Two analytically distinct patterns appear in this record. First, the @sumibtw disable follows the standard 'account integrity' citation with no content identified — the notice states 'it still doesn't follow our Community Standards on account integrity,' with 'still' implying a prior review determination that the reporter does not describe. The absence of any content-level explanation for an account that had been operating for an unspecified period is consistent with automated integrity-signal detection rather than content enforcement. Second, the @sumilivin cycling pattern — suspended at least five times, with at least one documented reinstatement — is a strong false positive signal. A human review reinstating an account after an automated suspension, followed by the automated system re-suspending the same account, indicates the integrity classifier is persistently marking this account as a violation target despite human reviewers concluding otherwise. The reinstate-re-suspend cycle also undermines the claim that appeals are effective: an account that has been suspended, reviewed, reinstated, and suspended again four or more additional times is not being effectively cleared by the review process. Third, the Meta Verified subscription angle is notable: Meta markets Verified as providing access to real human support and enhanced account protection. Two months of no resolution for a paid Verified subscriber — despite following all support instructions — suggests either that the premium support tier lacks the authority to override integrity enforcement decisions, or that human support agents lack the ability to influence outcomes in AI-flagged cases. This is a distinct failure mode from the baseline: not just that appeals are ignored, but that paid support channels specifically marketed as providing recourse also fail to produce one.