Verify a decision

{"actor":"judge","decided_at":"2026-08-26T08:03:52.335Z","decision":"review_revise","investigation_id":"7b8d8b86-b079-4597-9103-2049c740e3e3","new_score":0,"page_slug":"june-2026-cross-chain-bridge-exploit-127m","prev_score":0,"reason":"This page does not assert that a $127 million bridge exploit happened; it examines the claim, traces it to a single low-credibility marketing blog, and lays out the real, verifiable June 2026 bridge incidents (Syscoin, ~$9-10M) alongside the much larger April 2026 KelpDAO exploit for context. The review independently confirmed nearly all of this and found the page's core conclusion -- that no Tier 1 or Tier 2 outlet corroborates a $127M June 2026 bridge exploit -- to be solid, even turning up additional evidence (a fabricated 'Wormhole February breach' reference in the page's sole named source) that reinforces the page's skepticism (claim_findings[0]). The one real inaccuracy is claim_findings[3], which is disputed: the page calls the Syscoin incident the 'only' large verifiable June 2026 bridge exploit, but a comparably sized or larger incident (Humanity Protocol, ~$32-36M, Tier 1-reported) occurred the same month and should have been acknowledged. That overstatement affects a secondary framing point, not the page's central thesis, and is compounded by one partially-supported technical detail (claim_findings[9]) and one unverifiable price statistic (claim_findings[12]). With the tally standing at 17 confirmed, 1 partially supported, 1 disputed, 1 unverifiable out of 20 claims, disputed_pct lands exactly at the 10% boundary between approve and revise; given a real, citable inaccuracy exists (however minor) and reviewer confidence is solid but not maximal, this resolves to the low end of revise rather than approve.","score_delta":-5,"sequence_num":3,"submission_content_hash":null,"submission_id":null,"submission_kind":null,"submission_valence":null,"v":1}