Tim Berners-Lee, the father of the web who gave away the technology for free, says that we are at an inflection point with data privacy and AI. But before he makes that point, he reminds us that we are the product:

Today, I look at my invention and I am forced to ask: is the web still free today? No, not all of it. We see a handful of large platforms harvesting users’ private data to share with commercial brokers or even repressive governments. We see ubiquitous algorithms that are addictive by design and damaging to our teenagers’ mental health. Trading personal data for use certainly does not fit with my vision for a free web.

On many platforms, we are no longer the customers, but instead have become the product. Our data, even if anonymised, is sold on to actors we never intended it to reach, who can then target us with content and advertising. This includes deliberately harmful content that leads to real-world violence, spreads misinformation, wreaks havoc on our psychological wellbeing and seeks to undermine social cohesion.

And about that fork in the road with AI:

In 2017, I wrote a thought experiment about an AI that works for you. I called it Charlie. Charlie works for you like your doctor or your lawyer, bound by law, regulation and codes of conduct. Why can’t the same frameworks be adopted for AI? We have learned from social media that power rests with the monopolies who control and harvest personal data. We can’t let the same thing happen with AI.

I’m happy that the conversation around the design talent crisis continues. Carly Ayres, writing for It’s Nice That picks up the torch and speaks to designers and educators about this topic. What struck me—and I think what adds to the dialogue—is the notion of the belief gap. Ayres spoke with Naheel Jawaid, founder of Silicon Valley School of Design, about it:

“A big part of what I do is just being a coach, helping someone see their potential when they don’t see it yet,” Naheel says. “I’ve had people tell me later that a single conversation changed how they saw themselves.”

In the past, belief capital came from senior designers taking juniors under their wing. Today, those same seniors are managing instability of their own. “It’s a bit of a ‘dog eat dog world’-type vibe,” Naheel says. “It’s really hard to get mentorship right now.”

The whole piece is great. Tighter than my sprawling three-parter. I do think there’s a piece missing though. While Ayres highlights the issue and offers suggestions from designer leaders, businesses need to step up and do something about the issue—i.e., hire more juniors. Us recognizing it is the first step.

Jason Spielman put up a case study on his site for his work on Google’s NotebookLM:

The mental model of NotebookLM was built around the creation journey: starting with inputs, moving through conversation, and ending with outputs. Users bring in their sources (documents, notes, references), then interact with them through chat by asking questions, clarifying, and synthesizing before transforming those insights into structured outputs like notes, study guides, and Audio Overviews.

And yes, he includes a sketch he did on the back of a napkin.

I’ve always wondered about the UX of NotebookLM. It’s not typical and, if I’m being honest, not exactly super intuitive. But after a while, it does make sense. Maybe I’m the outlier though, because Spielman’s grandmother found it easy. In an interview last year on Sequoia Capital’s Training Data, he recalls:

I actually do think part of the explosion of audio overviews was the fact it was a simple one click experience. I was on the phone with my grandma trying to explain her how to use it and it actually didn’t take any explanation. I’m like, “Drop in a source.” And she’s like, “Oh! I see. I click this button to generate it.” And I think that the ease of creation is really actually what catalyzed so much explosion. So I think when we think about adding these knobs [for customization] I think we want to do it in a way that’s very intentional.

Chatboxes have become the uber box for all things AI. The criticism of this blank box has been the cold start issue. New users don’t know what to type. Designers shipping these product mostly got around this problem by offering suggested prompts to teach users about the possibilities.

The issue on the other end is that expert users end up creating their own library of prompts to copy and paste into the chatbox for repetitive tasks.

Sharang Sharma writing in UX Collective illustrates how these UIs can be smarter by being predictive of intent:

Contrary, Predictive UX points to an alternate approach. Instead of waiting for users to articulate every step, systems can anticipate intent based on behavior or common patterns as the user types. Apple Reminders suggests likely tasks as you type. Grammarly predicts errors and offers corrections inline. Gmail’s Smart Compose even predicts full phrases, reducing the friction of drafting entirely.

Sharma says that the goal of predictive UX is to “reduce time-to-value and reframe AI as an adaptive partner that anticipates user’s intent as you type.”

