1. Problem
No tool maintains a model of what a team currently believes to be true — which is why every assistant that sends meeting notes and reminders eventually gets muted.
User
A PM at a mobile games studio running live ops (Tamarind Games, fictional — no real company data).
Trigger
A message asserts something about a referent — a launch date, a success metric, a scope decision — where a live claim from a different person already conflicts with it.
Outcome
The conflict surfaces with both source messages and the reasoning, before it reaches execution.
Grounded in, attributed, not asserted
- Olivier Courtemanche (former EA PM): misaligned definitions of success, specs executed without shared reasoning, contested ownership, and a “data truth trap” — two people reading the same metric to opposite conclusions.
- Kowalyshyn & Scheutz (Tufts): a taxonomy — unsupported beliefs, false beliefs, belief contradictions, omissions — used as a frame only; their reported percentages are lab-study numbers on a two-person task and are not cited as if they transferred to this domain.
Explicit assumption
No games-studio PM was interviewed to validate this. What would be validated first, given the chance: whether contradiction (what this system detects) or omission (what it deliberately does not) is the failure mode PMs actually notice first in practice.
Success criteria
| Metric | Measurable from this corpus? |
|---|---|
| Detection lead time | Yes |
| Caught before vs. after execution | Yes |
| Time to reconcile | Partially |
| Rework avoided | No — needs studio-supplied revenue/cost inputs |