What is HAX (Human-AI Experience) design?

HAX (Human-AI Experience) design is the discipline of designing AI-powered products that people understand, trust, and choose to come back to. hax.academy teaches it through two paired artifacts: the 7 Sins of AI Product Design (what to avoid) and the 6 HAX Principles (what to do instead).

What are the 7 Sins of AI Product Design?

Neglect, Obscurity, Gluttony, Mediocrity, Vanity, Tyranny, and Envy. Each sin names a common failure pattern in AI products — from bolting AI onto features that did not need it, to copying the chat-box pattern when the product needed something else entirely.

What are the 6 HAX Principles?

Empathy First; Automation vs. Augmentation; Transparency & Trust; Control & Reversibility; Feedback & Learning; and Responsibility & Safety. Together they map to the AI user journey and turn HAX from theory into shippable product decisions.

Who teaches at hax.academy?

Christian Kuhn — author, educator, and speaker with 20+ years of UX fieldwork across Pharma, Healthcare, Fintech, and Logistics. He also runs Optimizer.pt AI Literacy and TheStarter.io, and speaks at conferences including Harvard University DCE, webinale Web Week, Code Talks Hamburg, and AI Connect Riga.

What formats does hax.academy offer?

Workshops (90 minutes to full-day, 09:00–17:00), a blended course (16h live + 8h recorded + 6h project work), keynote talks, webinars, recorded videos and podcasts, and 1:1 consulting engagements for teams shipping AI products.

In which languages is hax.academy content available?

English by default; German on request for workshops and courses.

Does human-centered AI design actually improve ROI?

Yes. Correctly applied human-centered AI design (HAX principles plus Human-in-the-Loop governance) can generate a median ROI of +372%. Companies that allocate 25% or more of their AI budget to training achieve a 2.4× higher median ROI (+442%) than those with no training investment. Source: Denis Atlan, AI ROI Analysis: Evidence from 200 B2B Deployments (2022–2025), https://doi.org/10.5281/zenodo.17795133.

How do I book a workshop, course, or consulting engagement?

Use the Request Consulting form on hax.academy or email ck@hax.academy directly. Engagements typically start with a scoping call to align on outcomes, audience, and format.

hax.academy

Human-AI Experience Design

Learn: Human-AI Experience (HAX) Design

We teach designers, product managers, and developers how to ship AI products people understand, trust, and choose to come back to, through the HAX framework and AI Product Design Patterns: workshops, courses, and consulting built on twenty years of fieldwork.

Explore the courses

Why HAX

We have regressed to the command line.

0:00 / 0:00

Parts of this video are AI-generated. Can you spot which?

In twenty years of interface work, one lesson holds: the best interfaces disappear. You don't think about them, you just do things.

AI products broke that. The command line is back in the foreground, only now the parser gets moody and invents things. User tests keep proving the same point: the UX is not good.

Tool fluency is not product design fluency. Prompting is an afternoon of self-teaching. Designing AI products people understand, trust, and return to is the work nobody teaches well. Most AI failures are choice-of-application failures, not execution failures. Done right, human-centered AI design (HAX principles plus Human-in-the-Loop governance) returns a median ROI of +372%, and teams putting 25% or more of their AI budget into training see 2.4× higher median ROI.

hax.academy exists to fix that.

The framework

Name the disease. Ship the cure.

One diagnostic lens, six load-bearing principles, and a five-phase map that tells you where each decision lives.

The 7 Sins of AI Product Design

Neglect — designing AI as an afterthought, bolted onto a product that did not need it

Obscurity — building features users cannot find, cannot understand, or cannot trust

Gluttony — eating data, attention, and screen real estate without giving proportional value

Mediocrity — shipping outputs that are fine, when the alternative is doing nothing

Vanity — designing for the demo, not the daily use

Tyranny — removing user control in the name of “smart” defaults

Envy — copying the dominant chat-box pattern when the product needed something else entirely

A diagnostic frame, inspired by Kirstin Berman's Irrational Labs. Most AI products you can name commit at least three.

The 6 HAX Principles

Empathy First

Design for the user's real intent, not the AI's capabilities.

Try-out learning path

Automation vs. Augmentation

Choose deliberately, every time.

Grammarly shows suggestions side by side and lets you pick — you stay the author.

Transparency & Confidence

Calibrate trust, don't manufacture it.

Google Flights doesn't say “73.4% confidence.” It says “prices are unlikely to drop.”

Real Control & Editability

AI output should never feel final.

Canva lets you apply an AI effect and revert it in one click.

Graceful Failure

AI fails differently than software fails; design for it.

Replit AI admits the bug, shows what it's trying, and lets you step in.

Mental Model Shaping

People don't have a model for AI yet; you give them one.

Apple Photos delivers a whole mental model in three placeholder words: “Photos, People, Places…”

Curated from the public canon: Microsoft HAX Toolkit, Google People + AI Guidebook, Apple HIG, GitHub, the Shape of AI, Build for Mars (Peter Ramsey's UX Research library), Real Life AI Design Pattern from aiverse.design, and pressure-tested with international client work by Optimizer.pt, and Senior designer/developer alumni.

The AI Product Journey

Every principle has an address on the journey and will be demonstrated with Real-World Examples and trained with realistic exercises (Miro + Design-Tool + AI-Tools).

How to design AI products

For anyone new to the field, here is what the work actually covers. The workshop trains each of these dimensions through real product examples and hands-on exercises.