Imagine a little widget that appears within the chatbox as you type. Kind of a cool idea.

Thinking about this morning’s link about web forms, if you abstract why it’s so powerful, you get to the point of human-computer interaction: the computer should do what the user intends, not the buttons they push.

Matt Webb reminds us about the DWIM, or Do What I Mean philosophy in computing that was coined by Warren Teitelman in 1966. Webb quotes computer scientist Larry Masinter:

DWIM is an embodiment of the idea that the user is interacting with an agent who attempts to interpret the user’s request from contextual information. Since we want the user to feel that he is conversing with the system, he should not be stopped and forced to correct himself or give additional information in situations where the correction or information is obvious.

Webb goes on to say:

Squint and you can see ChatGPT as a DWIM UI: it never, never, never says “syntax error.”

Now, arguably it should come back and ask for clarifications more often, and in particular DWIM (and AI) interfaces are more successful the more they have access to the user’s context (current situation, history, environment, etc).

But it’s a starting point. The algo is: design for capturing intent and then DWIM; iterate until that works. AI unlocks that.

Forms is one of the fundamental things we make users do in software. Whether it’s the login screen, billing address form, or a mortgage application, forms are the main method for getting data from users and into computer-accessible databases. The human is deciding what piece of information to put into which column in the database. With AI, form filling should be much simpler.

Luke Wroblewski makes the argument:

With Web forms, the burden is on people to adapt to databases. Today’s AI models, however, can flip this requirement. That is, they allow people to provide information in whatever form they like and use AI do the work necessary to put that information into the right structure for a database.

How can it work?

With AgentDB connected to an AI model (via an MCP server), a person can simply say “add this” and provide an image, PDF, audio, video, you name it. The model will use AgentDB’s template to decide what information to extract from this unstructured input and how to format it for the database. In the case where something is missing or incomplete, the model can ask for clarification or use tools (like search) to find possible answers.

I believe purity tests of any sort are problematic. And it’s much too easy to throw around the “This is AI slop!” claim. AI was used in the main title sequence for the Marvel TV show Secret Invasion. But it was on purpose and aligned with the show’s themes of shapeshifters.

Anyway, Daniel John, writing in the Creative Bloq:

[Lady] Gaga just dropped the music video for The Dead Dance, a song debuted in Season 2 of Netflix’s Wednesday. Directed by Tim Burton, it’s a suitably nightmarish black-and-white cacophony of monsters and dolls. But some are already claiming that parts of it were made using AI.

John shows a tweet from @graveyardquy as an example:

i didn’t think we’d ever be in a timeline where a tim burton x lady gaga collab would turn out to be AI slop… but here we are

We need to separate quality critiques from tool usage. If it looks good and is appropriate, I’m fine with CG, AI, and whatever comes next that helps tell the story. Same goes for what we do as designers, ’natch.

Gaga’s song is great. It’s a bop, as the kids say, with a neat music video to boot.

Here’s a fun visual essay about a artist Yufeng Zhao’s piece “Alt Text in NYC.” It’s a essentially a visual search engine that searches all the text (words) on the streets of New York City. The dataset comprises of over eight million photos from Google Street View! Matt Daniels, writing for The Pudding:

The result is a search engine of much of what’s written in NYC’s streets. It’s limited to what a Google Street View car can capture, so it excludes text in areas such as alleyways and parks, or any writing too small to be read by a moving vehicle.

The scale of the data is immense: over 8 million Google Street View images (from the past 18 years) and 138 million identified snippets of text.

Just over halfway down the article, there is a list of the top 1,000 words in the data. Most are expected words from traffic signs like “stop.” But number twenty-five is “Fedders,” the logo of an air-conditioner brand popular in the 1950s to the 1990s. They’re all over the exteriors of the city’s buildings.

Best viewed on your computer, IMHO.

Josh Miller, CEO, and Hursh Agrawal, CTO, of The Browser Company:

Today, The Browser Company of New York is entering into an agreement to be acquired by Atlassian in an all-cash transaction. We will operate independently, with Dia as our focus. Our objective is to bring Dia to the masses.

