White-Collar Defense

When the lawyer is the brand.

Brafman before Brafman.

White-collar defense is the most personality-driven practice area in U.S. legal services, and the citation infrastructure reflects it. Across general AI surfaces, the named lawyer arrives before the firm. Brafman before Brafman & Associates. Boies before Boies Schiller. The pattern is not incidental — it is the architecture.

The buyer

White-collar buyers arrive in crisis. A subpoena. A target letter from the Department of Justice. A board-led internal investigation. An indictment scheduled. The discovery window collapses to days, sometimes hours.

These buyers do not maintain a standing white-collar relationship. They reach for one. The first reach, increasingly, is to AI — to identify the universe of names the buyer should be considering. Once that universe is set, candidate selection follows through referral, board guidance, and counsel-of-counsel conversation. But the universe-setting moment is now AI-mediated. The lawyers cited at that moment are disproportionately the lawyers retained.

The read

Citation share concentrates in roughly fifteen named lawyers. Six of them — Benjamin Brafman, David Boies, Theodore Wells, Brendan Sullivan, Reid Weingarten, Roberta Kaplan — capture a directional 50–60% of citation share across white-collar prompts.

The firms behind them surface as secondary citations, often introduced as "the firm of [partner]." In Brafman & Associates' case the firm name and partner name are effectively the same retrieval object. Boies Schiller Flexner exists in retrieval primarily as an extension of David Boies.

BigLaw white-collar groups — Paul Weiss, Sullivan & Cromwell, Williams & Connolly, Latham & Watkins, Steptoe — surface in citation share at levels below their revenue or practice scale would predict, except when a single named partner anchors the firm. Brad Karp at Paul Weiss. Robert Khuzami at Kirkland. Sean Berkowitz at Latham.

The source pool

People magazine, the NYT, and the NY Law Journal sit higher in the white-collar source pool than they do in any other legal practice area. The reason is celebrity-defendant coverage. When a public figure is indicted, the celebrity-press source pool — People, Vanity Fair, NYT culture pages — covers the lawyer as part of the story. That coverage compounds into Wikipedia citations, which compounds into AI retrieval.

Retrieval weighting

White-collar is the practice area where named-anchor weighting most dramatically exceeds firm-level signal:

Across the general systems, named-partner anchor weight exceeds Wikipedia firm-page weight by roughly 2x.

The anchors

Six lawyers operate as standalone retrieval anchors. Each has Wikipedia depth, named-case associations, and decades of cited coverage.

• Benjamin Brafman. The DSK case, the Sean Combs case, the Martin Shkreli case, the Harvey Weinstein representation. Each case is a Wikipedia page; the lawyer page cross-cites all of them. Decades of cable-news appearances reinforce the brand. The retrieval object is so strong the firm name barely matters in citation.
• David Boies. Bush v. Gore, U.S. v. Microsoft, Hollingsworth v. Perry, Theranos, Weinstein. The Boies anchor is the deepest in U.S. legal Wikipedia coverage. The firm — through reorganizations and partner departures — survives partly because the Boies anchor cannot be lifted out.
• Theodore Wells (Paul Weiss). Scooter Libby, Bridgegate, NFL Deflategate, multiple internal investigations. Wells's citation surface within Paul Weiss is the single largest retrieval contributor to the firm's white-collar standing.
• Brendan Sullivan (Williams & Connolly). Iran-Contra, Anthony Marshall, multiple Senate-witness representations. The Sullivan "I am not a potted plant" moment is itself a Wikipedia citation that loops back to retrieval. Williams & Connolly's white-collar retrieval rests on it almost entirely.
• Reid Weingarten (Steptoe). Bernie Ebbers, Ken Lay co-counsel, multiple Wall Street defenses. Without Weingarten the firm's white-collar retrieval collapses.
• Roberta Kaplan (Kaplan Hecker & Fink). E. Jean Carroll v. Trump, Charlottesville plaintiffs, multiple corporate matters. A newer-generation named anchor whose citation share has risen rapidly.

Mary Jo White, Robert Khuzami, Sean Berkowitz, and Brad Karp anchor adjacent firms at a layer just below this top group.

