In legal AI, should companies make incumbent law firms software-powered partners, or build AI-native competitors that attack the service layer directly?

Emad Mostaque · May 17, 2023

Harvey's approach — going to a law firm, using it as distribution, and building your system inside their system — is the right model for AI startups

Rather than retargeting an existing market, you go in and transform an incumbent, and organizations are elastic and plastic enough right now to let you do it as a skunkworks lab

33:16 20VC: Why the AI Bubble Will Be Bigger Than The Dot Com Bubble, Why AI Will Have a Bigger Impact Than COVID, Why No Models Used Today Will Be Used in a Year, Why All Models are Biased and How AI Kills Traditional Media with Emad Mostaque, Founder & CEO @

Aatish Nayak · Apr 11, 2025

AI products should partner with the producer of professional work to serve the end consumer rather than going direct to the consumer

Humans don't simply trust AI — they trust other humans using AI

Scope: depends on the domain; framed as the more likely right answer, not universal

51:52 20Product: How Scale AI and Harvey Build Product | Why PMs Are Wrong: They are not the CEOs of the Product | How to do Pre and Post Mortems Effectively and How to Nail PRDs | The Future of Product Management in a World of AI with Aatish Nayak

Max Junestrand · Aug 15, 2025

Selling AI into law firms is not cannibalising the firms — you partner with them so they win and you take a share of their upside

The service market is where the value is, so helping firms deliver work far faster (e.g. ten hours to fifteen minutes in an AM Law 200 pilot) creates a bigger pie to share rather than destroying the customer

12:47 20VC: 15 Term Sheets in 7 Days and Choosing Benchmark | Harvey vs Legora: Who Wins Legal and How to Play When You Have $600M Less Funding | Are AI Models Plateauing Today | Building a 9-9-6 Culture From Stockholm with Max Junestrand

Max Junestrand · Aug 15, 2025

There is a huge share of value in the application layer, especially on the service side rather than only software

The legal software market is $20B while the legal services market is a trillion, so the right strategy is to work with the best firms to play for the service market

11:45 Trillion dollar service market dwarfs 20b software market so target services via partnership

Max Junestrand · Aug 15, 2025

The winning play in legal AI is making the best law firms and attorneys software-powered, not building AI-native competitors that attack the service layer directly

Companies like Crosby are playing for the pure service part; Legora's differentiated bet is that the best attorneys go software-powered

13:47 Software powered incumbents beat ai native attackers

Harry Stebbings · Aug 15, 2025

To build a $100B business in legal you have to be software-powered; building the next-generation law firm gives more certainty but a capped upside

14:57 Software powered required for 100b scale pure service play caps upside