3 patterns for this goal
Precedent-currency failures happen when a legal research agent cites a case as controlling law without checking whether it has been overruled, when a research agent identifies an overruled status in its own analysis but never communicates that status to a downstream drafting agent through the structured handoff schema, or when a legal research agent fails to surface contrary authority across multiple circuits or jurisdictions despite knowing a circuit split exists on the cited proposition. Precedent-currency failures are particularly damaging because they make a brief or memo’s foundation of authority appear sound when the cited cases are in fact superseded, limited, or contradicted by later decisions that the research step could have retrieved but didn’t. Because LLM research agents have access to legal-research databases and citator services but operate without explicit post-generation verification, superseded authority can appear in filed documents without detection until opposing counsel flags it, at which point the filing’s credibility is irreparably harmed and strategic position is compromised.
Precedent-currency failures are failures of independent verification at the output stage. The superseded-authority pattern stops after retrieving a case without checking current citator status. The circuit-split pattern ranks results by topical relevance without requiring an explicit search for contrary authority. The handoff pattern relies on a citation list with no validity-status field to capture nuance the research agent identified. The fix recurs across all three patterns: add a mandatory verification layer that doesn’t trust the generation step’s own confidence or schema. Run live citator checks before output, implement dual-path research (primary + explicit contrary-authority search), enforce structured validity fields with reconciliation before handoff.
Implement a real-time citator-check gate before output: after the research agent generates candidate citations, pass all citations through a live citator service (Shepard’s, KeyCite, LexisNexis) to check for overruling, limiting, or negative treatment. Capture the citator status (good law, overruled, limited, questioned) and include it in the output. If citator is unavailable, mark as UNVERIFIED and require attorney confirmation before use — see Superseded Case-Law Citation.
Implement dual-path research: Path 1 does primary topical search ranked by relevance; Path 2 runs explicit contrary-authority queries using negation terms and opposite-holding indicators. Cross-reference results against a Circuit-Split Registry of known splits. For any citation on a question with known splits, output must explicitly disclose the split, list all known split positions, and specify which jurisdiction’s rule is binding and which are persuasive — see Circuit Split Blindness in Citation Selection.
Extend the citation-list handoff schema to include point-specific validity fields: {case_id, citation, original_holding, validity_status: {overall_status, point_specificity: [{proposition, validity_on_point, overruling_authority}]}}. Validation gate: if research analysis contains validity-limitation language, require population of all point_specificity entries before handoff. Reconciliation check: scan research analysis for validity keywords; if found, verify corresponding entries in schema. If mismatch, escalate for correction — see Multi-Agent Handoff Drops Overruled-Citation Flag Between Research and Drafting Agent.
| Pattern | Mechanism |
|---|---|
| Circuit Split Blindness in Citation Selection | Contrary authority across multiple circuits exists but research does not surface it; brief presents holding as settled law when split actually exists |
| Multi-Agent Handoff Drops Overruled-Citation Flag Between Research and Drafting Agent | Research agent identifies overruled status in analysis but structured citation-list schema has no validity field; drafting agent cites case as good law |
| Superseded Case-Law Citation | Case cited as controlling precedent has been overruled, vacated, limited, or superseded without agent detecting current legal status |
Total: 3 patterns
Legal Research Agent Cites a Precedent as Settled Law Without Flagging That a Circuit Split (or Equivalent Jurisdictional Disagreement) Exists, Leading to a Brief That Misrepresents the Strength of the Authority
A Legal-Research Agent That Identifies, in Its Own Analysis, That a Specific Case It Surfaced Has Been Overruled on the Exact Point It Was Going to Be Cited For Hands Off a Structured List of Candidate Citations to a Drafting Agent That Strips That Overruled-Status Context, So the Drafting Agent Cites the Case as If It Were Still Good Law
Agent Cites a Case as Controlling Precedent Without Detecting That It Has Been Overturned, Limited, or Superseded by Later Authority