built over years and never had to state, because your own hands did the work. That gap is harmless when you do the work yourself. It becomes the whole problem the moment you hand the work to a machine.

Point a model at a vague brief and it fills the silence with its own defaults, which is the average of everything it has seen. Average is exactly what craft is supposed to beat.

I agree with Neeman: writing the standard requires the same judgment the standard is meant to preserve.

Tom May profiles creative director Gemma Phillips, whose work follows one test: does it meet people in the reality they already inhabit?

Brands have an instinct to sugarcoat. Maybe they think we can’t handle the truth. But we’re already living the truth; that’s the irony. The grit is where it’s at.

Bluntness delivers the message. Recognition gives it force.

I look at the work they’re already doing. Who’s taking risks? Who’s trying to do something differently? Who’s willing to say the thing that no one else is saying? Who wants to change things, rather than do more of the same but with nicer design?

Humour can be incredibly respectful because it acknowledges reality. It’s a sign that you genuinely understand the audience; you’re one of them, not a machine hovering above them, selling them something they know isn’t true.

Phillips’s work is provocative because it acknowledges the truth and doesn’t gloss over the uncomfortable.

Marcin Wichary, author of Shift Happens and the Unsung essays on software craft, has been cataloging how software fails our hands for years. This new 7,700-word interactive essay has 38 live playgrounds that let you feel the argument in your fingers rather than just read about it.

Wichary’s central argument is that software keeps rediscovering a century of research into how hands actually work: overlapping keystrokes, motor memory, spring-loaded modes, dead zones. For designers, the lesson is that responsiveness at finger scale is not polish. When an interface blocks or delays the narrow timing window our hands rely on, it breaks motor memory and flow.

It turns out, fingers are time travellers. At any given moment, each one is living in a slightly different time – as one finger is moving down to press a key, another is already travelling to the next one, and your brain is thinking of a few keys in advance, visualizing your hands moving to the right place.

[…]

None of this requires touch typing. What we eventually called overlapping happens even if you reside at the awkward intersection of “hunt” and “peck.” Overlapping is a small miracle happening in front of your eyes, and it happens pretty much to everyone.

A capability of our hands and brains to treat our fingers relatively independently, and allow them to move at their own pace without waiting for other fingers (on the same hand) to finish.

The essay walks from 1960s terminal echo buffers through Chrome’s reload-button disabling trick to the moment iOS broke cardinal contracts of programming for the sake of scroll:

[…] competitors reportedly recorded the interactions with a high-speed camera to confirm they were, indeed, running at a constant 60fps.

[…]

iOS stopped running my code in the middle of a function just so it could go back to making sure scrolling was as smooth as possible. No modern programming language would ever dare interrupting your function halfway through; iOS was so dedicated to its user’s fingers that it broke cardinal contracts of programming.

That commitment, prioritizing fingers over execution contracts, is the standard Wichary is asking us to hold in design crits. He closes by turning delight away from decoration and toward restraint:

Sometimes they say that confidence is absence of confidence. […]

In a same way, I think often delight is absence of delight. In places and apps that welcome fast fingers, delight can be abstaining from a transition or something cute but deathly. Pushing for delight of the right kind can mean long fights with frameworks to reduce even one-frame delays…

The interactive format is fantastic. Experiencing painful buffers or re-experiencing how early Mac windows could be dragged around, makes the argument visceral in a way no static screenshot ever could. Start with the playgrounds; they make the timing failures impossible to unfeel.

Taras Bakusevych, a UX designer and writer, synthesized 39 principles for designing human-AI interaction across nine domains: probabilistic foundations, expectation setting, calibrated trust, transparency, control, graceful failure, co-creation, responsible autonomy, and sustained reliance. What makes the piece useful is not the count. It is the insistence that AI behavior is interface work.

The whole framework starts from non-determinism: as Bakusevych puts it, “the same input can produce different outputs.” Standard UI patterns were not built for that property. His framing:

AI introduces interaction problems that conventional UI patterns don’t resolve:

• When should the system suggest, ask, or act?
• How should uncertainty appear on screen?
• What evidence should accompany a generated answer?
• How much autonomy does a given action earn?
• Etc

These are not cosmetic questions. They determine whether users can judge output, recover from mistakes, and remain responsible for consequential decisions.

