A law firm's expert knowledge has been turned into a proprietary AI platform that can be sold externally. The AI re‑engineering of professional services firms is moving from internal cost reduction to an independently valued asset.
Facts
- September 14, 2026, the United States' largest personal injury law firm Morgan & Morgan announced it will invest at least USD 1,000,000,000 in AI and technology over the next ten years; since 2021 it has invested roughly USD 300,000,000 to build the internal AI platform MX2.
- MX2 currently has close to 5,000 monthly active users; the firm has 1,100+ lawyers and roughly USD 2,400,000,000 in annual revenue.
- The platform has entered the case production chain: automatic acquisition of medical records and police reports, structuring of medical records, generation of case materials, demand letters, trial preparation, and pattern searching across a nationwide case database.
- The company says it has automatically processed hundreds of thousands of demand letters and hundreds of thousands of medical record requests.
- Management did not disclose specific layoff numbers, per‑capita revenue, or EBITDA uplift, but clearly stated that benefits are reflected in case velocity, staffing efficiency, and settlement outcomes.
- Starting at 2027, Morgan & Morgan plans to sell MX2 to other law firms; previously the company hired JPMorgan to explore a minority equity sale.
Analysis
Possible evolution path: traditional professional services firm → distill expert experience into a proprietary AI platform → improve labor efficiency → productize internal capabilities → attract external capital → AI platform becomes an independent valuation source. Once the platform is sold externally, the valuation metric shifts from “law firm profit” to “software and data assets.”
This line mirrors the Ardian line from the same day: Ardian is a GP turning its investment memory into an Agent, while Morgan & Morgan is the operator productizing its professional expertise for external sale. Both point to a shift—the value created by AI-native reconstruction is moving from “internal cost reduction” to assets that can be priced and traded separately.
Counterevidence
- USD 1,000,000,000 is the investment commitment for the next ten years, not an incurred expense; the approximately USD 300,000,000 already invested is also self‑reported by the company and unaudited.
- Monthly active users are close to 5,000, and hundreds of thousands of demand letters lack a denominator and baseline, making it impossible to convert them into genuine productivity per person or unit cost improvements.
- Layoffs, per‑capita revenue and EBITDA uplift were not disclosed; the benefit description remains at process metrics such as case velocity, lacking verifiable operating results.
- Selling MX2 to other law firms means sharing capabilities with peers, which may weaken its relative advantage; data confidentiality and professional responsibility constraints in the legal industry also limit the platform’s transferability and sale pace.
Source: FDEPE tracking database. Original material: Reuters 2026-09-14 “Law firm Morgan & Morgan touts $1 billion AI investment, plans to sell platform to other firms”: reuters.com ↗; same day Bloomberg Law “John Morgan Will Spend $1 Billion on AI, Build Tool for Sale”. The figures in the text are the company’s own statements and have not been audited.