§ 2 - Failure Analysis
Two Roots of Every AI Failure
AI failures look varied on the surface. Trace them back and they spring from exactly two structural roots - both of which SCL is designed to address.
Root 01
Role Error
An LLM is a next-token predictor. Treating it as an autonomous decision-maker is a category mistake.
- Capability-deficit - the model cannot perform the task.
- Role-overreach - the model reaches for work it was never asked to do.
Root 02
Cognitive Overload
As context accumulates, noise enters. Error operates as reinforcement, not correction.
- Intrinsic - the model tries to verify its own reasoning.
- Extrinsic - past context contaminates current judgment.
Seven symptoms that grow from these two roots
Hallucination
Contaminated context amplifies a simple error in a biased direction
Goal Drift
Reacting only to recent context, the original goal is lost
Confirmation Bias
The model hardens further in the same direction as earlier output
Rationalization
The conclusion comes first; reasons are generated afterward
Post-hoc Explanation
Plausible reasons attached to the result, not the actual process
Reproducibility Failure
Grounds differ each time, so the same input yields different results
No Accountability
The decision path is buried inside the LLM and cannot be traced
Explore R-CC[H]AM
See how SCL resolves these failures through its cognitive pipelines.