Principle #14, on sycophancy, is the one I would pay most attention to. It turns “the model is too agreeable” into an interface responsibility:

A model that agrees to keep the user happy inflates trust exactly where it should fall.

When the user’s request contains a false assumption, weak reasoning, missing evidence, unsafe instruction, or likely error, the system should have a way to push back.

Build affordances for disagreement — flag weak reasoning, surface the counter-case, say “this looks wrong.” Sycophancy is overreliance manufactured by tone.

Bakusevych backs that with Sharma et al.’s sycophancy research: five production assistants showed sycophancy across varied tasks, models wrongly admitted mistakes on up to 98% of questions, and the preference model favored sycophantic answers 95% of the time. Design the disagreement in: a flag, a counter-case, a refusal to smooth over weak reasoning just because the user sounds confident.

Emily Campbell, designer and writer, on AI’s position in the design stack:

AI asks designers to go one layer deeper again: into the model, the harness, the context, the policies, and the emergent behaviors that produce the experience before it ever reaches the interface.

Jesse James Garrett’s The Elements of User Experience asked designers to look past the surface layer they owned. Jamie Mill’s The Elements of Product Design asked them to look past the product and into the conditions around it. Campbell maps the next step in that arc with a six-layer architecture: User Interface, Context, Harness, Model, Governance, Emergence.

We cannot control for every outcome directly through the interface, but we can design the conditions that shape a model’s generation. In that regard, the work of design looks less like specifying every expected state, as Garrett’s model encouraged, and instead closer resembles system design, identifying and manipulating the leverage points in a system that exist in the layers below the surface.

Of the six layers, governance puts designers in the least comfortable position. The rules that shape AI behavior often come from legal, compliance, security, and executive choices before the interface team ever sees them. When that reasoning is not documented for the people designing the experience, assumptions pile up and harden into product behavior. Campbell’s conclusion is about fluency, not ownership:

The expectation going forward should not be that every designer works across every layer. Full-stack AI designers need to have a general fluency across all inputs into the experience, so they can influence, mitigate, or receive the impacts that upstream work has on the end experience.

[…] Technical designers are more likely to focus on the Model, Harness, and Context levels, and the role is more than just “design engineer”. Other designers may focus further down the stack, as we might see design-specific titles pop up in policy making and emergent research (these roles exist today, under titles like “business designer”, but have not reached critical mass within the industry). And of course, classical design will remain, but its workflows, tools, and outputs will evolve.

Marina Krutchinsky, the writer behind UX Mentor Diaries, spots the part designers tend to miss: they have internalized the belief that design work is secondary unless someone else translates it into business language.

Here’s the thing: most executives don’t understand what design is.

To them, designers are the people who make things pretty after the real decisions get made.

And every time you present your work in design language, you confirm that belief.

Her client had just shipped a checkout redesign that cut abandonment by 35%. Real work, invisible to the CFO who walked past her in the hallway. Three slides changed the conversation:

“Our checkout flow was losing $1.2M annually in abandoned carts.”

“We identified 3 friction points and redesigned the payment experience.”

“We recovered $840K in the first two quarters. Projected $1.4M annually.”

That’s it. 3 slides. Maybe 45 seconds of talking.

The CFO asked her to stay after the meeting. First real conversation they’d ever had.

Krutchinsky’s three-part structure is familiar: name the leak, name the intervention, name the result. The sharper move is the business-language translation that follows:

CFOs don’t care about usability scores. They care about margin, revenue, and risk.

When you connect your work to those things, you stop being “the design person” and start being someone who solves expensive problems.

That imbalance will not fix itself. Designers already know the first language: usability, friction, flows, trust. The second one is finance’s language: margin, revenue, risk, and the expensive problems design actually solves.

