 THE HEARING / SOURCE-USE CLARIFICATIONS



 Match the limit to the claim.
 Supplement to the original source register. Original evidence texts and numerical caveats are
 retained.


   Item / source               Clarification in the revised game


   RU1 / RU-A                  Levada produces the survey. Its report of ballot exclusion does not adjudicate registration
                               grounds or compare treatment of applicants.


   RU2 / RU-B                  The sentences are 2023 background to 2024. The report does not measure the amount of
                               electoral capacity lost before/after them.


   RU5 / RU-E                  Check whether an original detention number counts people or incidents. The card does not
                               turn those units into a national competitiveness measure.


   MY1 / MY-A                  Plurality in a district and a majority of seats are different units. Aggregate conversion does
                               not establish biased boundaries or why votes were geographically distributed.


   MY3-5 / MY-B, MY-D          IPU election summaries supply narrative context. DOJ material records legal allegations
                               and action, not a conviction or a measured voting response. Shared narratives are not
                               independent checks.


   MY6 / MY-E, MY-F            Later reform can show post-turnover opportunities or constraints. It cannot cause the
                               earlier May victory.


   MW3 / MW-B, MW-C,           MW-D is now also associated with MW3. Legal analysis and the direct-judgment retrieval
   MW-D                        caveat accompany the court sequence. Invalidation, remedy and majority interpretation
                               are distinct; personal culpability is not established.


   MW4 / MW-C                  A joint ticket does not track individual vote transfer. The reported 2020 winner had both a
                               majority and the largest share; strategic effects of the threshold need separate evidence.


   MW5-6 / MW-C, MW-E          Implementation reporting does not establish compliance motives. Reported shares
                               concern different contests. Conflicting exact official totals remain disclosed and
                               unexplained.


   RS2-3 / RS-A                Real alternatives can coexist with unequal conditions. Selected broadcast content does not
                               measure equal reach, individual exposure or persuasion.


   RS5-6 / RS-A                Pending and rejected remedies differ; the window does not establish permanent
                               impossibility. Ballot-to-seat conversion alone does not prove unfair campaigning or
                               unequal enforcement.

 Use the original register for complete titles, dates, locators and links. The game retains those records. A caveat is
 analytically useful when it narrows the particular claim you make, rather than merely appearing cautious.




THE HEARING / SECOND INDEPENDENT AI REVIEW ROUND / 30 OCTOBER 2026                                                           1
