20VC: How Model Performance is Plateauing | Two Key Rules for Effective Deal-Making | Company Building Lessons from Keith Rabois, Brian Halligan and Pat Grady | Why Enterprise AI Adoption is Years Off with Harvey CEO Winston Weinberg
20VC: How Model Performance is Plateauing | Two Key Rules for Effective Deal-Making | Company Building Lessons from Keith Rabois, Brian Halligan and Pat Grady | Why Enterprise AI Adoption is Years Off with Harvey CEO Winston Weinberg
20VC: How Model Performance is Plateauing | Two Key Rules for Effective Deal-Making | Company Building Lessons from Keith Rabois, Brian Halligan and Pat Grady | Why Enterprise AI Adoption is Years Off with Harvey CEO Winston Weinberg
Building a successful AI-driven enterprise company demands more than technical innovation—it requires disciplined leadership, strategic hiring, and a deep understanding of long-term value creation. In this conversation, Winston Weinberg shares hard-earned insights from scaling Harvey, a high-growth B2B platform transforming professional services with AI.
Winston emphasizes daily physical discipline as a foundation for mental resilience, advocating for routines like fasted morning runs to maintain clarity under pressure. He highlights the importance of focusing on top-priority tasks as companies scale, distinguishing real progress from mere activity. Fundraising success with top-tier firms like Sequoia and a16z came from early relationship-building, cold-emailing OpenAI with AI-generated legal questions to spark interest, and moving quickly in later rounds. VCs often fail at hiring guidance, so mission alignment and ownership mindset in candidates—especially in Europe, where gardening leave extends timelines—are critical. Enterprise AI adoption lags in productivity gains due to integration complexity, requiring shifts from sales-led to post-sales success models. Contrary to fears, AI won’t eliminate law firms but will expand their role in advising on AI risk and regulation. Finally, effective deal-making means knowing when to walk away, prioritizing irreplaceable talent over financial optics, and staying committed to principled, long-term vision.
10:13
10:13
A valuation multiple of 100 is iffy; 20–25 feels reasonable for end-of-year revenue.
25:05
25:05
Even if frontier models stopped improving, AI revenues could still explode due to low integration.
36:59
36:59
High ROI from AI products reduces the 'hostage' effect seen in legacy enterprise software
38:13
38:13
AI shifts spend from human labor to technology budgets
41:56
41:56
AI won't destroy jobs; it creates new work in legal and regulatory fields
46:07
46:07
Ownership allows employees to identify and fix problems independently as the company scales
51:45
51:45
AI researchers care about working on hard and interesting problems and will leave if company direction changes.
54:17
54:17
Listen more than you speak, as deal-making is about understanding people
1:00:10
1:00:10
Used AI-generated legal questions from a subreddit to cold-email OpenAI leadership
