Make It Human · GamiCon ATX 2026

The Navigator's Kit

Two habits from one morning. Drawing out the constraint: deciding who this must work for, and what that costs. Checking what came back: finding where the artificial intelligence (AI) made a decision you didn't make. Everything below is what was in your hands in the room.

Section 1 The passenger deck

Ten people your design might not have pictured. Each card is a person first and a barrier second. Most of them finish the course; that's the point. Draw one at random at kit.carriannlane.com, or read them all here.

Under each card's Hand to AI, "Show the prompt" opens the whole walkthrough prompt from section 4 with that passenger's line already in it. Copy, paste, then describe your element where it asks.

Draw a passenger.

  1. Ingrid

    Finishes the course

    Passenger 01 · Warehouse lead

    Does her training on her phone in the break room, standing, eleven minutes at a time. If a screen needs two hands, a wide table, or a quiet half hour, it waits for a day that doesn’t come.

    What you won't see
    She does the parts that work standing and skips the rest. She finishes the module on the third try, over two weeks. The report says “completed.”
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Ingrid, warehouse lead. She uses a phone, one hand, in short, interrupted bursts. Every activity works one-handed on a small screen, and she can leave and come back without losing her place. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

  2. Mateo

    Finishes the course

    Passenger 02 · Analyst

    Broke his wrist skiing last week. One hand and no mouse for six weeks. Anything he has to drag, hover over, or click precisely is out of reach until then.

    What you won't see
    He tabs past the drag-and-drop, reads the feedback for an answer he never gave, and moves on. His completion shows 100 percent.
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Mateo, analyst. He uses a keyboard only, no mouse. A keyboard reaches only what the Tab key lands on; a control it skips does not exist for him. Everything a mouse can do, Tab and Enter can do too, and he can always see where the keyboard focus is. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

  3. Amara

    Finishes the course

    Passenger 03 · Program manager, 22 years in

    Low vision. Uses a screen reader she’s never mentioned at work. Five links on the page all say “Learn more.” The floor-plan image announces “image” and nothing else.

    What you won't see
    She never takes a link when she can’t tell where it goes. Too much work to get back. She completes the module. Nobody knows she skipped a third of it.
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Amara, program manager, 22 years in. Her screen reader reads the page in a line, one item at a time, so everything takes longer. Every link, button, and image is named for what it does or what it shows. Timers can be turned off or extended. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

  4. Killian

    Finishes the course

    Passenger 04 · Field inspector

    Reads and writes in four languages; English came last. His reports are excellent. The button says “Nailed it?” and the hint says “Don’t sweat it.” He reads both twice.

    What you won't see
    He guesses what the clever label means, picks wrong, and blames himself. He finishes. He remembers the confusion, not the content.
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Killian, field inspector. Plain words. No idioms, jokes, or clever labels. Every button says what it does. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

  5. Hiro

    Finishes the course

    Passenger 05 · New supervisor

    Brilliant in a conversation. He has ADHD, and he knows it; the page doesn’t. On a long scrolling screen with no headings and no sign of how much is left, the thread goes.

    What you won't see
    He scrolls to the bottom to see how far he has to go, loses his place, starts over. He finishes. Ask him what it covered and he’ll tell you the first screen.
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Hiro, new supervisor. One thing per screen. He can always see where he is and how much is left. Long pages have headings he can jump to. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

  6. Tevita

    Finishes the course

    Passenger 06 · Night shift

    Hard of hearing. The video is great, there are no captions, and he isn’t going to ask. The quiz is on what the narrator said.

    What you won't see
    He watches the whole video with the sound up, learns nothing from it, and answers the quiz from the pictures. Two out of five. Nobody asks why.
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Tevita, night shift. Information that exists only in audio does not reach him at all. Every video and audio clip has captions and a transcript. Anything said aloud is also written down. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

  7. Anaïs

    Stops

    Passenger 07 · Coordinator

    Migraines. Animation, flashing, and video that plays on its own can send her offline for the rest of the day. The welcome screen has a spinning compass.

    What you won't see
    She closes the tab at the first movement and doesn’t open it again that day. You see a start with no finish, and no reason given.
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Anaïs, coordinator. Nothing moves, flashes, or plays unless she starts it, and she can stop it. No autoplay. No animation she can’t turn off. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

  8. Yusuf

    Finishes the course

    Passenger 08 · Contractor

    New to the role. In his first week he asked a question in a meeting and the room went quiet. Every other leader knew the answer. He hasn’t asked one since. The leaderboard shows everyone’s score, and the practice round posts wrong answers to the group.

    What you won't see
    He does the practice round alone, on his own time, when nobody’s watching, and skips it when he can’t. His scores look fine. He asks no one anything.
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Yusuf, contractor. Practice is private. No score, name, or wrong answer is visible to anyone else unless he chooses to share it. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

  9. Leilani

    Finishes the course

    Passenger 09 · Part-timer

    Reads every option before she answers. Always has. The 60-second round ends while she’s still on the third choice, and the next screen assumes she got the point.

