
Intelligent relationships - reflecting on the future of smart devices
About
During the talk, we will reflect on the future of intelligent devices, speculating on their role in our lives. How can the relationship with our devices be different? How can it be more playful? All the reflections will be grounded in a will to design democratic, inclusive, and engaging experiences with emerging technologies by gradually building trust and understanding and promoting human participation. Let's pause and re-think.
Watch the full talk
Watch this WaysConf session, then continue with related talks or explore the current programme.
Talk in brief
A reflection on the relationship between people and increasingly autonomous smart products. The talk argues that design now operates inside socio-technical systems where algorithms influence work, relationships, and everyday decisions. Trust cannot depend on a single acceptance screen: consent should be contextual and revisable, power should be visible, and intelligent behavior should remain understandable. Voice interfaces show both the promise and danger of product personalities. A healthier future treats the interaction as an evolving relationship in which customers can discover capabilities, question decisions, and change boundaries over time.
Key takeaways
- 01
Design relationships rather than isolated interactions
Smart products participate in social and technical systems, so their effects extend beyond a single task or interface state.
Watch from 3:28 - 02
Let trust grow through understandable behavior
Autonomous systems need predictable feedback and explanations because customers cannot inspect every model or automated decision.
Watch from 6:18 - 03
Replace blanket consent with contextual choices
Permission requested at the moment of use is easier to understand and revise than one broad agreement during installation.
Watch from 8:55 - 04
Make changing power relationships visible
Customers should know what the system can do next, what information it uses, and how they can limit or reverse that action.
Watch from 11:39 - 05
Use conversation to improve discoverability
Relational interaction styles can reveal smart capabilities gradually while giving people a natural way to ask and correct.
Watch from 17:28
Video chapters
- 1:03Stepping back from everyday design production
The opening invites designers to examine the longer-term role their products play in people’s lives.
- 3:28Life inside complex socio-technical systems
Algorithms and connected products are framed as active parts of work, hobbies, and relationships.
- 6:18Trust under automation and machine learning
Increasing autonomy makes gradual understanding and predictable system behavior more important.
- 8:55The limits of one-time blanket consent
Lengthy terms and broad installation permissions fail to support informed, continuing choices.
- 11:39Contextual permission and visible next steps
The design focus moves from accepting everything to understanding one relevant action at a time.
- 14:42Voice, personality, and experience aesthetics
Conversational products force teams to consider tone, identity, and the expectations created by human-like behavior.
- 17:28Relational interaction as future discovery
Older HCI models offer ways for smart products to reveal capability while customers retain agency.
Read edited transcript highlights
These concise notes were edited from automatic captions and checked against the talk structure. They are not a verbatim transcript.
Smart products participate in social systems
Technology is no longer a separate tool used after a social decision. Algorithms shape what people see, who they contact, how work is organized, and which opportunities appear. Designers should therefore examine the network around an interaction: who benefits, who is affected, what feedback is available, and how the system changes behavior over time.
Watch from 3:28Trust develops through repeated understandable encounters
People cannot directly inspect the reasoning of every machine-learning model or autonomous agent. They judge the system through outcomes, explanations, and consistency. A trustworthy product communicates uncertainty, shows relevant causes where possible, and behaves predictably enough that customers can form an accurate mental model rather than relying on a friendly surface.
Watch from 6:18Consent should match the moment and purpose
A single installation screen often asks for location, microphone, storage, and broad terms before the customer understands why they are needed. Contextual consent asks for a specific permission when its benefit and consequence are visible. The choice should remain reversible, and refusal should not silently remove unrelated value from the product.
Watch from 8:55Agency depends on knowing what happens next
In a relationship with an intelligent product, power changes as the system learns, predicts, and acts. Customers need to see the next proposed action, the information behind it, and the available boundary. This makes automation negotiable rather than mysterious and gives people a practical route to correct or stop behavior that no longer serves them.
Watch from 11:39Conversation can reveal capability without pretending humanity
Relational interaction styles can help customers discover what a smart product can do and ask for clarification in familiar language. The design should not imply emotions or understanding the system does not possess. A useful conversational character supports orientation and correction while remaining honest about the product’s limits and automated nature.
Watch from 17:28