UI patterns

AI features need their own interface grammar: suggestion chips, confidence indicators, compare modes, inline drafts, and undo surfaces. We map each pattern to the user decision it supports.

Input types

Voice, text, image, document, gesture, or sensor feed. The right input depends on context, error cost, and user intent. We choose inputs deliberately, not default to the chat box.

Flows

AI changes the sequence of a task. We design when the system acts, when it asks, when it hands back control, and what happens when the output is wrong.

Language, wording & labels

Words shape trust. We replace vague AI labels with specific promises, replace confidence scores with plain-language likelihoods, and write error messages that help users recover.

Scaffolding

Onboarding, empty states, examples, and progressive disclosure teach users how to work with the AI without reading a manual.

Behavior, not just screens

The visible UI is only half the product. We design the underlying behavior: when the model runs, what it returns, how it fails, and how it explains itself.

These dimensions are drawn from the public canon: Microsoft HAX Toolkit, Google People + AI Guidebook, Apple HIG, the Shape of AI, aiverse.design, and real-world AI UX audits from Optimizer client work.

AI Product Design Patterns

We are using Real-World AI UX examples. Screencast Videos and UI Screenshots from our Partner aiverse.design and our own Collection and Use-Cases. Additionally we offering "Learning Pathways" curated with Peter Ramsey's UX research library: Build for Mars.

Vizcom — expressive input. Source: aiverse.design

Learn it. Apply it. Ship it.

Choose your depth.

In person · full day

The Full-Day Workshop

Six load-bearing principles. One full day. Your product on the table. Audit a real AI product with the 7 Sins, redesign it with the 6 HAX Principles, and prototype the unhappy paths.

09:00–17:00 · groups of 4–5 · you leave with an audit, a redesign brief, and a working prototype

Next workshop
→

Online · live · 2 weeks

The 2-Week Live Course

Designing Human-Centered AI Products. Four live lessons across two weeks, the fastest route from prompt-dependency to a product-ready HAX practice. Limited Special Price for Future Product Days Conference Ticket Holder.

4 × 2-hour live sessions · recordings included · certificate

See the live course →

Online · hands-on · 4 weeks

The 4-Week Intensive

Design Patterns for Human-Centered AI Experiences. Thirty hours across four weeks with live classes, recorded AI Pattern- and Use case deep-dives, and a group project that runs the full framework end to end.

16h live + 8h recorded + 6h project work · certificate · taught in English. German on request

See the intensive →

For teams & events

Webinars

The framework, delivered to your team or your conference, from the 20-minute conference format to the two-hour deep dive with Q&A.

remote · English or German · tailored to your product context

On demand

Videos & Podcasts

Recorded talks and discussions, watch or listen the thinking before you buy the training.

free · new recordings added after each conference season

Watch on Youtube →Listen on Spotify →

Consulting

Work with us directly

A 7 Sins audit of your AI product, a HAX redesign brief your team can ship, or hands-on enablement inside your org, built on our fieldwork in Pharma, Healthcare, Fintech, and Logistics.

Engagements start with a 30-minute scoping call

Where to find us

The next six months.

Jul 28, 2026
Guest lecture — Dave Platt's UX Engineering course, Harvard University DCE
Online
Details →
Aug 22, 2026
Talk & 90-minute workshop — AI Connect
Riga, Latvia
Details →
Sep 22, 2026
90-minute workshop — Future Product Days
Copenhagen
Details →
Oct 6–29, 2026
The 4-Week Intensive — TheStarter
Online, English
Details →
Oct 26–30, 2026
Talks + full-day workshop — webinale Agentic Web Week
Munich
Details →
Nov 4–5, 2026
Talk — Code Talks Conference
Hamburg
Details →
Nov 9–18, 2026
The 2-Week Live Course — Future Product Days
Online, Mon + Wed 18:00–20:00 CET
Details →
Feb 24–25, 2027
Talk & full-day workshop — prompt:UX
Berlin
Details →

Want a private date for your team?

What practitioners say

Rated by people who ship AI for a living.

5.0 / 5

Conference average: 4.34
Webinale, Berlin 2026

4.67 / 5

“Don't Make Me Prompt”,
DevSum, Stockholm 2026

4.64 / 5

Nordic UX Aarhus,
Denmark 2026

“The frameworks you built around trust, transparency, control, and graceful failure … are now part of how I evaluate every AI decision I encounter at PepsiCo.”

Nathy Tanner

PepsiCo

1 / 4

Taught and spoken at

Harvard University DCEAI Connect.lvPorto Tech HubFuture Product DaysWebinale Web WeekDevSumNordic UXGerman UPATheStarter.ioISEP PortoWe.Test AthensCode<>TalksPortoAX Optimizer.pt AI Literacy

Who is behind hax.academy

Christian Kuhn

Bridging Human Needs and AI Capability through Behaviorally Informed UX Design

Christian has over 20 years of international UX experience and currently leads the UX Center of Competence at Optimizer in Porto, Portugal. He drives the development of award-winning, user-centered product solutions through UX research and Behavior informed design. He contributes to Harvard University DCE, Singularity University, Nomura Research Institute Japan, and many more. He is a published author on UX Design, AI Product Design and Behavior Design topics. He is teaching UX Leadership, Behavior Design and Human-Centered AI Experiences (HAX) at TheStarter.io. PortoAX Activist. Jury Member of the UX Nordic Award 2026.