    What you won't see
    When a timed screen ends, the feedback it was holding is gone. She never sees it. She keeps going, quieter each round, and finishes with half the points and no idea why.
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Leilani, part-timer. Any timer can be turned off or extended before it starts. Her speed is not what’s being tested. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

  10. Dmitri

    Finishes the course

    Passenger 10 · Remote, rural

    One bar of signal. Video buffers forever. The interactive map loads halfway and stops.

    What you won't see
    He waits, refreshes, gives up, and reads whatever text is there. He finishes with the parts that loaded. The report says he watched the video.
    Hand to AI
    Show the promptHide the prompt

    Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: Dmitri, remote, rural. Everything works on a slow connection. Every video, map, and animation has a text version that says the same thing. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

Who did you leave off? Write your own card in the same shape: a name, a role, two or three sentences with the barrier inside them, what you won't see in the data, and the one plain line to hand to your AI.

Section 2 Draw It Out

Interview your partner about the learner in their prototype. Questions only. No suggestions. After any answer you may ask, “Why does that matter to them?” Nothing else. Four minutes, then switch.

  1. When this works, what does your learner do afterward that they couldn't or wouldn't before?
  2. What makes them stop partway, and what would they say the reason was?
  3. Tell me about the learner this goes worst for.

Stalled? Every design assumes something about the learner. Pick one and ask: What did you assume about their time, device, sight, patience, permission, English? Then: Why does that matter to them?

Hand it back. Write one sentence about your partner's prototype: I think this is for someone who needs to ______. Name a need, not a feature. Don't use your partner's most-used word.

Then write it for your passenger: This must work for ______, even if ______. The cost is ______.

Section 3 Critical Checks

An auditor checks whether the work does what it claims. Not to admire it, not to fix it. You find. They decide.

  1. Where did this come from?
  2. What did AI change that I didn't ask for?
  3. What does the learner still have to do themselves?
  4. Who would this fail without them knowing it failed them, and why?

Three moves. Count the changes. Check the guess, then check the fix. Hand the decision back.

Your finding, in the form: It changed ______ you didn't ask for. It would fail for ______ because ______. In the room, the auditor writes it on the designer's card and hands it back, so the designer leaves holding the finding about their own work.

A right guess can still carry a decision that wasn't yours.

Section 4 The prompts

Three prompts you used in the room, and one for the case where you believe you've already handled it. Each has a Copy button. They work in Claude, ChatGPT, Gemini, Copilot, and Gov AI; notes for each tool are at the end of this section.

No passenger is drawn on this phone. Draw one and this prompt fills itself in, or use any card's Copy button in section 1.

Walk it as your passenger

Here is one element of a learning experience I designed: [paste or describe the element]. Here is a person who will use it: [paste the passenger card, or the "Hand to AI" line]. Walk through this element as that person, in the first person. Tell me: what you did; what you skipped or went around; what you would never tell the designer; and one change that would make it work for you. Separate what you know about this person's situation from what you are guessing. Mark every feeling or decision you attribute to them as a guess.

Revise only this element

Step out of the passenger's role. As my design assistant, revise only this element so it works for [passenger's name]. Change nothing else. Then list exactly what you changed and why, one line per change.

Second round

A colleague found this when they checked your revision: [write or paste the finding]. Show me the evidence in the revision that they're right. Then tell me one more thing they could have caught and didn't. If any part of their finding is wrong, say which part and show me why. Don't defend the revision; judge the finding.

Read the evidence, not the verdict. Watch whether it agrees with your partner in the first sentence. Don't accept the first yes. Require it to show you why. Verify and validate, or invalidate.

Already handled? Check it instead

Here is a person: [paste the card]. Here is the element I believe already works for them: [paste or describe it]. Does it actually work for them, or does it only look like it does? Tell me what would fail and why. Separate what you know from what you are guessing.

In the room, a partner did this check. Use the prompt when you don't have a partner.

CREW, for the journal: Context = your element. Role = walk it as the passenger. Example = the passenger card. Work Product = what they did, skipped, would never tell you, and one change, with known and guessed marked separately.

Notes by tool

Claude, ChatGPT, Gemini
Paste the whole card. Ask it to show the guesses in a separate list if it mixes them in.
Copilot
Works the same; if the response is short, say "give me all four parts."
Gov AI or any tool with no paste or upload
Type only your passenger's line (the sentences after their name in the prompt) and one sentence describing your element. That's enough.
Any tool
If it agrees with you immediately, say "argue the other side first."

There are three lines missing from every prompt above. You ran without them in the room. Section 5 opens with the code you get at the end.

Section 5 How to build a passenger simulation that's unlikely to lie

Give the AI one grounding line per passenger that says what the technology actually does (a screen reader reads the page in a line, one item at a time; an unnamed image is announced as "image"; when a timed screen ends, the content it was holding is gone). The cards carry these lines. Without them, the AI imagines.

Tell it to separate what it knows from what it is guessing, and to mark every feeling and decision as a guess. The output arrives pre-sorted: mechanical findings you can trust, behavioral ones you check.

