Most players find out the exit layer exists the wrong way. They try to withdraw something, it doesn't move the way they expected, and then the reverse engineering starts. What's RORS. Why does staking position affect this. What is my Trust Score and why does it matter right now when it didn't matter five minutes ago.
That sequence of discovery is the problem. Not the system itself.
The three layer architecture is genuinely well thought out. RORS constrains what qualifies as transferable value in the first place. Staking routes where that value sits inside the ecosystem before it can move anywhere. Trust Score sits at the exit and decides whether it leaves with you at all. Each layer is doing real work. Together they create a withdrawal environment that's more protected than almost anything else running in Web3 gaming right now.
But protection that's invisible until you need it isn't really protection from the player's perspective. It's a surprise. And surprises at the withdrawal layer are where player trust breaks hardest because that's the moment real money is involved.
The enterprise problem is the same thing at a different scale. External studios evaluating Stacked as an anti-cheat integration aren't going to take behavioral accuracy on faith. They have procurement processes specifically built to stress test vendor claims. Right now the only verifiable evidence that Stacked works is that it works inside Pixels farming. One game. One interaction pattern. One behavioral baseline.
A combat game has completely different human behavior rhythms than a farming loop. A strategy title has different tells. The models trained on Pixels harvesting cycles are not automatically accurate when dropped into a real time action environment and any enterprise studio technical team is going to flag that gap immediately.
What closes it is published cross-genre benchmark data before it gets asked for in a sales meeting. True positive rates across different game types. False positive rates. Edge cases where the model needed adjustment. That package isn't a weakness disclosure. It's the only evidence format that enterprise procurement actually respects.
The internal transparency case points the same direction. Players don't know specifically what behavior feeds their Trust Score, how individual actions are weighted, or how close they are to any threshold that changes their withdrawal experience. That information gap between what Stacked knows about a player and what the player knows about themselves is where quiet frustration builds into public criticism.
Showing players their own behavioral data doesn't break the anti-cheat logic. Bots don't self-correct because they see a dashboard. Humans do. And humans adjusting behavior in response to visible feedback is exactly what the system should want.
The proof gap is the same problem whether you're talking to a player who just hit a withdrawal wall or a studio deciding whether to integrate. Both need Pixels to show its work. Neither has enough right now.