Super interesting acquisition here. There is zero overlap as far as I can tell. Atlassian’s move is out of left-field. Dia’s early users were college students. The Browser Company more recently opened it up to former Arc users. Is this bet for Atlassian—the company that makes tech-company-focused products like Jira and Confluence—around the future of work and collaboration? Is this their first move against Salesforce? 🤔

DOC is a publication from Fabricio Teixeira and Caio Braga that I’ve linked to before. Their latest reflection is on interfaces.

A good user interface is a good conversation.

Interfaces thrive on clarity, responsiveness, and mutual understanding. In a productive dialogue, each party clearly articulates their intentions and receives timely, understandable responses. Just as a good conversationalist anticipates the next question or need, a good interface guides you smoothly through your task. At their core, interfaces translate intent into action. They’re a bridge between what’s in your head and what the product can do.

Reflection is the best word I’ve found to describe these pieces. They’re hype-free, urging us to take a step back, and—at least for me—a reminder about our why.

In the end, interfaces are also a space for self-expression.

The ideal of “no interface” promises ultimate efficiency and direct access—but what do we lose in that pursuit? Perhaps the interface is not just a barrier to be minimized, but a space for human expression. It’s a canvas; a place to imbue a product with personality, visual expression, and a unique form of art.

When we strip that away, or make everything look the same, we lose something important. We trade the unique and the delightful for the purely functional. We sacrifice a vital part of what makes technology human: the thoughtful, and sometimes imperfect, ways we present ourselves to the world.

Interesting piece from Vaughn Tan about a critical thinking framework that is disguised as a piece about building better AI UIs for critical thinking. Sorry, that sentence is kind of a tongue-twister. Tan calls out—correctly—that LLMs don’t think, or in his words, can’t make meaning:

Meaningmaking is making inherently subjective decisions about what’s valuable: what’s desirable or undesirable, what’s right or wrong. The machines behind the prompt box are remarkable tools, but they’re not meaningmaking entities.

Therefore when users ask LLMs for their opinions on matters, e.g., as in the therapy use case, the AIs won’t come back with actual thinking. IMHO, it’s semantics, but that’s another post.

Anyhow, Tan shares a pen and paper prototype he’s been testing, which breaks down a major decision into guided steps, or put another way, a framework.

This user experience was designed to simulate a multi-stage process of structured elicitation of various aspects of strongly reasoned arguments. This design explicitly addresses both requirements for good tool use. The structured prompts helped students think critically about what they were actually trying to accomplish with their custom major proposals — the meaningmaking work of determining value, worth, and personal fit. Simultaneously, the framework made clear what kinds of thinking work the students needed to do themselves versus what kinds of information gathering and analysis could potentially be supported by tools like LLMs.

This guided or framework-driven approach was something I attempted wtih Griffin AI. Via a series of AI-guided prompts to the user—or a glorified form, honestly—my tool helped users build brand strategies. To be sure, a lot of the “thinking” was done by the model, but the idea that an AI can guide you to critically think about your business or your client’s business was there.

Designer Tey Bannerman writes that when he hears “human in the loop,” he’s reminded of a story about Lieutenant Colonel Stanislav Petrov, a Soviet Union duty watch officer who monitored for incoming missile strikes from the US.

12:15 AM… the unthinkable. Every alarm in the facility started screaming. The screens showed five US ballistic missiles, 28 minutes from impact. Confidence level: 100%. Petrov had minutes to decide whether to trigger a chain reaction that would start nuclear war and could very well end civilisation as we knew it.

He was the “human in the loop” in the most literal, terrifying sense.

Everything told him to follow protocol. His training. His commanders. The computers.

But something felt wrong. His intuition, built from years of intelligence work, whispered that this didn’t match what he knew about US strategic thinking.

Against every protocol, against the screaming certainty of technology, he pressed the button marked “false alarm”.

Twenty-three minutes of gripping fear passed before ground radar confirmed: no missiles. The system had mistaken a rare alignment of sunlight on high-altitude clouds for incoming warheads.

His decision to break the loop prevented nuclear war.

Then Bannerman shares an awesome framework he developed that allows humans in the loop in AI systems “genuine authority, time to think, and understanding the bigger picture well enough to question” the system’s decision. Click on to get the PDF from his site.

