Product teardown · behavioral AI

Rock, Paper, Scissors and Behavioral Patterns

How systems learn a quirk you don't even know you have — and why that's a gold mine.

One day, while ordering from Subway through a food-delivery app, I noticed something odd: I couldn't find my usual coupon.

I have a habit. For any order above ₹300, I check the price across different apps and pick the best one. But when I'm in the office I usually don't bother — even if no discount coupon is available, I still place the order, because it's about time. For the first few days I saw the coupon. After a few days, it stopped appearing. Strangely, that same week, when I was working from home, the coupon showed up again.

Were they really acting on my behavior pattern? Maybe I was wrong; maybe I was mistaken. But it kept me thinking.

9:41▢ ▢ ▮
SubwayCart
At the office · you don't compare
Veggie Delite + Cookie₹420
Apply couponNo offers available
To pay₹420
Predictable buyer → coupon withheld
9:41▢ ▢ ▮
SubwayCart
Working from home · you compare
Veggie Delite + Cookie₹420
Coupon SAVE60 applied−₹60
To pay₹360
Comparison-shopper → coupon shown

01A different kind of pattern

A behavior pattern is different from a user-segmented flow — for example, iOS users being shown a higher price. The key difference is that this is a quirk or pattern the user themselves isn't aware of, and a system identified it.

Take the famous market-basket analysis: at Walmart, shoppers tend to buy beer and baby diapers together on Fridays, and placing them side by side lifts sales. But that's a system spotting a behavior the customer already knows they do. The interesting question is the other one — what about a quirk the user probably isn't even aware of?

02The game: Rock, Paper, Scissors

To explore this, I built the most famous game of all — Rock, Paper, Scissors — and made the camera read your real hand, not a button (a button only trains the system on your clicking habits).

Play it yourself
The Read — Camera Duel

Show a real Rock, Paper or Scissors to your camera. Play a few games and watch the AI's win rate climb as it learns your quirk.

▶ Launch the game
Try to fight it
If you want to combat the system, you can deliberately show different behavior. I tried it — it was fun, and it worked, but it just stretched the game out (it took maybe 15–20 matches to adapt). The catch: when we order food or move through an app, we don't fight the system. Our normal behavior simply continues — and the more we use it, the more it sees.
Objective
The game learns a quirk you might not be aware of — and for B2C companies, that data is a gold mine.

03Where this goes — a light map

The coupon I saw (and didn't see) is just the entry point. The same idea — read a quirk you can't perceive, then ship a personalized "delta" in the flow — scales in three tiers.

Tier 1 Happening now

Offers tuned to how you behave, not who you are — your hesitation, scroll speed, what you abandon in the cart. Coupons withheld from people predicted to buy anyway; nudges timed to a detected mood. (My Subway coupon lives here.)

Tier 2 Near future

The flow builds itself around you — AI-generated screens unique to each person, and a reinforcement-learning loop (exactly like this game) that discovers the nudge sequence that converts you. Friction made as hard as you'll personally tolerate.

Tier 3 The frontier

The board flips. When your own AI agent shops for you, sellers can't read you directly — so they target the agent instead, and predict life events (a move, a new job) to pre-position the offer before you've decided.

9:41▢ ▢ ▮
Checkout
Flight · BLR → GOA₹4,120
Add refundable + insurance →
Continue with basic fare
Highlighted path+₹640
Tier 1 · the cross-sell glows, the basic option fades
9:41▢ ▢ ▮
For yougenerated
layout tuned to your taps
Add ₹49 ₹9950% off
Add ₹39 ₹79just for you
Compare priceshidden
Pay now₹508
Tier 2 · the screen itself is built to your weak spots
9:41▢ ▢ ▮
Your agent
You: order my usual, cheapest option
App A · ₹420no coupon
App B · ₹398+fees
App C · ₹360✓ chosen
Agent paid₹360
Tier 3 · your agent is unreadable — so they target it

04How the game learns your quirk — technical spec

The system keeps several running models of you and updates all of them on every throw. Each one predicts your next move from a different angle:

favorites · your most-picked shape follow-ups · after your last 1 / 2 / 3 moves reaction · to the AI's last move win-stay / lose-shift

For every prediction it considers three layers of second-guessing — counter you, counter your counter, counter that — so it still wins when you try to outsmart it. It scores every model-and-layer combination on how well it has predicted you recently (across a fast and a slow memory) and plays whichever is winning. A pure-random strategy sits in the pool as a floor: the moment you crack its pattern, it falls back to unbeatable randomness. Everything is stored on the device and carries across games, so the read of you keeps sharpening.

predictable play → AI wins 90–100%
random play → ~33%, unbeatable

That gap is the entire point. Predictability is exploitability. The more legible your behavior, the more the system beats you — and in a real store, that "win" is your wallet. The defense in the game is the same as in the app: be aware, and don't be readable.

The Read · behavioral-pattern case study predictability = exploitability