Across the AmLaw 200 — the top 200 US law firms by gross revenue — AI engines rank legal expertise through a pecking order that materially diverges from the rankings practitioners and recruiters use.
The 5W Legal Discovery Index, published by 5W AI Communications in 2026, measured how the five major AI engines render the AmLaw 200 across seven practice cuts: M&A, Litigation, Intellectual Property, Restructuring/Bankruptcy, Antitrust/Regulatory, Capital Markets, and Labor & Employment. The methodology mirrors the broader Reputation Index program — sixty-plus retrieval-intent prompts per practice cut, scoring across Accuracy, Sentiment, Completeness, Consistency, and Control.
The results scramble the traditional AmLaw 200 hierarchy in legible patterns. Three of those patterns matter most for managing partners and law-firm marketing leaders.
Pattern 1: Practice depth beats firm scale. Across all seven practice cuts, the AI engines do not consistently rank by firm revenue, headcount, or office footprint. They rank by practice-level primary-source density. A firm with a small but deeply published M&A practice — sustained partner-authored thought leadership, ranking-organization placements in Chambers and Best Lawyers, sustained client-side and deal-side editorial coverage, named-deal recognition across the trade press — outranks a larger firm whose M&A practice publishes less.
The implication is that the AmLaw 200 ranking captures aggregate firm revenue, while AI engine rendering captures practice-level signal density. The two are loosely correlated at the top end (the largest firms typically have deep practices in multiple cuts) and weakly correlated through the middle (mid-tier firms with one or two deep practices outrank larger generalists in those practices).
Pattern 2: Boutiques punch above their weight in specialist practices. Across Intellectual Property, Antitrust/Regulatory, and Labor & Employment, specialist boutiques with sustained primary-source publishing dominate AI engine rendering in their practice cuts despite materially smaller aggregate revenue than the AmLaw 50 firms they outrank. The mechanism is identical to what the Trust Map of America documented in consumer categories — depth of primary-source signals at the local level (in this case, the practice level) outweighs scale.
The implication for managing partners at boutique firms is that AI-mediated discovery may be substantially more accessible than traditional recruiting and referral channels. The implication for managing partners at AmLaw 50 firms is that aggregate revenue does not translate automatically to AI authority in any specific practice cut — each cut requires its own primary-source investment.
Pattern 3: The Litigation cut is dominated by anchor cases. Across the litigation practice cut, the engines render firms through the most-rendered case in the firm's recent corpus. A firm anchored to a Supreme Court victory, a high-profile securities case, an antitrust verdict, or a constitutional argument carries that anchor across every identity prompt. The mechanism is identical to the Anchor Event Era pattern documented at the principal level — the engine compresses identity around the most-rendered event.
The litigation cut also produces the largest engine-to-engine variance. ChatGPT, Claude, Perplexity, and Gemini diverge sharply on litigation rankings even when given identical prompts. The engines weight different recency signals, different source corpora, and different ranking-organization signals. Multi-engine visibility — already established as the operating framework for consumer brand AI authority — applies with particular force to litigation practices.
The structural finding. Across the seven practice cuts, the AmLaw 200 reorders consistently along three axes. Practice depth dominates firm scale. Boutiques outrank larger firms in their specialist cuts. Anchor cases compress firm-level reputation in litigation specifically.
For managing partners reading this, the implications converge into a single category of investment that most AmLaw 200 firm marketing departments have under-resourced relative to its impact.
The buildable signals. Each practice cut the Index measured responds to four primary-source signal categories.
Practice-level partner publishing. Sustained partner-authored thought leadership — Harvard Law Review, Yale Law Journal, Stanford Law Review, and NYU Journal of Legislation and Public Policy placements; sustained partner presence on the Harvard Law School Forum on Corporate Governance and the Columbia Blue Sky Blog; Above the Law contributor columns; Bloomberg Law analytical commentary; Law360 expert columns. Each compounds across years into a primary-source corpus the engines retrieve from.
Ranking-organization placements. Chambers USA, Chambers Global, Best Lawyers, US News Best Law Firms, Benchmark Litigation, Legal 500. The engines treat these as authoritative practice-level signals. Investment in ranking submissions is now infrastructure investment, not marketing tactic.
Named-deal and named-case coverage. Trade-press coverage of specific deals and cases, with named-partner attribution, that the engines retrieve as confirmation of practice expertise. Sustained named-deal coverage is the single strongest correlate of AI engine ranking across the M&A and Capital Markets cuts the Index measured.
Wikipedia accuracy. The engines retrieve heavily from Wikipedia at the firm and partner level. Accurate, up-to-date, well-cited Wikipedia entries materially affect engine rendering across every practice cut.
The category implication. Legal services is the professional-services category in which AI-mediated discovery has the longest pre-engagement consideration window. General counsels and litigation prospects research outside counsel choices for months before engaging. The first frame of that research is increasingly the AI answer — and the AI answer reflects practice-level signal density rather than firm-level revenue or headcount.
The Legal Discovery Index findings are now in circulation among AmLaw 200 marketing partners. The next twelve months of legal-services marketing budget will reflect what the Index found — investment shifting from brand campaigns and general firm marketing toward practice-level primary-source infrastructure that the engines retrieve from.
The AmLaw 200 ranking captures aggregate revenue. The AI engine pecking order captures practice depth. The two will converge over time — not because the engines will change, but because the firms that learn to build practice-level primary-source corpora will displace the firms that do not.
The pecking order is published. The mechanism is now known.