Simon Sherwood, writing in The Register:

Amazon Web Services CEO Matt Garman has suggested firing junior workers because AI can do their jobs is “the dumbest thing I’ve ever heard.”

Garman made that remark in conversation with AI investor Matthew Berman, during which he talked up AWS’s Kiro AI-assisted coding tool and said he’s encountered business leaders who think AI tools “can replace all of our junior people in our company.”

That notion led to the “dumbest thing I’ve ever heard” quote, followed by a justification that junior staff are “probably the least expensive employees you have” and also the most engaged with AI tools.

“How’s that going to work when ten years in the future you have no one that has learned anything,” he asked. “My view is you absolutely want to keep hiring kids out of college and teaching them the right ways to go build software and decompose problems and think about it, just as much as you ever have.”

Yup. I agree.

This post from Carly Ayres breaks down a beef between Michael Roberson (developer of an AI-enabled moodboard tool) and Elizabeth Goodspeed (writer and designer, oft-linked on this blog) and explores ragebait, putting in the reps as a junior, and designers as influencers.

Tweet from Michael Roberson

The tweet earned 30,000 views, but only about 20 likes. “That ratio was pretty jarring,” [Roberson] said. Still, the strategy felt legible. “When I post things like, ‘if you don’t do X, you’re not going to make it,’ obviously, I don’t think that. These tools aren’t really capable of replacing designers just yet. It’s really easy to get views baiting and fear-mongering.”

Much like the provocative Artisan campaign, I think this is a net negative for the brand. Pretty sure I won’t be trying out Moodboard AI anytime soon, ngl.

But stepping back from the internet beef, Ayres argues that it’s a philosophical difference about the role friction in the creative process.

Michael’s experience mirrors that of many young designers: brand audits felt like busywork during his Landor internship. “That process was super boring,” he told me. “I wasn’t learning much by copy-pasting things into a deck.” His tool promises to cut through that inefficiency, letting teams reach visual consensus faster and spend more time on execution.

Young Michael, the process is the point! Without doing this boring stuff, by automating it with AI, how are you going to learn? This is but one facet of the whole discussion around expertise, wisdom, and the design talent crisis.

Goodspeed agrees with me:

Elizabeth sees it differently. “What’s interesting to me,” Elizabeth noted, “is how many people are now entering this space without a personal understanding of how the process of designing something actually works.” For her, that grunt work was formative. “The friction is the process,” she explained. “That’s how you form your point of view. You can’t just slap seven images on a board. You’re forced to think: What’s relevant? How do I organize this and communicate it clearly?”

Ultimately, the saddest point that Ayres makes—and noted by my friend Eric Heiman—is this:

When you’re young, online, and trying to get a project off the ground, caring about distribution is the difference between a hobby and a company. But there’s a cost. The more you perform expertise, the less you develop it. The more you optimize for engagement, the more you risk flattening what gave the work meaning in the first place. In a world where being known matters more than knowing, the incentives point toward performance over practice. And we all become performers in someone else’s growth strategy.

…Because when distribution matters more than craft, you don’t become a designer by designing. You become a designer by being known as one. That’s the game now.

As a follow-up to yesterday’s item on how Google’s AI overviews are curtailing traffic to websites by as much as 25%, here is a link to Nielsen Norman Group’s just-published study showing that generative AI is reshaping search.

Kate Moran, Maria Rosala and Josh Brown:

While AI offers compelling shortcuts around tedious research tasks, it isn’t close to completely replacing traditional search. But, even when people are using traditional search, the AI-generated overview that now tops almost all search-results pages steals a significant amount of attention and often shortcuts the need to visit the actual pages.

They write that users have developed a way to search over the years, skipping sponsored results and heading straight for the organic links. Users also haven’t completely broken free of traditional Google Search, now adding chatbots to the mix:

While generative AI does offer enough value to change user behaviors, it has not replaced traditional search entirely. Traditional search and AI chats were often used in tandem to explore the same topic and were sometimes used to fact-check each other.

All our participants engaged in traditional search (using keywords, evaluating results pages, visiting content pages, etc.) multiple times in the study. Nobody relied entirely on genAI’s responses (in chat or in an AI overview) for all their information-seeking needs.

