Synthetic Media & Provenance · Capstone brief

PROV-06 · Satire vs. deception

Draw the editorial line between labeled satire and misleading synthetic media, then test your line on three headlines that sit right on it. A rule that only handles easy cases is not a rule.

The question

Draw the editorial line between labeled satire and misleading synthetic media, then test your line on three headlines that sit right on it. A rule that only handles easy cases is not a rule.

System / materials

Three genuinely hard edge cases, written or found: a satirical piece that travels without its label, a synthetic image with obvious visual cues that stop being obvious when cropped, an AI-assisted illustration of a real event. Your publication's existing satire or opinion policy, if one exists. Published cases where satire was mistaken for reporting, cited.

Expected failure modes

A line drawn so it happens to permit everything the student wants to make. Assuming context travels with the piece — screenshots and reshares strip it, and that is the normal case rather than the exception. Ignoring what a reader who has never heard of your publication would conclude. Confusing a label that exists with a label that survives.

Done looks like

A decision rule: the line stated in one or two sentences a section editor could apply; the three edge cases run through it with the verdict and reasoning for each, including any case the rule handles badly; a revision showing how the rule changed after the hard cases, or an argument for why it held; and a note on what happens when a piece travels without its context.

Five C's

CT: testing a rule against cases designed to break it. CR: choosing edge cases in good faith rather than easy ones. CO: a peer applies the rule to the same three and disagreements are reconciled or documented as genuine ambiguity. CM: a rule an editor applies without a meeting. CZ: the reader who believes a piece that was meant as a joke.

Mentor role

An opinion or standards editor reviews the rule against the edge cases. Standing instruction: reject a rule that permits every case the student proposed. School-supervised.

Rubric calibration

R1: one rule, one publication, three cases. R2: reasoning reproducible by another editor. R3: comparator is the current policy or its absence. R4: context stripping addressed. R5: the rule is one or two sentences. R6: names what the publication will not publish regardless of labeling.

Two ways this goes wrong

(a) Three easy cases and a rule that was never at risk. (b) A rule that assumes the label travels with the image.

Credit lane fit

Lane A immediately (journalism, English, or ethics cross-listing). No verified credit claim.