The screen that speaks for the house

Thinking about how Homeview could be a social object, but misses the point. But opens up a sharper view of what is needed. And also more handpicked news from last on human-AI-things relationships as always.

The screen that speaks for the house
"A screen that speaks for the house, a social object connecting family life, showing it in use", acc Midjourney

Dear reader!

This edition has a slightly different format. Monday is usually the day I take time to compose and finalize the newsletter, but this Monday I ran a workshop at Dutch Design Week. So I spoke this introduction into my phone on the drive back from Eindhoven. The triggered thought was already set up earlier in the week. Normally I go back and forth on it a few times; this week I kept that to the bare minimum. The same goes for the links: they are the ones I came across, not the result of going through all the sources I usually follow.

Week 409: Dutch Design Week workshop and thoughts on a social object in your house

Looking back on the week, it was busy with all kinds of things, including two workshops. One client one, with a focus on speculative design in context of the Cities of Things Design Manifesto, and yesterday one at Dutch Design Week, where I connected the same speculative design canvas method with a Wijkbot making exersize / performance. It was a lot of fun!

It was a small group of eight, a good mix of people from academia, practice, and smaller and larger agencies, from The Netherlands, USA, and Germany. After a presentation that sparked some good discussion, they used speculative design to imagine critical objects for the street, with Lego as a building material.

The two teams went in different directions. One imagined an object that turns leftover furniture and other things on the street into a social experience, a kind of moving café. It stayed rather conceptual, but that suited the exercise. The other team made a wearable that turns the grim outside world into a happy place at the push of a button. The hybrid future world-building they have been exploring in the first round was a clear driver. The Cities of Things manifesto was maybe less clearly represented, something to think about for the setup.

My plan was for these Lego designs to become a clear starting point for building with the Wijkbot. I think that still holds, but the translation from a speculative object to a working prototype needs more time than we had. The testing on the street gave good energy again, though there was too little time and too few passers-by to really test the interactions. We had some nice reflections afterwards with new guests for the drinks. The experiment of connecting the two workshop formats has a lot of potential to explore designing with immersive AI, but would benefit from a longer format, maybe in two chapters. Something to think about a bit more.

On to the triggered thought. The social behaving object is returning.

This week’s triggered thought: The screen that speaks for the house

Apple introduced a set of home devices last week. Next to a new Apple TV and HomePod, the HomeView got most of the attention: a square 7-inch screen on a small dome, or flat against the wall. It is cute. The hinge, the shape, the way it sits on a kitchen counter give it a personality. What it is for is less clear.

On Dithering, John Gruber and Ben Thompson described it as something like CarPlay for the home. The comparison helps, partly because it shows where it breaks. In the car, the relation is clear. There is one driver, one task, and the phone’s content needs a different form while you drive. The home is not one person’s context. Several people live there together, with routines, small negotiations, and things that happen to all of them at once.

Their point is that the home doesn’t need apps. It needs services that respond to situations. The doorbell rings and the screen becomes a doorbell monitor. Someone is cooking and it holds the recipe they found on their phone. It could be a family organizer without taking over what everyone does on their own device.

Apple frames HomeView as “central to the home and personal to everyone in it.” That sentence holds the tension. The device recognizes up to six faces and switches to each person’s messages and calendar. The household becomes six individuals taking turns at a screen. It is shared, but it is not yet social.

What would make it social? A social object is more than something people share. It plays an active role in a social structure: a family, a neighbourhood, a community. It holds something of the group that no single member owns, and it shapes how members relate to each other. A family calendar on the fridge does this in a modest way.

One detail in HomeView points somewhere interesting. The interface adapts to your distance, showing less from across the room and more as you come closer. That is the kind of move Apple once made with the iPhone, turning physical qualities like touch and location into a software experience. The hardware seems to understand the room. The software still thinks in apps and accounts.

I recognize this problem from my own work. In the Grid, a speculative scenario I use to explain how a home full of intelligent things might behave, the house has its own rhythms and predictions. Plants are cared for, energy use is balanced against goals, and only now and then does the system surface to the people living there. In my visualizations those moments appear on two handheld screens, one for the plants, one for energy. Looking at them again, they are dashboards and settings panes. Recognizable, but they miss what the scenario is about. When I present it, I always explain that the screens should show the relation between the people and the things. I never designed that relation into the image itself.

So the open question is not only Apple’s. If a house starts to observe, act and predict, what does its screen show? The status of devices, or the state of a relationship? And whose voice is it? Does the house address the family as a whole, whoever happens to stand in front of it, or speak on behalf of the person who set it up?

There is a hopeful version, where a shared object supports family life instead of splitting it into accounts. There is also a version where the entry ticket to the household is your face. What about the guest, the child too young for an account, the cleaner, the neighbour who drops by? A home is never only its registered members.

Maybe the design challenge for this kind of device is less about the screen and more about how to represent a relation: between people, between people and their things, and between a household and the intelligence it lives with. Who gets to shape that relation is worth asking before these devices arrive in kitchens next month.

Notions from last week’s news

This week, a bit shorter list. And with fewer reflections, due to time squeeze.

From the frontier: Self-replicating models on the edge, Open-weight agents, Amazon new-bees, Google and Meta catching up,

Human-AI relations

Teaming up in times of agentic coding.

