We took a blended approach at
@easepractice with our MSO-PC stack, which is a mix of AI-native legal compliance services human professional services.
It can costs a practice >$200k to form a 50 state MSO-PC with nearly $500k-$1MM in ongoing maintenance costs annually. 😬
The global professional services market (law, consulting, accounting, marketing, etc.) is worth over $6 trillion annually. It’s a massive white-collar sector, and AI-native firms are emerging to replace or radically reshape it.
The end state is relatively clear: companies will interface directly with intelligent agents to get work done. But the path there is still uncertain, since the underlying tech isn’t yet robust enough to handle complex workflows end to end.
Three main strategies to attacking this huge market are emerging:
⸻
1. Sell to existing service firms (e.g., Harvey)
Pro: Highest distribution; large firms already own the customer relationships
Con: End customers will eventually prefer working directly with in-house agents instead of outside entities
Harvey provides AI tools for major law firms, augmenting workflows like contract analysis, drafting, and legal research.
⸻
2. Build an AI-native service firm (e.g., Crosby)
Pro: Best long-term product experience, fully designed around AI-first delivery
Con: Tough distribution; starting from zero without the brand or pipeline of traditional firms
Crosby is rebuilding management consulting from the ground up as an AI-native firm, handling research, synthesis, and strategy work with a human-in-the-loop model that prioritizes speed and quality over headcount.
⸻
3. Roll up legacy firms and augment with AI (e.g., Crete)
Pro: Medium distribution; you get trusted client relationships plus the ability to embed AI deeply
Con: Capital-intensive and operationally complex; requires transforming legacy systems while scaling new ones
Crete is acquiring accounting firms across the United States and embedding OpenAI-powered tools into their workflows, streamlining tasks like audit prep, memo drafting, and data reconciliation at scale.
⸻
The winning strategy will depend on:
• How quickly AI capabilities improve
• Whether clients prioritize quality or familiarity
• Who solves distribution without diluting the agent-first experience