Legal AI Solution Design MapInterpreting benchmarks for legal work and system design.
Menu

FairLex

Assurance · Model or response · Open · ECtHR, US, Swiss, Chinese, Indian · 2022

Fairness benchmark across five legal systems testing demographic parity and equal opportunity over protected attributes.

Reported results

No comparable score table is recorded in this collection.

What the benchmark measures

Fairness benchmark across five legal systems testing demographic parity and equal opportunity over protected attributes.

The evaluated unit is a model response or component output. Read the source for the exact prompt, tool and harness conditions.

How it is scored

Group fairness metrics (demographic parity, equal opportunity) alongside task accuracy.

Scores remain in the original unit. They are not normalised or combined with results from other benchmarks.

Sources

Tests classification fairness across several legal systems. It does not directly assess whether generated drafting changes with party demographics.