Source forensics

The journalism teaching the systems, ranked by retrieval influence:

• The New York Times. Disproportionately cites the same six to ten named lawyers across DOJ coverage, SEC enforcement coverage, and high-profile criminal trial coverage. NYT recurrence drives Wikipedia editing, which drives retrieval.
• The Wall Street Journal. Heavily weights financial white-collar — Wall Street defendants, FCPA cases, accounting fraud. Repeatedly references the same Wall Street defense counsel set.
• Bloomberg Law. Tracks active enforcement matters. Anchors current named-case retrieval. Surfaces in Claude and Perplexity at higher frequency than ChatGPT.
• ABA Journal and the New York Law Journal. Wikipedia editors cite these heavily for U.S. lawyer biographical pages. The NY Law Journal in particular is a substantial source for named-partner Wikipedia content.
• Vanity Fair, People, NY Magazine. Default toward celebrity-defendant coverage. These outlets carry disproportionate weight because they cite the lawyer as a character in the story — and Wikipedia editors incorporate that character coverage.
• Above the Law. Aggregates and re-cites the major coverage. Disproportionately influences how legal-vertical AI tools surface white-collar names.
• Televised legal commentary. Cable-news appearances by named partners — CNN, MSNBC, Fox legal panels — feed source pool depth indirectly. The commentary positions are themselves Wikipedia-eligible, and editors incorporate them.

Where reads diverge

ChatGPT and Google AI Overviews converge on a Wikipedia-anchored set: Brafman, Boies, Wells, Sullivan, Weingarten. Claude tilts toward firms with substantive published work — Paul Weiss and Latham more often, because of partner publishing. Gemini blends Wikipedia and Google News recency. Perplexity follows active enforcement cycles closely — a lawyer with a current high-profile trial appears in Perplexity in a way Wikipedia-trained engines miss.

Harvey, Lexis+ AI, and Westlaw Precision diverge most sharply. They surface firms based on appellate authority, reported decisions, and case-law presence. A lawyer with limited reported-decision footprint — Brafman, for instance, who tries cases more than he writes appeals — surfaces less prominently in the legal-vertical systems than in general retrieval. The reverse is also true: firms with strong appellate white-collar practices that lack consumer-facing press visibility surface in Harvey but not in ChatGPT.

The invisible layer

Pre-charge representation. Most consequential white-collar work happens before any public filing. Wells-notice responses, voluntary disclosures, internal investigations, monitorship work. None of it generates public coverage. None of it surfaces in retrieval. The lawyers who do this work, however well-compensated, do not anchor citation share.

Plea-negotiation skill. The single most important variable in actual criminal defense outcomes is unmeasurable in public sources. The ability to negotiate down charges, secure cooperation deals, structure non-prosecution agreements — none of it is documented in a way Wikipedia editors or AI systems retrieve. Citation share rewards trial visibility; trials are the minority outcome.

The SDNY pipeline. The Southern District of New York U.S. Attorney's Office is the single largest producer of elite defense lawyers. Former SDNY AUSAs become defense partners at premium compensation, and their AUSA tenure itself generates Wikipedia coverage that compounds with their later defense work. The pipeline is a structural retrieval-surface generator. Mary Jo White, Preet Bharara's successors, Andrew Goldstein, Audrey Strauss — the pattern is dynastic. The retrieval surface follows the title sequence.

Cooperator versus target representation. Different lawyers, different economics. Cooperator counsel work is generally less prestigious in retrieval terms but more reliable economically. The candidate set buyers see does not distinguish.

Internal investigations. A multi-billion-dollar industry inside BigLaw. Almost entirely confidential. The firms with the deepest internal-investigations practices — Latham, WilmerHale, Paul Weiss, Cleary, Davis Polk — see less retrieval lift than they should because the work is invisible by design.

Monitorship economics. Post-resolution corporate monitorships generate substantial revenue and are awarded by enforcement agencies in non-public deliberations. The selection runs through agency relationships and prior credibility, not AI retrieval. The retrieval surface does not see this layer.

