{"id": 7912, "prediction_verbatim": "MSFT outperforms QQQ over 48h [DIRECTION: up]. BULL CASE (stated above): US cloud capex leadership dominates China AI deterioration narrative. BEAR CASE (stated above): single-fund decline + narrative-level 'ROI comparison' do not confirm institutional repositioning without on-chain or fund-flow confirmation. [FALSIFY: MSFT underperforms QQQ or the two move in tandem (within 1.2% relative spread) over 48h].", "created_at": "2026-07-21 10:32:31", "direction": "up", "timeframe": "48h", "score": 0.17, "outcome": "Wrong \u2014 MSFT -3.0% vs QQQ +1.3% \u2014 MSFT trailed QQQ by 4.3%", "resolved_at": "2026-07-23 10:35:07", "chain_hash": "1b7d5777f9c186e5d71774fa8c004216d689b6222cdcbd6e87f213d0e0e04773", "chain_tx": "3bL91P2nUzeyET5fuUTmgK63U3tMD2UrzgV636Zgqrui97Yr6Eavj6WMbAHXJFe9bd8KchjakZLNBRAvFaFA1dsU", "verify": "sha256(f'{prediction_verbatim}|{created_at}|{id}') must equal chain_hash, and the Solana tx memo must read 'workshop:pred:<id>:<chain_hash>' with a timestamp BEFORE resolved_at. prediction_verbatim is the raw stored text, internal tags included \u2014 that is what was hashed."}