Rule — class-adjacent exposure
Adverse-impact statistical method + Rule 23 class-adjacent exposure
The statistical method plaintiff firms use to prove adverse impact in employment-discrimination and lending-law class actions now cuts the other direction. If an AI case-intake classifier over-routes one client segment into a punitive lane at a rate the same statistical test would flag as adverse impact, the same class-adjacent exposure attaches — against the plaintiff firm.
What the method actually asks
"A selection rate for any race, sex, or ethnic group which is less than four-fifths (4/5) (or eighty percent) of the rate for the group with the highest rate will generally be regarded by the Federal enforcement agencies as evidence of adverse impact."
EEOC Uniform Guidelines on Employee Selection Procedures — the four-fifths rule, the reference statistical test
"One or more members of a class may sue or be sued as representative parties on behalf of all members only if: (1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class."
Fed. R. Civ. P. 23(a) — class-certification prerequisites
What this means in plain English
Two moves link the statistical test to the firm's own exposure:
- The four-fifths test applies to any AI-driven selection. Case-intake triage is a selection. Routing a client to SIU/fraud-review vs Fast-Track Settle vs Standard Adjuster Review is a selection with financial consequences for the client. If the selection rate on one segment is less than 4/5 the rate on another, the pattern is prima facie adverse impact.
- The class-certification test is easy to meet when the AI is the common actor. Numerosity is trivial (hundreds of intakes per drift window). Commonality is baked in (same AI, same period, same failure mode). Typicality is the AI's routing pattern. Adequacy is the plaintiff-class counsel's job.
Historically, adverse-impact class actions targeted employers, lenders, and landlords. The 2024-2026 case law extended the same reasoning to any commercial actor whose AI produces segment-differential outcomes.
What triggers the exposure in the sample
C3 SIU/fraud-review rate 71% vs baseline segments' ~5-15%. That is not a four-fifths violation — it is a five-to-fifteen-fold ratio. Under any reasonable statistical framing, the pattern is adverse impact. Once one C3 client complains, the class-certification test is trivially met against the firm.
The wrinkle for a plaintiff PI firm: the same firm that would run the adverse-impact case against a defendant is now the defendant. The defense-bar's tactical playbook against class-cert is available — but the underlying pattern is not favorable.
What the $499 Snapshot shows against this rule
- Per-segment selection-rate measurement — the four-fifths test directly applied
- Distributional-shape severity classification — whether the drift is momentary or sustained
- Dated identification + 3 fix-first items including freeze/hold — the mitigation record
- Signed independent-verifier declaration — the third-party evidence class-adjacent counsel + carriers weigh differently than self-report
See the lane-rate chart that produces the four-fifths finding →
$499 Snapshot. 3 business days.
Per-segment measurement + dated determination on your firm's own AI-driven intake selection.
Buy $499
Snapshot credit applies to Baseline ($2,500) or Enterprise Attestation ($35-55K) upgrade within 30 days.