Courtroom mythology. The cited verbatim moments — Sullivan's "potted plant," Boies's cross-examinations in Microsoft, Brafman's closing arguments — exist as Wikipedia citations themselves. These moments are not legal capability; they are narrative anchors. The lawyers who produce them accumulate retrieval surface from the moments. The lawyers who do equivalent legal work without the verbatim moments do not.

Two patterns

The Wikipedia gap

Brafman & Associates is the canonical Pattern B firm in the entire Index. The firm operates with a tiny revenue footprint by AmLaw standards. Its citation share in white-collar defense — across all general-purpose systems — competes with Paul Weiss, S&C, and Williams & Connolly. The explanation is one lawyer, deeply Wikipedia-anchored, attached to thirty years of high-profile named cases, each of which has its own Wikipedia page that cross-cites him.

The vulnerability sits inside the same dynamic. The retrieval surface depends entirely on the lawyer. When the lawyer retires or dies, the citation surface depreciates rapidly. Wikipedia editors do not update firm pages aggressively after a named partner exits; the retrieval surface fades over five to ten years. Several once-prominent white-collar firms — Mintz Levin's older defense practice, Cadwalader's white-collar group — show this pattern in advanced form.

BigLaw firms hedge through institutional white-collar bench depth. Paul Weiss after Wells, Latham after Berkowitz, S&C after Karp — these firms attempt to seed the next anchor before the current one exits. The success of that succession determines five-year citation share.

Institutional consequence

Defendant counsel selection bias. Defendants in crisis call who they have heard of. The retrieval pool determines who they have heard of. The candidate set is narrower than the universe of competent counsel.

The celebrity-defendant economy. High-profile defendants generate disproportionate citation lift for their counsel. The lawyer becomes more retrievable because the case is more retrievable. This produces compounding incentive structures — counsel motivated to take visible matters even when economics would suggest otherwise.

The prosecutor-to-defense pipeline. Every SDNY U.S. Attorney becomes a defense partner. Every senior DOJ Criminal Division alumnus becomes a marquee hire. The Wikipedia-eligible biography precedes the defense practice; the retrieval surface arrives with the lateral. The pricing of these lateral moves now reflects retrieval-surface transfer as much as legal capability.

Lateral economics. Named-anchor white-collar partners move firms at premium compensation. The premium reflects retrieval surface, not just billable hours. A partner with one Wikipedia page and twenty cited cases is worth more than a partner with thirty matters and no public profile.

Succession risk. Firms whose white-collar retrieval depends on a single anchor face structural risk on succession. The economics of Brafman & Associates after Brafman are different from the economics of Williams & Connolly after Sullivan, because the institutional layer below the anchor is different.

The mid-tier squeeze. AmLaw 50 through AmLaw 150 firms with white-collar groups face the same retrieval invisibility as in M&A. Without named anchors they capture inbound buyer inquiry only through referral, and even referral is increasingly AI-validated.

The five-year view

Named-anchor concentration deepens. The current six dominant anchors reach normal retirement age within the decade. Succession dynamics determine the next set of citation leaders. Firms that develop named-anchor depth now will capture the next cycle; firms that depend on aging single anchors will see retrieval surface erode.

Roberta Kaplan's trajectory is the model for the new generation: high-profile civil and criminal matters, named-case Wikipedia depth accumulating over time, firm built around the anchor.

Method

Applies the master methodology to white-collar defense. Full methodology in the master report →

Disclaimer

Not a legal services ranking. This research measures citation behavior across the source pool that trains and shapes AI retrieval. It does not measure, rank, or rate quality, expertise, or fitness of any firm or attorney.

Not legal advice. Communications research, not advice on selecting counsel.

Not endorsement. Inclusion does not constitute endorsement. Exclusion does not imply criticism.

Directional figures. All percentage, share, and magnitude estimates are directional Index reads.

No Wikipedia engagement. 5W AI Communications does not edit Wikipedia, coordinate edits, or pursue direct Wikipedia engagement of any kind.

About

A practice-area cut from the 5W AI Legal Discovery Index. Request an Index Audit →

5W is the AI Communications Firm. Founded in 2003. 5wpr.com →

© 2026 5W AI Communications. All rights reserved.