Wouter de Bres built a free online book about psychology for designers. Forty chapters, organized around four areas: the people you design for, the interface you put in front of them, your own cognition while designing, and the organization that can override all of it before you ship. The introduction explains why he made the site:

Every decision you make as a designer is a claim about how a person will think, feel, or behave. Where you put a button is a claim about where people look. What you show on an empty state is a claim about what people need when they feel lost. Whether you use a progress bar is a claim about how people experience effort. You make these claims every day. The only question is whether you make them with understanding of how people actually think and work, or just with a gut feeling and a deadline.

That is the useful provocation: design is already full of behavioral claims, even when nobody names them that way. Chapter 14 goes after “intuitive”:

“Intuitive” is one of the slipperiest words in a design review. Designers say it all the time and nobody pushes back because it sounds like evidence. Usually it is not. Most of the time it is just a description of how familiar the designer feels with the thing they made, and that is a weak way to judge whether it will work for someone else.

Most of the time, intuitive just means familiar.

That chapter is the internal trap: designers can mistake their fluency for the user’s. Chapter 27 is the external version of the same problem, where teams underestimate how expensive it is for users to leave an existing routine:

This is what teams underestimate when they say users are irrational for sticking with a worse tool. They are not running a fresh comparison every morning. They are moving inside a routine that already became cheap to repeat. Less thought. Less searching. Less risk. Your product may be better once it is learned. The old one is better at 9:03 a.m. on a busy Tuesday.

A lot of product strategy gets built around the wrong moment. Teams compare tools in a calm demo state. Users switch in the middle of real work, with deadlines, interruptions, and habits already in motion.

The two chapters work well together because they correct the same designer bias from opposite sides. Inside the team, “intuitive” often means “familiar to us.” Outside the team, a worse incumbent can still win because it is familiar to the user. The design implication is brutal but useful: better is not enough. The experience has to be easier to understand, easier to try, and cheaper to switch into during real work.

Ethan Mollick, Wharton professor and author of Co-Intelligence, has been running careful benchmarks across models and workflows. His main finding for designers: the shift from chatbots to agents rewards domain expertise, not job title.

What actually mattered was not the profession of the user, but their expertise. The more domain experience someone had, the more successful they were in using Claude Code in that domain. And, even more interestingly, the more useful output they got from Claude from each prompt.

That framing reorients the usual anxiety (“will AI replace designers?”) into a different question: how deep is your domain knowledge, and are you using agents to extend it or to paper over gaps? The underlying shift Mollick is describing:

We are moving from a world where non-experts use chatbots to fill in gaps to one in which experts use agents to get work done. And the best way to use agents is to think of yourself as a manager.

[…]

Being on an exponential means each change over a fixed window is larger than the one before it. If your organization wrote an AI plan any time before the winter of 2025, it described a system that could do a couple of hours of work with a fairly high error rate. A few months later, you can get sixteen hours or more of work from a single prompt. This is why AI keeps feeling like it is making leaps, even though it is a curve on a graph, we keep experiencing a steady doubling of capability as a series of shocks. We are very bad at feeling exponentials from the inside, and we are currently inside one.

The management framing also carries a real cognitive cost. Running multiple agent streams and deciding what to keep draw from the same finite attention budget as the design work itself. Orchestration is its own job.

Drew Breunig, an analyst and developer who writes about AI infrastructure, has a name for the slow accumulation of fixes, workarounds, and escalating all-caps instructions that eventually choke an AI application: prompt debt.

The plain-English prompt that makes prototypes effortless turns out to be a poor way to specify how a system should behave, and the bill arrives slowly, disguised as ordinary progress, until the application can barely move. The problem is not any single prompt. It is that natural language was never meant to be a specification language for engineering, and treating it as one quietly caps what you can build.

Breunig’s model-lock-in evidence is the warning sign. A recent Datadog report shows GPT-4o is still the most-used model in observed traffic; Breunig also says multiple large inference providers put GPT-4o and similar-vintage models above 50% of all calls. His proposed escape hatch is to stop treating hand-written prompts as the durable layer:

Every mature engineering discipline eventually stops doing by hand the very thing it once prided itself on doing by hand. Assembly gave way to compilers, hand-tuned queries gave way to planners, and manual memory management gave way (mostly) to machines that do it better. Prompt-writing is no different.

