NY professional services hold seven of Vault's Top 10 — and Harvey AI in 250+ firms.

PAA

People also ask

• How does ABA Model Rule 1.6 confidentiality affect AI vendor architecture?

  ABA Model Rule 1.6 prohibits law firms from disclosing client information to a third-party vendor (including an AI vendor) without informed consent. Practical implication: vendor architecture must support no-training-on-firm-or-matter-data, no-cross-tenant-data-mixing, audit logs demonstrating confidentiality, and contractual no-disclosure clauses. ABA Formal Opinion 512 (July 2024) addressed generative AI specifically — lawyers must understand the technology, protect confidentiality, ensure competent representation, and comply with billing + supervision obligations. Every Big Law vendor contract now references Opinion 512 or its NY State Bar equivalents under 22 NYCRR Part 1200.

• What is Wachtell's PPP and how does that affect mid-market firm pricing?

  Wachtell Lipton Rosen & Katz holds the highest Big Law PPP at ~$8.3M in FY2023, followed by Cravath at ~$5.8M PPP (first-year associate base $225,000). Seven of Vault Top-10 firms headquarter in NYC: Wachtell, Cravath, Sullivan & Cromwell, Davis Polk, Skadden, Cleary Gottlieb, Paul Weiss. The mid-market wedge: own 1–2 niche practice areas where Big Law charges $1,800/hour and the 30–200-attorney firm charges $700–$1,200/hour, with comparable AI tooling closing the speed gap on first-pass diligence.

• Can the Voice Agent preserve Rule 1.18 prospective-client confidentiality?

  Yes. ABA Model Rule 1.18 prospective-client confidentiality is preserved by design. Inbound caller-ID + name pre-screen against the firm's matter and conflicts database; if a conflict trigger fires, the call is routed to a designated conflicts attorney rather than a generalist intake taker. The script avoids extracting `significantly harmful information' from a prospective client that would later disqualify the firm from representing the adverse party — Rule 1.18(c) carves out a preservation path if the conversation stays high-level. Audit-log retention aligned to NY Rules of Professional Conduct.

• Is Harvey AI the only Big Law LLM platform, or are there alternatives?

  Three reference platforms dominate. Harvey AI ($1.5B valuation, Kleiner Perkins-led Dec 2023 Series B) is deployed at A&O Shearman across ~3,500 lawyers in 43 offices and at PwC's ~4,000 legal professionals. Lexis+ AI (LexisNexis) launched October 2023 with 1,000+ law firm customers in 12 months. Westlaw Precision AI + Thomson Reuters CoCounsel (acquired Casetext for $650M in August 2023) competes directly. Most Am Law 100 firms deploy Lexis+ AI and Westlaw Precision AI in parallel rather than as alternatives.

• Does NY State Bar Rule 7.1 apply to AI-generated marketing copy?

  Yes — AI-generated marketing content does not get a pass under 22 NYCRR Part 1200. The firm is responsible for AI outputs as if drafted by a licensed attorney. Rule 7.1 prohibits false or misleading communications; Rule 7.2 governs advertising and specialty designation; Rule 7.3 governs solicitation including the 30-day no-solicitation period after personal injury; Rule 7.4 restricts `expert in [practice area]` claims unless ABA-approved certified; Rule 7.5 governs firm names and letterheads. Areza bakes Rule 7.1–7.5 review into the publish pipeline so AI-generated content cannot reach a public surface without attorney sign-off.

Frequently asked

• How does Areza handle ABA Model Rule 1.6 confidentiality and no-training-on-firm-data?

  Every NY law firm engagement starts with a vendor agreement that explicitly carves out client matter data from training corpora and from any retention beyond firm-controlled audit logs. We are a third-party service provider that does not retain or train on firm or matter data — period. The architecture: sub-processor list documented at engagement start, DPA with explicit no-training clause, audit logs retained for the period the firm requires (typically the longer of 7 years or the firm's NY Rules of Professional Conduct 22 NYCRR Part 1200 retention), no-cross-tenant-data-mixing, and contractual reference to ABA Formal Opinion 512 (July 2024) on generative AI in legal practice. For internal-only Knowledge Bot RAG over firm materials, we configure the firm's matter library + precedent bank + clause library + drafting templates as the corpus with strict no-training architecture and audit logs accessible to the firm's General Counsel + Office of General Counsel.

• Are you compliant with NY State Bar Rule 7.1–7.5 advertising rules?

  Yes. Every marketing surface we publish for a NY law firm goes through a Rule 7.1–7.5 review baked into the publish pipeline. Rule 7.1 (no false or misleading communications) — comparison claims are objectively verifiable; performance claims (`got our client $50M`) are representative + disclaimed; testimonials avoid prohibited language. Rule 7.2 (advertising) — `attorney advertising` labelling on every public communication where required. Rule 7.3 (solicitation) — no direct solicitation of persons known to be in need of legal services in a manner the rule prohibits. Rule 7.4 (specialty claims) — `expert in X' avoided unless the attorney is certified by an ABA-approved specialty organisation. Rule 7.5 (firm names + letterheads) — firm name + letterhead reflect actual partner roster + admissions. AI-generated marketing content does not get a pass; every AI-generated page requires an attorney sign-off before publish.

• How do you handle NYC Local Law 144 (AEDT) for a law firm's recruiting tooling?

