The story behind the cartoon
OpenAI published an updated Model Spec on October 29, reorganizing its public description of how its models should behave around five values: faithfulness to human intent, harm minimization, honesty and transparency, human agency, and care.
The company is explicit about the gap: the Spec describes behavior it is working toward, and its models do not yet fully reflect it. It is exploring new training methods to teach models the values, not just the rules.
The admission is the story: the rulebook is published, the five values are named and checked off — and the models themselves are still catching up to their own instruction manual.
The joke is the rulebook-vs-reality beat: a smiling robot holds up the enormous MODEL SPEC: 5 VALUES handbook, every value checked, while its other hand works red marionette strings on a furious tiny human — human agency, tabbed and checked, visibly violated.