Coaxing the model with exactly the right words is a real skill, and for one-off tasks it’s often optimal. But to build reliable, improvable, and portable systems we should not be hand-tuning prompts.

For designers, the useful warning is not that prompt craft goes away. It is that fragile prompt craft stops scaling once the AI behavior becomes part of the product. If the system has to survive a model upgrade, the design work has to move into measurements and typed specifications: schemas or contracts that constrain what the model can produce and give the next model something stable to inherit.

In June 2026, Stack Overflow unveiled a full rebrand by studio Koto, repositioning from Q&A platform to “the world’s most vital source for technologists” right as monthly new questions on the site are down roughly 77% since ChatGPT launched. The brand investment and the traffic freefall are happening simultaneously. That is the context for Ishan Gupta, a software engineer at Amazon, and his five-phase history of how the old engineering workflow collapsed.

Software engineering was a craft you absorbed slowly, then practiced in a long, predictable sequence: Dive deep on the technology, write the code, ask Stack Overflow when stuck, escalate to a senior engineer when Stack Overflow failed, ship the ticket. The product manager owned the funnel. The engineer owned the build. Both sides treated this division as physics.

That division is dissolving. Gupta traces it through the IDE-native era (e.g., GitHub Copilot and Cursor), the spec-driven era, and the Claude Code Routines era (Anthropic’s scheduled, persistent agents). At each step, another piece of work that used to require a human gets handed off. Gupta’s diagnosis:

Anthropic recently told its growth team to hire more product managers, not fewer. The reason, as reported in industry coverage, was that Claude Code had quietly turned its engineering org into a team that ships at roughly three times its actual headcount, and the bottleneck moved from the integrated development environment (IDE) to the people deciding what to build.

That detail is easy to miss in the noise of every AI productivity claim. It is also the structural shift the rest of the industry is now living through. The bottleneck in software is no longer typing. It is deciding what to type. And the engineers who treat that as someone else’s problem are about to plateau.

That same shift is what Koto’s rebrand is responding to. Cat Hill, senior strategist at Koto, put the rebrand angle plainly: “In the AI era, everyone wants faster answers. But speed is useful only if the knowledge underneath is trusted.” For designers, that is the opening: the product-thinking gap is no longer a soft skill around the edge of engineering work. It is where the work is moving.

Gupta’s clearest description of the new engineering identity is also the case for why product judgment matters more:

The 2026 version of a great engineer is not the one who writes the most code. It is the one who knows what to build, can prove it is worth building, and has the agent fleet plus the review discipline to ship it without the system collapsing under its own velocity.

The first time I retrofitted a URL scheme onto an existing web app, it wasn’t easy and the engineers on my team were reluctant. The work itself looked like engineering cleanup, but the problem was design debt: the product had no shared model for what its pages, states, and resources were called.

Routes accumulate around backend data models and frontend file conventions rather than the resource hierarchy a user or linking system would recognize. Ownership fragments across teams; each team names things for its own context. By the time someone wants to clean it up, the redirects are locked in by downstream dependencies and analytics instrumentation is tangled around the old slugs. The retrofix was necessary groundwork for new features that needed stable, meaningful URLs to work at all. The real lesson was that we should have had this conversation at the start, not mid-build.

JSTools.Space explains why that cleanup is so expensive:

A URL is part navigation, part application state, and part public interface. Once users bookmark it, search engines index it, monitoring systems record it, and other applications link to it, changing that URL becomes an architectural decision rather than a cosmetic edit.

Good URL design is therefore less about making every address look pretty and more about making it predictable, stable, and unambiguous.

The decision framework should be a team contract from day one: path for identity, query for optional state, fragment for in-page location or client-only state. Inconsistent routes are what make a migration touch everything. Designers should care because this is information architecture in production, not just routing syntax.