In many ways, I think this is smart. Unless “web search” is happening, I tend double-check ChatGPT and Claude, especially for anything historical and mission-critical. I also like Perplexity for that fact—because it shows me its receipts by giving me sources.

Jessica Davies reports that new publisher data suggests that some sites are getting 25% less traffic from Google than the previous year.

Writing in Digiday:

Organic search referral traffic from Google is declining broadly, with the majority of DCN member sites — spanning both news and entertainment — experiencing traffic losses from Google search between 1% and 25%. Twelve of the respondent companies were news brands, and seven were non-news.

Jason Kint, CEO of DCN, says that this is a “direct consequence of Google AI Overviews.”

I wrote previously about the changing economics of the web here, here, and here.

And related, Eric Mersch writes in a LinkedIn post that Monday.com’s stock fell 23% because co-CEO Roy Mann said, “We are seeing some softness in the market due to Google algorithm,” during their Q2 earnings call and the analysts just kept hammering him and the CFO about how the algo changes might affect customer acquisition.

Analysts continued to press the issue, which caught company management completely off guard. Matthew Bullock from Bank of America Merrill Lynch asked frankly, “And then help us understand, why call this out now? How did the influence of Google SEO disruption change this quarter versus 1Q, for example?” The CEO could only respond, “So look, I think like we said, we optimize in real-time. We just budget daily,” implying that they were not aware of the problem until they saw Q2 results.

This is the first public sign that the shift from Google to AI-powered searches is having an impact.

I enjoyed this interview with Notion’s CEO, Ivan Zhao over at the Decoder podcast, with substitute host, Casey Newton. What I didn’t quite get when I first used Notion was the “LEGO” aspect of it. Their vision is to build business software that is highly malleable and configurable to do all sorts of things. Here’s Zhao:

Well, because it didn’t quite exist with software. If you think about the last 15 years of [software-as-a-service], it’s largely people building vertical point solutions. For each buyer, for each point, that solution sort of makes sense. The way we describe it is that it’s like a hard plastic solution for your problem, but once you have 20 different hard plastic solutions, they sort of don’t fit well together. You cannot tinker with them. As an end user, you have to jump between half a dozen of them each day.

That’s not quite right, and we’re also inspired by the early computing pioneers who in the ‘60s and ‘70s thought that computing should be more LEGO-like rather than like hard plastic. That’s what got me started working on Notion a long time ago, when I was reading a computer science paper back in college.

From a user experience POV, Notion is both simple and exceedingly complicated. Taking notes is easy. Building the system for a workflow, not so much.

In the second half, Newton (gently) presses Zhao on the impact of AI on the workforce and how productivity software like Notion could replace headcount.

Newton: Do you think that AI and Notion will get to a point where executives will hire fewer people, because Notion will do it for them? Or are you more focused on just helping people do their existing jobs?

Zhao: We’re actually putting out a campaign about this, in the coming weeks or months. We want to push out a more amplifying, positive message about what Notion can do for you. So, imagine the billboard we’re putting out. It’s you in the center. Then, with a tool like Notion or other AI tools, you can have AI teammates. Imagine that you and I start a company. We’re two co-founders, we sign up for Notion, and all of a sudden, we’re supplemented by other AI teammates, some taking notes for us, some triaging, some doing research while we’re sleeping.

Zhao dodges the “hire fewer people” part of the question and instead, answers with “amplifying” people or making them more productive.

Ben Davies-Romano argues that the AI chat box is our new design interface:

Every interaction with a large language model starts the same way: a blinking cursor in a blank text field. That unassuming box is more than an input — it’s the interface between our human intent and the model’s vast, probabilistic brain.

This is where the translation happens. We pour in the nuance, constraints, and context of our ideas; the model converts them into an output. Whether it’s generating words, an image, a video sequence, or an interactive prototype, every request passes through this narrow bridge.

It’s the highest-stakes, lowest-fidelity design surface I’ve ever worked with: a single field that stands between human creativity and an engine capable of reshaping it into almost any form, albeit with all the necessary guidance and expertise applied.

In other words, don’t just say “Make it better,” but guide the AI instead.

That’s why a vague, lazy prompt, like “make it better”, is the design equivalent of telling a junior designer “make it intuitive” and walking away. You’ll get something generic, safe, and soulless, not because the AI “missed the brief,” but because there was no brief.