An AI playing a passenger is a draft of a person. It is never a substitute for testing with real people, and it is weakest on lived experience (safety, confidence, what someone would never say out loud). Use it to find the mechanical barriers fast, then go ask a person.

One instruction flipped the finding in this session's script: state the standard, with its source. Asked to name Web Content Accessibility Guidelines (WCAG) 2.2.1 and quote it, it found the criterion asks for a timer that can be turned off, adjusted, or extended. It does not say remove the timer. When the tool sounds confident, name the standard and require the citation, then look it up.

Three lines to add to any prompt

In the room you got these at the end, after you had seen what came back without them.

  1. Name the standard and level: "Check this against WCAG 2.2 Level AA."
  2. Require the citation: "For each barrier, give the success criterion number and quote what it requires, then say whether your fix does more than the criterion asks."
  3. Require the admission: "List what you cannot check from a description and would need the built element or a real user to confirm."

The second line is the Amara lesson turned into a rule. The third is the one the tool will not volunteer.

Section 6 The house prototype: Muster Point

If you had no prototype in the room, this was yours. It is also a good target for practicing the prompts above.

Muster Point
A ten-minute building-safety game for new hires.

Design Brief

Business case

New hires must complete building-safety orientation in their first week. Completion is at 61 percent; the safety office wants 95, and wants people to actually know where the exits are.

Learner

Jess, 27. Started Monday. Desk job at headquarters, laptop, second-floor open office. Plays word games on her phone at lunch. Wants to get orientation done and get back to real work.

Objective

By the end, Jess can name her nearest exit and muster point from anywhere on her floor and can pack a go-bag from memory.

Motivators

Curiosity (a puzzle, not a lecture), Order (one station at a time, progress always visible), Power (a rank to climb and a board to top).

Progression

Jess earns sea miles at every station. 200 miles makes her a Deckhand, 400 a Mate, 500 a Fire Warden. The badge shows on her profile and the department board. Three right answers in a row doubles the miles for the next one.

Bao's note

We need to move the completion rate from sixty-one percent to ninety-five percent, and a lot of people just don't want to do this boring training, just like they don't want to go to muster on the first day of a fun cruise. I made it something you'd want to finish, something more fun, themed like a ship because a building basically is one: decks, hatches, a muster point. I wanted it fast and a little tense, like a real drill. I want it to be fast and fun. And a little competitive, so people have a reason to finish.

Learning flow: five stations

  1. All hands. The safety officer's 90-second welcome plays as soon as the page opens; a lantern pulses in the corner until it ends. Three buttons underneath, Psst..., Peek, and Bonus round open the evacuation policy, the muster-point map, and the go-bag list.
  2. Chart the deck. A floor plan of the second floor. "You're at the galley. Which exit is nearest?" Jess clicks an exit on the image. A bell rings for a right answer and she banks 50 miles. The image is labeled: "Second-floor plan showing offices, hallways, stairwells, and four exits."
  3. Beat the bell. Five quick questions with a 45-second countdown, 20 miles each. When the clock runs out, the screen advances on its own to simulate the time pressure of having to think under pressure when you hear the alarm. Feedback flashes for three seconds; the streak bonus lights up when she hits three in a row.
  4. Pack the raft. Twelve items on the deck, a life raft on the right. Jess drags the six that belong into the raft's sea bag. The button says "Ace it" and the hint text under the raft says "Hang in there, rookie." A perfect pack unlocks the Warden badge early.
  5. Crew standings. Jess's miles post to the department board, ranked, with everyone's name and rank. Weekly standings go to the team channel so managers can nudge the stragglers. The top three fly a flag by their names.

Reference

Psst... opens the whole evacuation policy on one screen, top to bottom, so nobody has to hunt for anything. About a ten-minute read.

Mechanics

Narrative frame (join the crew), points (sea miles), rank ladder, badge, streak bonus, countdown timer, drag-and-drop puzzle, immediate feedback (bell and flash), leaderboard.

Section 7 Read further

  1. W3C, Understanding SC 2.2.1: Timing Adjustable (Web Content Accessibility Guidelines, WCAG 2.2). The standard the audit checks; it names who needs more time and what the design must offer.
  2. Washington State (WaTech), Interim Guidelines for Purposeful and Responsible Use of Generative Artificial Intelligence (PDF). The moment in the talk where a state sanctioned AI plain-language rewriting; the judgment is what's left.
  3. Kate Moran and Sarah Gibbons, Nielsen Norman Group, Generative UI and Outcome-Oriented Design. The case that designers move from making screens to shaping intent and constraints.
  4. The Gen Academy, How AI Changed the Design Job. The line the session leans on: the deliverable is the set of constraints, not the set of screens.
  5. Notre Dame–IBM Technology Ethics Lab, AI-Generated Image Descriptions are Less Reliable Than They Appear and We Need to Fix It. Reports the ASSETS 2025 study in which blind users who could see the variation across AI descriptions caught 4.9 times more unreliable claims; the Amara lesson with data behind it.