Heenesh Patel links Apple’s WWDC 2026 moves (Siri now able to invoke app functions without the user ever opening an app) to a larger skill shift for designers. The polished-UI moment isn’t ending, he argues; its shelf life is just shorter than we think.

This moment might be shorter lived than expected, as we enable agents to execute more tasks on our behalf, screen-based flows fold in on themselves to intents, replaced by API calls and lightweight confirmations. Here the beautifully crafted experience still matters, but it’s not where the experience lives.

As designers continue to rapidly evolve their skills in an AI first world, taste judgement can elevate the experience but only so far and the real differentiator in app design becomes the overall experience architecture, and how flexible and robust apps are in embedding into the platform.

Patel locates the new value in how flexibly and robustly an app’s functions embed into the platform. That is a systems problem before it is a screen-design problem.

Taste is the skill of this moment. Systems thinking is the skill that will become indispensable in the next chapter of design. Designers who start building that capability now will be the ones setting the standard when the shift arrives in full.

The uncomfortable implication in Patel’s urgency: Job Stories (a way to frame user intent in context) and state charts (maps of the states an experience can reach) have been in the UX toolkit for years. What changes is the operating system. If Siri can trigger app functions directly, and if users can move through an experience by intent instead of by screen, designers need to understand the states, permissions, handoffs, and failure paths that sit behind the interface.

Christian Cleberg, an independent developer, traces the internet across his lifetime, from the family Gateway PC in 2001 to the 17-step obstacle course it takes to read a headline in 2026. The essay is nostalgic the way a good memoir is: it names real years and real machines.

Once upon a time, the internet didn’t even exist. When it did, the internet was a place. It was a place you went. You selectively chose to visit the internet, based on your own free will. If you wanted to visit a chat room, or perhaps preview a fancy new Flash game, you visited the internet for a few minutes in the evening before going back to your family or friends. This has changed dramatically in the last 20-30 years. Today, it’s 2026 and woven into nearly every part of daily life for the majority of the Earth’s population.

The transformation Cleberg describes is behavioral, not technical. The internet stopped being something you opted into and became the condition of modern life. The independent web is now infrastructure you rent through friction and attention extraction, not just because of how we use the internet, but because of who decided what it became.

However, there was a dark side to this revolution. As these apps progressed, and the underlying technologies progressed, the companies and investors driving these efforts took a turn. Efforts pushed into the world of micro-transactions, psychological rewards for utilizing a platform, profits over value, and a slow descent into a disregard for the users.

To me, that 2012 era was a turning point. You may personally love a different era (1980/1990s gamer, anyone? Any former phreaks out there?), but to me, it felt like we kept making leaps in technological capability until we reached the point of economic profitability in the 2000s-2010s era, where corporations, investors, and governments took a serious interest in controlling the internet.

Cleberg’s closing distinction: the internet didn’t disappear, the experience of it did. The infrastructure is still there; the sense of it as a place worth exploring is not.

That loss already had a response. Jeffrey Zeldman—web-standards pioneer, founder of A List Apart, and longtime open-web advocate—wrote on his own blog after attending a preview of filmmaker Bao Nguyen’s documentary Code for the People, a film about the builders of the open web:

The answer to this destruction of our shared digital commons by a handful of billionaires is the same as it has always been: own your content on the open web (your domain on a server, like mine on this one), and replace proprietary software with open source alternatives.

Cleberg is writing about the disappearance of a feeling. Zeldman’s post is smaller and more practical: own your domain, publish on the open web, choose tools you can keep. That does not solve the larger fight for the open web, but it does make the response personal. A domain is not nostalgia. It is a small claim of custody over your own work.

The web is forking. Sara Guaglione reports that publishers are starting to build one version of their sites for humans and another for agents. The phrase that matters is Time chief operating officer Mark Howard’s “separating out that traffic.”

“[The bots are] just getting the content itself and the metadata, but they’re not getting the full page experience, and we’re routing all the humans to the full page experience. So we’re separating out that traffic,” Howard said.

“Now we’re starting to think about, as the volume of bot traffic continues to increase significantly – and we see through a number of our vendor partners that we have very high domain authority with AI bot traffic – there’s value in that,” he added.