Without clear stakes, a defined brand voice, and rich context, the system will fill in the blanks with its default, most average response. And “average” is rarely what design is aiming for.

And he makes a point that designers should be leading the charge on showing others what generative AI can do:

In the age of AI, it shouldn’t be everyone designing, per say. It should be designers using AI as an extension of our craft. Bringing our empathy, our user focus, our discipline of iteration, and our instinct for when to stop generating and start refining. AI is not a replacement for that process; it’s a multiplier when guided by skilled hands.

So, let’s lead. Let’s show that the real power of AI isn’t in what it can generate, but in how we guide it — making it safer, sharper, and more human. Let’s replace the fear and the gimmicks with clarity, empathy, and intentionality.

The blank prompt is our new canvas. And friends, we need to be all over it.

Yesterday, OpenAI launched GPT-5, their latest and greatest model that replaces the confusing assortment of GPT-4o, o3, o4-mini, etc. with just two options: GPT-5 and GPT-5 pro. The reasoning is built in and the new model is smart enough to know what to think harder, or when a quick answer suffices.

Simon Willison deep dives into GPT-5, exploring its mix of speed and deep reasoning, massive context limits, and competitive pricing. He sees it as a steady, reliable default for everyday work rather than a radical leap forward:

I’ve mainly explored full GPT-5. My verdict: it’s just good at stuff. It doesn’t feel like a dramatic leap ahead from other LLMs but it exudes competence—it rarely messes up, and frequently impresses me. I’ve found it to be a very sensible default for everything that I want to do. At no point have I found myself wanting to re-run a prompt against a different model to try and get a better result.

It’s a long technical read but interesting nonetheless.

My former colleague from Organic, Christian Haas—now ECD at YouTube—has been experimenting with AI video generation recently. He’s made a trilogy of short films called AI Jobs.

You can watch part one above 👆, but don’t sleep on parts two and three.

Haas put together a “behind the scenes” article explaining his process. It’s fascinating and I’ll want to play with video generation myself at some point.

I started with a rough script, but that was just the beginning of a conversation. As I started generating images, I was casting my characters and scouting locations in real time. What the model produced would inspire new ideas, and I would rewrite the script on the fly. This iterative loop continued through every stage. Decisions weren’t locked in; they were fluid. A discovery made during the edit could send me right back to “production” to scout a new location, cast a new character and generate a new shot. This flexibility is one of the most powerful aspects of creating with Gen AI.

It’s a wonderful observation Haas has made—the workflow enabled by gen AI allows for more creative freedom. In any creative endeavor where the production of the final thing is really involved and utilizes a significant amount of labor and materials, be it a film, commercial photography, or software, planning is a huge part. We work hard to spec out everything before a crew of a hundred shows up on set or a team of highly-paid engineers start coding. With gen AI, as shown here with Google’s Veo 3, you have more room for exploration and expression.

UPDATE: I came across this post from Rory Flynn after I published this. He uses diagrams to direct Veo 3.

For the past year, CPG behemoth Unilever has been “working with marketing services group Brandtech to build up its Beauty AI Studio: a bespoke, in-house system inside its beauty and wellbeing business. Now in place across 18 different markets (the U.S. and U.K. among them), the studio is being used to make assets for paid social, programmatic display inventory and e-commerce usage across brands including Dove Intensive Repair, TRESemme Lamellar Shine and Vaseline Gluta Hya.”

Sam Bradley, writing in Digiday:

The system relies on Pencil Pro, a generative AI application developed by Brandtech Group. The tool draws on several large language models (LLMs), as well as API access to Meta and TikTok for effectiveness measurement. It’s already used by hearing-care brand Amplifon to rapidly produce text and image assets for digital ad channels.

In Unilever’s process, marketers use prompts and their own insights about target audiences to generate images and video based on 3D renders of each product, a practice sometimes referred to as “digital twinning.” Each brand in a given market is assigned a “BrandDNAi” — an AI tool that can retrieve information about brand guidelines and relevant regulations and that provides further limitations to the generative process.

So far, they haven’t used this system to generate AI humans. Yet.