Design For AI

Design For AI

Design For AI

Thinking through the design problems that AI products create.

Not how to use AI as a designer.

How to design for it.

Thinking through the design problems that AI products create.

Not how to use AI as a designer.

How to design for it.

Thinking through the design problems that AI products create.

Not how to use AI as a designer.

How to design for it.

About

How AI is rewriting the rules of UX.

How AI is rewriting the rules of UX.

Design challenges that AI products create. Not how to use AI as a tool, but what happens to UX when the system is non-deterministic, when trust has to be earned rather than assumed, and when the interface has to make invisible infrastructure feel inevitable.

Design challenges that AI products create. Not how to use AI as a tool, but what happens to UX when the system is non-deterministic, when trust has to be earned rather than assumed, and when the interface has to make invisible infrastructure feel inevitable.

One question throughout: what does it actually mean to design for AI?

One question throughout: what does it actually mean to design for AI?

The Articles

The problems. The patterns. The practice.

The problems. The patterns. The practice.

Each article explores a design challenge that AI products create.

Each article explores a design challenge that AI products create.

  • 2025

    Systems

    Trust

    Most AI products fail at the translation layer

    Most AI products fail at the same point. Not technically. The model works. But somewhere between what the system does and what the user understands, something breaks. That gap has a name.

    5 min read

  • 2025

    Interfaces

    Behavior

    Designing AI is not designing screens

    Most designers working on AI products are doing the same job they've always done. Mapping flows. Building components. Designing states. That work still matters. It's just not enough anymore.

    5 min read

  • 2025

    Behavior

    Trust

    The UX of uncertainty

    Uncertainty in an AI system is not an edge case. It's the baseline condition of every output the system produces. The question is never whether it exists. It's whether the design acknowledges it.

    5 min read

  • 2025

    Interfaces

    Sustems

    The best AI interface is the one users forget exists

    Invisible doesn't mean minimal. It means absent from the user's conscious attention. In AI products, earning that invisibility is harder than it has ever been. And more important.

    5 min read

  • 2025

    Agency

    Trust

    Designing the handoff between human and AI

    The most critical moment in an AI product is not when the user interacts with the system. It's when control changes hands. That moment is almost universally badly designed.

    5 min read

  • 2026

    Context

    Behavior

    Context is the new content

    In traditional products, content exists before the user arrives. In AI products, what the user sees depends on who they are and what they've done before. That changes everything about how design works.

    5 min read

  • 2026

    Onboarding

    Context

    What onboarding looks like when the product learns

    Classical onboarding teaches the user how to use the product. AI product onboarding has to do something harder: begin the process of the product learning the user. Most teams design only one side of this.

    5 min read

  • 2026

    Trust

    Strategy

    Trust is not a UX pattern. It's an architecture decision.

    Every design team working on an AI product has had the same conversation. The product feels untrustworthy. The solution proposed is always surface work. It's almost never the real problem.

    5 min read

  • 2026

    Behavior

    Agency

    Designing for the second decision

    Most design work is obsessed with the first interaction. In AI products, the decision that determines whether a user stays or leaves almost never happens first. It happens second.

    5 min read

  • 2026

    Systems

    Strategy

    Why the best AI products have the least visible AI

    The AI products people actually integrate into how they work have something in common. The AI in them is almost invisible. This is not a coincidence. It's a design consequence.

    5 min read

  • 2025

    Systems

    Trust

    Most AI products fail at the translation layer

    Most AI products fail at the same point. Not technically. The model works. But somewhere between what the system does and what the user understands, something breaks. That gap has a name.

    5 min read

  • 2025

    Interfaces

    Behavior

    Designing AI is not designing screens

    Most designers working on AI products are doing the same job they've always done. Mapping flows. Building components. Designing states. That work still matters. It's just not enough anymore.

    5 min read

  • 2025

    Behavior

    Trust

    The UX of uncertainty

    Uncertainty in an AI system is not an edge case. It's the baseline condition of every output the system produces. The question is never whether it exists. It's whether the design acknowledges it.

    5 min read

  • 2025

    Interfaces

    Sustems

    The best AI interface is the one users forget exists

    Invisible doesn't mean minimal. It means absent from the user's conscious attention. In AI products, earning that invisibility is harder than it has ever been. And more important.

    5 min read

  • 2025

    Agency

    Trust

    Designing the handoff between human and AI

    The most critical moment in an AI product is not when the user interacts with the system. It's when control changes hands. That moment is almost universally badly designed.

