Investment Bankers Built A $2B AI To Replace Themselves | Term Sheet

Key Concepts

• Rogo: An AI platform designed to automate high-finance workflows, specifically for investment banking.
• Agentic AI: AI systems that can "think before they act," using tools to perform multi-step, end-to-end workflows (e.g., querying data, updating Excel, sending emails).
• Partner and Compete Model: The dynamic where AI startups utilize foundational models (OpenAI, Anthropic) while simultaneously building specialized, competing applications on top of them.
• Semantic Parsing: An older, rule-based technique for interpreting natural language, which has been superseded by deep learning and Large Language Models (LLMs).
• High Finance Reinvention: The shift from manual, labor-intensive "factory" models of banking toward an apprenticeship-based model focused on high-trust, relationship-driven advice.

---

1. Main Topics and Key Points

• The Evolution of Investment Banking: Gabe Stangle, founder of Rogo, argues that the "drudgery" of banking—valuation, data gathering, and PowerPoint creation—is being automated. This allows bankers to focus on the "human" elements: negotiation, strategy, and relationship management.
• The "Step Change" in AI Capabilities: Stangle identifies three technological leaps:
  1. GPT-3/3.5/4: Enabled natural language generation.
  2. Reasoning Models (e.g., o1): Allowed models to "think" before acting, enabling the use of external tools.
  3. Agentic Workflows: The ability to chain hundreds of actions (querying data from FactSet/S&P, updating models, drafting communications) to complete a week-long task in minutes.
• Market Impact: Stangle suggests that AI will democratize access to capital markets, potentially allowing smaller, entrepreneurial bankers to serve the 300,000+ American businesses currently underserved by traditional Wall Street firms.

2. Real-World Applications

• M&A Processes: Rogo aims to reduce the time required to consummate an acquisition from months to 48 hours by automating valuation and buyer outreach.
• Lazard Case Study: Lazard, the firm where Stangle began his career, is now a paying customer. They utilize Rogo to integrate AI into their existing workflows, demonstrating that even storied, traditional institutions are actively seeking to reinvent their operating models.

3. Methodologies and Frameworks

• The "Human-in-the-Loop" Philosophy: Despite automation, Stangle emphasizes that high-stakes transactions (like buying a home or a multi-billion dollar company) remain emotional and personal. AI handles the "blocking and tackling," while humans handle the trust and negotiation.
• The "Build vs. Partner" Dilemma: For large banks, the choice is between building custom AI agents in-house or partnering with specialized platforms like Rogo that are already integrated into financial systems of record.

4. Key Arguments and Perspectives

• The "Flash Boys" Parallel: Stangle compares the current AI shift to the automation of the New York Stock Exchange floor. While the nature of the jobs changed, the volume of market participation exploded. He argues that while specific "analyst" tasks will disappear, the aggregate number of finance professionals may actually increase as the market expands.
• The "Consensus" Trap: Stangle notes that early venture capital investors rejected Rogo because they viewed it through the lens of their own current needs, failing to see the long-term vision of a transformed industry. Keith Rabois (Khosla Ventures) was the sole investor to back the company, favoring the contrarian bet.

5. Notable Quotes

• "There's certain types of workflows that will never need to be done by a human again, and I think that's a good thing." — Gabe Stangle
• "I don't like consensus bets. I like to be contrarian." — Keith Rabois (as recounted by Stangle)
• "A deal is never just a deal. It's never just about money changing hands. It's emotional." — Gabe Stangle

6. Data and Research Findings

• Goldman Sachs Case Study: In 2000, the New York equity trading desk had 600 traders. By 2017, only two remained, supported by 200 engineers. Stangle uses this to illustrate that while specific roles vanish, the organization survives by reinventing its core functions.
• SpaceX IPO: The video notes that SpaceX is preparing for what is expected to be the largest IPO in history, with revenue rising alongside multi-billion dollar losses and a unique pay package for Elon Musk tied to colonizing Mars.

7. Synthesis and Conclusion

The transition to AI in high finance is not merely about cost-cutting; it is a fundamental shift in the "arteries of innovation." By automating the analytical drudgery of investment banking, AI is enabling a return to the original purpose of the industry: the financing of growth and innovation. While the "analyst" role as it existed 20 years ago is becoming obsolete, the industry is poised for a period of entrepreneurial expansion, where AI-enabled bankers can reach previously untapped segments of the global economy. The ultimate takeaway is that the most successful firms will be those that use AI to move away from "factory-model" deal-making and back toward high-trust, relationship-based advisory services.