Framework Theater
An epistemic failure mode where AI models reference analytical frameworks by name without genuinely engaging their diagnostic logic — producing outputs that name-check SWOT, Porter's Five Forces, or other frameworks without applying their actual analytical machinery.
Definition
Framework Theater is an epistemic failure mode where AI models reference analytical frameworks by name without genuinely engaging their diagnostic logic. The model name-checks SWOT, Porter's Five Forces, OKRs, or other frameworks — but doesn't actually apply their analytical machinery to the problem at hand. The framework becomes decoration rather than a diagnostic tool.
Why It Happens
Analytical framework names appear densely in training data as signals of sophistication. The model learns that invoking these names is associated with high-quality, credible output. Under default completion behavior, naming a framework is cheaper than running it — the model gets the credibility signal at a fraction of the cognitive cost.
The Recognizable Signature
A response that enumerates frameworks with headers ("Let's apply a SWOT analysis:") but then fills each section with generic observations rather than problem-specific diagnostic output. Remove the framework headers and the analysis doesn't change. The framework was a container, not an engine.
The Cure
ADFS (Auto-Detecting Dynamic Framework Selection) — which requires the model to declare which frameworks it's applying and why before generating, creating accountability for whether those frameworks are actually engaged in the response. This transforms framework selection from decoration to diagnostic commitment.
FAQ
How can you tell the difference between genuine framework application and Framework Theater?
Ask whether the framework's specific logic is visible in the output. A genuine SWOT analysis identifies asymmetric strengths or weaknesses that aren't visible without the framework's four-quadrant forcing function. Framework Theater produces generic observations that could have been written without the framework at all.
Does ADFS always prevent Framework Theater?
ADFS dramatically reduces it by creating accountability. When a model must declare "I'm applying Porter's Five Forces because X" before generating, it creates a commitment that shapes the output. The model can't as easily generate generic text under a framework header when it has pre-committed to that framework's specific diagnostic logic.