← The impacts · AItopiaOrAImageddon?

candidate · p2026-09-06-302 · not-found

Product recalls or defects found early by a model

set aside as

NOT FOUND

the researcher's note

Product recalls or defects found early by a model. Nothing found that names a learned model AND counts people. NHTSA's own process document (DOT HS 812 984, Risk-Based Processes for Safety Defect Analysis and Management of Recalls) describes analyst review of the ODI complaint database, not a fitted model; NHTSA publishes recall reach (2019: 966 recalls, 38.6 million vehicles) but attributes it to no model. No public CPSC document describes an ML triage of incident reports with an outcome. FDA Sentinel's signal methods are epidemiological; the machine-learning work over FAERS is research papers, not a deployed system with a people effect.

proposed by

researcher (opus), brief 2026-09-06, severe-benefits

received

2026-09-06T17:40:15-04:00

harvest report

impacts/harvest/2026-09-06-severe-benefits.md

Rulings, in order

1. 2026-09-06T17:40:15-04:00 · researchernot-found — Product recalls or defects found early by a model. Nothing found that names a learned model AND counts people. NHTSA's own process document (DOT HS 812 984, Risk-Based Processes for Safety Defect Analysis and Management of Recalls) describes analyst review of the ODI complaint database, not a fitted model; NHTSA publishes recall reach (2019: 966 recalls, 38.6 million vehicles) but attributes it to no model. No public CPSC document describes an ML triage of incident reports with an outcome. FDA Sentinel's signal methods are epidemiological; the machine-learning work over FAERS is research papers, not a deployed system with a people effect.

{
 "id": "p2026-09-06-302",
 "received": "2026-09-06T17:40:15-04:00",
 "proposer": "researcher (opus), brief 2026-09-06, severe-benefits",
 "kind": "candidate",
 "sha": "033f33eed0a48f9f",
 "data": {
  "name": "Product recalls or defects found early by a model",
  "note": "Product recalls or defects found early by a model. Nothing found that names a learned model AND counts people. NHTSA's own process document (DOT HS 812 984, Risk-Based Processes for Safety Defect Analysis and Management of Recalls) describes analyst review of the ODI complaint database, not a fitted model; NHTSA publishes recall reach (2019: 966 recalls, 38.6 million vehicles) but attributes it to no model. No public CPSC document describes an ML triage of incident reports with an outcome. FDA Sentinel's signal methods are epidemiological; the machine-learning work over FAERS is research papers, not a deployed system with a people effect.",
  "list": "NOT FOUND"
 },
 "from_file": "impacts/harvest/2026-09-06-severe-benefits.md"
}