  We map the bias-audit obligation across the firm's Workday + Greenhouse + Lever ATS layer, configure the candidate-notification flow on the recruiting marketing surface (the 10-business-day notice + the `opt-out and request alternative process` language the statute requires), and surface the summary results of the most recent bias audit as plain HTML with the publication date. We do not perform the bias audit ourselves — the statute requires an independent auditor with no financial interest in the AEDT or the employer. For clients that need both marketing-site work and bias audit, we refer to a specialised AEDT auditor (Holistic AI, BABL AI, Crowe LLP, BSI are the active names in the NYC market). Big Law NYC recruiting teams have been operating under this statute since July 2023 — the compliance cadence is established, but mid-market firms growing rapidly into NYC may have gaps.

• Does the Voice Agent support Rule 1.18 prospective-client conflicts pre-screen?

  Yes. The Voice Agent is built to preserve ABA Model Rule 1.18 prospective-client confidentiality. Inbound caller-ID + name pre-screen against the firm's matter and conflicts database; if a conflict trigger fires, the call is routed to a designated conflicts attorney rather than a generalist intake taker. The script avoids extracting `significantly harmful information' from a prospective client that would later disqualify the firm from representing the adverse party — Rule 1.18(c) carves out a path the firm can preserve if the conversation stays high-level. Audit-log retention is compliant with NY Rules of Professional Conduct and ABA Model Rule 1.18 commentary. PEP / sanctions screening on inbound caller (for white-collar + regulatory practice intake) escalates to a human conflicts officer inside 30 seconds.

• What about NY Public Health Law Article 27-F + HIPAA for firms with healthcare practice groups?

  Big Law firms with healthcare practice groups (Sullivan & Cromwell, Skadden, Cleary, Cravath, and the mid-market healthcare boutiques) handle PHI in M&A diligence, regulatory comments, FCA + Anti-Kickback litigation, and IRS Form 990 / nonprofit-healthcare matters. Vendor architecture must support BAA-covered inference for healthcare-matter work or explicitly carve out healthcare matters from AI-tool use. Areza signs a Business Associate Agreement on a per-matter basis for healthcare engagements; we route healthcare-matter inference through AWS Healthcare-eligible regions or equivalent BAA-covered providers. NY Public Health Law Article 27-F (HIV confidentiality) adds a state-specific layer for any HIV-related healthcare matter — the disclosure floor is higher than HIPAA. We do not default-include healthcare matters in standard fintech or PE engagements without explicit scoping.

• How fast can a Voice Agent in US English go live for a NY law firm?

  14 days from kick-off for the standard configuration: inbound handling, prospective-client-conflicts pre-screen, jurisdiction screening (we cannot give advice in NY if not admitted in NY), calendar booking via Calendly with attorney-level routing, CRM hand-off (Salesforce, HubSpot, Clio, MyCase supported), and outbound reminders in US English. Add a week for bilingual EN + ES overlay (the standard for immigration + family-law + personal-injury practices where Spanish-language intake is material). Add two weeks for practice-area-specific compliance scripts — securities + M&A (the SEC + FINRA-adjacent boilerplate), litigation (the SDNY + EDNY filing-deadline + procedural-rule overlay), restructuring (chapter 11 + chapter 15 + Bankruptcy Code overlay), healthcare (HIPAA + NY Art 27-F overlay), tax (IRS Circular 230 overlay). The 14-day baseline assumes you can provide your intake recordings, FAQs, and the names of the three buyer questions you hear most often.

• What pricing should a NY mid-market law firm expect for an Areza engagement?

  Foundation starts at USD $5,200 for a 2–4 week conversion-first build with NY State Bar Rule 7.1–7.5 + ABA Model Rule 1.6 + 1.18-aligned practice-area pages, structured-data attorney bios with Lawyer + Person + LegalService schema, FAQPage markup for `do I need a lawyer for X` queries, llms.txt configured for ChatGPT + Perplexity + Claude allow-list, ADA WCAG 2.1 AA compliance, NY SHIELD-aligned cookie banner. AI Search retainer starts at USD $430/month with named-target citation tracking against Chambers + Vault + Above the Law + Law360 + Bloomberg Law. Voice Agent for inbound intake + Rule 1.18 conflicts pre-screen + Calendly handoff adds USD $1,300–$1,900/month. Knowledge Bot from USD $315/month for the public-facing FAQ + thought-leadership surface; firm-internal Knowledge Bot variant with matter-library RAG adds USD $800–$1,500/month. A typical NY mid-market firm engagement combines Foundation + AI Search + Voice Agent at USD $7,200–$10,200 setup plus USD $1,600–$2,700/month.

• How does Areza differ from a NY Big Law firm's in-house tech team or a Big-4 NYC consulting practice?

  Big Law in-house tech teams (Cravath, Davis Polk, Skadden, Sullivan, Wachtell all have one) handle the firm's internal AI tooling — Harvey + Lexis+ AI + Westlaw Precision AI + iManage + Relativity integration, Microsoft 365 Copilot rollout, internal Knowledge Bot RAG over matter libraries. They do not handle marketing + AI-search citation infrastructure; that work is split between the CMBDO + Director of Marketing + outside agencies. Big-4 NYC consulting practices (Deloitte, EY, KPMG, PwC) open enterprise re-platforming envelopes above USD $500K with 10–30 FTE delivery teams — they serve corporate in-house legal + tax + compliance teams more than law firms themselves. Areza is purpose-built for the AI-search + agentic-automation + voice layer at mid-market NY law firms (30–300 attorneys) — the firms that Tier-1 consultancy pricing filters out and that need to compete with Cravath on first-pass diligence speed without paying Cravath's overhead. The honest split: hire the firm's in-house tech team for Harvey integration, hire a Big-4 NYC practice for enterprise re-platforming, and bring Areza in for the marketing + AI-search + Voice Agent + Workflow Ops layer where the systems-first approach compounds.