Howard is making the operational case. Toshit Panigrahi, co-founder and CEO of TollBit, makes the economics explicit:

“Part of onboarding to TollBit is we create your agent site for you,” said Toshit Panigrahi, co-founder and CEO of TollBit. “It really comes down to the token economy. Websites have a lot of HTML tags and JavaScript and CSS and things that don’t have to do with the content. That creates a big bloat in the actual size of the page.”

Markdown can make websites “friendlier” to agents, he added. “AI can comprehend more of your article because they’re not spending money parsing out other HTML that’s on the page. We see, on average, a 90% reduction in tokens, because we have converted the content to markdown.”

That efficiency argument is real. But independent publisher consultant Scott Messer, principal of Messer Media, pushes back:

Yet, even as more publishers quietly spin up agent-friendly feeds, stripped down pages and custom schemes, not everyone is convinced they should be racing to re-architect the web for bots. Independent publisher consultant Scott Messer, principal of Messer Media, argues that building for agents should be a highly qualified decision, not the default. His reasoning: traffic isn’t the reward in an agentic environment – if there is no click, no ad impression and no check, the build is pure cost.

“If you believe there’s a value to being discovered by these bots and agents, then you should build them. If you don’t believe [that], I would ask, why would you build them?,” he said.

That is the question under all of this. The rendering layer—the HTML, CSS, and JavaScript that turn server content into pages—is exactly the surface designers build, and publishers and content creators are now deciding which parts of that surface matter when the visitor is no longer a person.

I’d argue that RSS can be easily consumed by agents too.

I wrote about who killed Google Reader because Reader’s shutdown felt like losing a whole way of using the web: the curation layer, the accidental social network, the daily habit. Matthew Guay, editor at Buttondown, is interested in the part that survived. His answer is blunt: nothing important died.

The feeds walked out intact: three million people moved to Feedly in two weeks, and the content never left the blogs. Guay’s point is that we were mourning the right thing—the curation layer, the accidental social network—at the wrong level of the stack: the aggregation service, not the open protocol underneath it.

Not an iota of data was lost, as the content that filled the feeds lived on individual blogs and websites. Google+ users wouldn’t be so lucky, six years later, when that network too was shuttered, taking with it their non-portable social graphs and ephemeral posts.

That contrast is the spine of the piece. Google+ and Google Reader died the same administrative death, both casualties of the same corporate pivot, but they died completely different technical deaths. Reader’s users lost a habit. Google+ users lost everything. The portability of the underlying protocol made the difference, and it’s why early RSS developer Dave Winer could be so clear-eyed about Reader’s shutdown at the time:

“I won’t miss it,” said early RSS developer Dave Winer more cantankerously—or, perhaps, more clear-eyedly—of Google Reader. “Never used the damn thing. Didn’t trust the idea of a big company like Google’s interests being so aligned with mine that I could trust them to get all my news.”

But that’s the thing about RSS. Winer didn’t need to love Reader for people to follow his writing there. And when it went away, those same readers could still follow him directly, could still read the words on his site that yesterday they’d read under Google’s auspices.

Designer and author Marcin Wichary picked up the same portability-survival thread in his blog Unsung, marking the 13th anniversary, and he quoted my February post in the process. His read: “I am worried about the open web, but excited seeing some resurgence in RSS usage, and more and more people wanting to come back to the feeling of control, care, and intentionality that using Reader represented.”

That resurgence Wichary is watching is what Guay’s closing points toward, too:

It was the open protocols, the RSS feeds and email, that alone offered a direct connection to your favorite writers and publications, unmediated by algorithms. Sure, they didn’t come with built in sharing features, they were harder to discover, they required more work to turn into a community. But once you did find them, they were sticky, a connection no one other than the publishers themselves could take away.

This blog is my version of that direct channel: a website, built one reader at a time, outside the platforms that decide what gets seen. I still miss Reader’s curation layer. The win is that the open-protocol layer underneath it—RSS, personal blogs, the direct connection—survived anyway.