    5 min read

  • 2026

    Context

    Behavior

    Context is the new content

    In traditional products, content exists before the user arrives. In AI products, what the user sees depends on who they are and what they've done before. That changes everything about how design works.

    5 min read

  • 2026

    Onboarding

    Context

    What onboarding looks like when the product learns

    Classical onboarding teaches the user how to use the product. AI product onboarding has to do something harder: begin the process of the product learning the user. Most teams design only one side of this.

    5 min read

  • 2026

    Trust

    Strategy

    Trust is not a UX pattern. It's an architecture decision.

    Every design team working on an AI product has had the same conversation. The product feels untrustworthy. The solution proposed is always surface work. It's almost never the real problem.

    5 min read

  • 2026

    Behavior

    Agency

    Designing for the second decision

    Most design work is obsessed with the first interaction. In AI products, the decision that determines whether a user stays or leaves almost never happens first. It happens second.

    5 min read

  • 2026

    Systems

    Strategy

    Why the best AI products have the least visible AI

    The AI products people actually integrate into how they work have something in common. The AI in them is almost invisible. This is not a coincidence. It's a design consequence.

    5 min read

  • 2025

    Systems

    Trust

    Most AI products fail at the translation layer

    Most AI products fail at the same point. Not technically. The model works. But somewhere between what the system does and what the user understands, something breaks. That gap has a name.

    5 min read

  • 2025

    Interfaces

    Behavior

    Designing AI is not designing screens

    Most designers working on AI products are doing the same job they've always done. Mapping flows. Building components. Designing states. That work still matters. It's just not enough anymore.

    5 min read

  • 2025

    Behavior

    Trust

    The UX of uncertainty

    Uncertainty in an AI system is not an edge case. It's the baseline condition of every output the system produces. The question is never whether it exists. It's whether the design acknowledges it.

    5 min read

  • 2025

    Interfaces

    Sustems

    The best AI interface is the one users forget exists

    Invisible doesn't mean minimal. It means absent from the user's conscious attention. In AI products, earning that invisibility is harder than it has ever been. And more important.

    5 min read

  • 2025

    Agency

    Trust

    Designing the handoff between human and AI

    The most critical moment in an AI product is not when the user interacts with the system. It's when control changes hands. That moment is almost universally badly designed.

    5 min read

  • 2026

    Context

    Behavior

    Context is the new content

    In traditional products, content exists before the user arrives. In AI products, what the user sees depends on who they are and what they've done before. That changes everything about how design works.

    5 min read

  • 2026

    Onboarding

    Context

    What onboarding looks like when the product learns

    Classical onboarding teaches the user how to use the product. AI product onboarding has to do something harder: begin the process of the product learning the user. Most teams design only one side of this.

    5 min read

  • 2026

    Trust

    Strategy

    Trust is not a UX pattern. It's an architecture decision.

    Every design team working on an AI product has had the same conversation. The product feels untrustworthy. The solution proposed is always surface work. It's almost never the real problem.

    5 min read

  • 2026

    Behavior

    Agency

    Designing for the second decision

    Most design work is obsessed with the first interaction. In AI products, the decision that determines whether a user stays or leaves almost never happens first. It happens second.

    5 min read

  • 2026

    Systems

    Strategy

    Why the best AI products have the least visible AI

    The AI products people actually integrate into how they work have something in common. The AI in them is almost invisible. This is not a coincidence. It's a design consequence.

    5 min read

The Author

Written from practice,
not from observation.

Written from practice,
not from observation.

Every article here comes from a design decision I had to make in production. Not a case study, not a thought experiment. A real product, real users, real constraints. The more I design for AI, the more I realize how much of this territory is still unmapped.

Every article here comes from a design decision I had to make in production. Not a case study, not a thought experiment. A real product, real users, real constraints. The more I design for AI, the more I realize how much of this territory is still unmapped.

These articles are my attempt to map it.

These articles are my attempt to map it.

Design for AI

Thinking through the design problems that AI products create.

Not how to use AI as a designer. How to design for it.

© 2026 Design for AI. All rights reserved.

Design for AI

Thinking through the design problems that AI products create. Not how to use AI as a designer. How to design for it.

© 2026 Design for AI. All rights reserved.

Design for AI

Thinking through the design problems that AI products create. Not how to use AI as a designer. How to design for it.

© 2026 Design for AI. All rights reserved.

Create a free website with Framer, the website builder loved by startups, designers and agencies.