AI Changed How We Code. It Hasn't Changed How We Work Together.
An intern’s notes on distributed teams, shared context, and a first attempt at keeping the meaning behind meetings
The AI Challenge is Coming into Focus | HESA
Morning all. I have some thoughts today on how AI is likely to affect teaching and assessment in higher education. A few weeks ago, I was at an event hosted by RBC, which was kind of a fireside chat with Anthropic’s Head of North American Government Affairs, Brian Peters. This was shortly after Anthropic released […]

Expecting goofy AI characters.

https://arstechnica.com/tech-policy/2026/10/disney-found-a-way-to-buy-character-ai-without-buying-its-lawsuits/

Writing skills will find new roles.

Board Agent
Automattic is always 5–15 years ahead of other companies, but things happen faster now. I predict other companies will start scrambling to add great writers to their boards, because, well, look at …

Loss of human connection as the dangerous offset.

I’m a pediatrician. Loss of human connection is the biggest threat to kids, not AI | Megan Moreno
Amid worries about children’s relationship to tech, the solution is often spending more time talking openly with each other

A humanist AI code.

Humanist AI Code of Conduct | Microsoft AI
We make responsible AI to empower people's lives.
Computers Made of Human Brain Tissue Identify Braille Letters In Breakthrough
Scientists discovered that human brain organoids, which are small neural clusters derived from stem cells, could identify the Braille alphabet with a high degree of accuracy.

Agent super intelligence

Can a Million AI Agents Create Superintelligence?
Ramez Naam gives a skeptic’s take on agent swarms.

Physical AI

Cyber-attacks in fridges, the new IoT fridge meme?

https://apnews.com/article/france-cyberattack-supermarkets-refrigeration-systems-78a813a8efd466b360655339f96083dc

A hacker broke into John Deere and SharkNinja with an army of AI agents
The AI agents infiltrated more than 100 websites.

Context engineering.

Context Engineering Is Knowing When to Start Over
Old instructions and earlier mistakes can derail your AI chats. Here’s how to reset without losing useful work.

A robot scale-up production line

How ADI is helping robotics startups scale - The Robot Report
Robotics businesses need help keeping up with innovations as they scale. Analog Devices Inc. or ADI is offering its Emerging Tech Hub.

A tipping point?

A woman was arrested in Philly for blocking a delivery robot
Carolyn Marvin was charged with ‘disorderly conduct’ for protesting delivery robots taking jobs.

In AI in human senses perspective.

Filtered for ears and memory
Posted on Saturday 17 Oct 2026. 796 words, 7 links. By Matt Webb.

Humans impersonating machines.

Roooobot. Huuuuuman.
This woman does some amazing things with a series of white rings — lots of precision optical illusions. It’s mesmerizing watching people moving like machines —
Daimon Robotics building the infrastructure layer for tactile intelligence - The Robot Report
Daimon Robotics is connecting tactile sensing, real-world data collection, simulation, and AI models to help robots perform more reliably.

Tech in civic societies

Edge AI for smart cities.

Edge AI for Smart Cities: From Real-Time Insights to Cognitive Urban Systems - Wiley Science and Engineering Content Hub
Download this complimentary White Paper today! This white paper explains how edge AI and physical AI help cities move beyond passive monitoring. It shows how urban infrastructure can sense, reason, and act on real-time conditions across public safety, mobility, and infrastructure management. What you will learn about: How the shift from smart cities to cognitive cities depends on infrastructure that reasons and acts, rather than infrastructure that only monitors. Why edge AI supports low-latency decisions, keeps sensitive data within city boundaries, and reduces dependency on centralized cloud infrastructure. How computer vision, multimodal sensing, and generative and agentic systems work together to support public safety, mobility, and infrastructure applications. How a tiered edge architecture assigns monitoring, planning, and action tasks across the far edge, regional, and cloud layers, and what durability, security, and interoperability requirements this places on hardware and software. Click 'LOOK INSIDE' to Download the PDF Now.

More crisis signals. Or not?

The AI bet is running straight into a capital crunch
DBS’ Tan Su Shan, the CEO of Singapore’s largest bank, sees a vibe shift, with investors starting to question AI returns.
View: Why the AI bubble fears are overblown
Money is not going to be the thing that finally causes the AI train to go off the rails.
CoreWeave’s debt is becoming the whole AI industry’s problem
Does anyone else hear that ticking sound?

World leveling small AI?

World Bank says ‘small AI’ could transform developing countries
Bank’s chief knowledge officer says the tech could speed up economic development from centuries to decades

Agency crisis.

The biggest threat from AI is the sense that we have no say in it. To that I say game on | Peter Lewis
The tech bros and their highly paid sock puppets are banking on us giving in to their ‘inevitability doctrine’. But in Australia, and around the world, resistance is rising

Energy challenges

https://arstechnica.com/science/2026/10/batteries-have-a-small-carbon-footprint-we-can-make-it-smaller/

How the present eats the future.

When The Present Eats The Future | NOEMA
The Achilles heel of consumer democracies.

Paper for the week

Characterizing Agentic Flooding of Government Services

Schmitz, C., Hammond, L., & Chan, A. (2026, October). Characterizing Agentic Flooding of Government Services. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (Vol. 9, No. 4, pp. 2266-2278).

https://doi.org/10.48550/arXiv.2608.16603

What’s up for the coming week?


I did not check all the calendars, but I captured some during the week. Of course, there is more DDW, and I will join Design+AI symposium today with ao Maria Luce Lupetti keynoting, present a conversation starter at the ESC dinner, see Kars Alfrink at CLICKNL digital innovation sessions, w workshop ‘Bewoners in-the-loop’ and join the Next Nature design debate.

Also this week, what I will miss. Algorithmic decision-making in Pakhuis Amsterdam. Next week, I will be on a panel.

Have a great week!