AI Is a Services Revolution

🌈 Abstract

The article discusses how the economy has shifted from agriculture and manufacturing to a services-based economy, and how this transition is ripe for disruption by AI. It explores the potential impact of AI on various services industries, such as legal, healthcare, education, and banking, and provides a framework for building successful AI startups in these domains. The article also discusses the current state of AI adoption in enterprises and the role of consultants in helping companies navigate the AI landscape.

🙋 Q&A

[01] Weekly writing about how technology and people intersect

1. What are the author's day and night activities?

• By day, the author is building Daybreak to partner with early-stage founders.
• By night, the author is writing Digital Native about market trends and startup opportunities.

2. What is the purpose of the article? The article discusses how technology, particularly AI, is intersecting with and disrupting various services industries.

[02] One funny thing about Large Language Models

1. What is the key capability of Large Language Models (LLMs)? LLMs are very good at language, as they work by digesting and synthesizing large amounts of text to generate outputs.

2. How does the author relate the capabilities of LLMs to the shift in the economy? The author notes that as the economy has shifted from agriculture and manufacturing to a services-based economy, the language-centric nature of many services industries makes them ripe for disruption by AI and LLMs.

[03] The shift in the U.S. economy

1. What are the key industries that have grown in the U.S. over the past 50 years? The key industries that have grown are healthcare, education, financial services, business services, and professional services.

2. How does the author describe the shift in the types of jobs in the U.S. over time? The author notes that in the 1950s, fewer Americans were working on farms and more were working in factories, whereas in the 2020s, fewer Americans are working in factories and more are working in desk-centric knowledge work.

[04] The threat of AI automation

1. What industries does Goldman Sachs expect to be most impacted by AI automation? According to Goldman Sachs research, the industries most exposed to AI automation are administrative office work, legal work, and architecture & engineering.

2. How does the author describe the pace of AI adoption in enterprises? The author notes that enterprises are slow to adopt new technology, with most not expecting to deploy AI solutions before 2026. There is a distinction between pilots/experiments and actual deployments, with most enterprises focused on the former rather than the latter.

[05] The opportunity for AI startups

1. What is the author's formula for a compelling AI startup? The author suggests that a successful AI startup should: 1) Pick a large, text-heavy services industry, 2) Use LLMs to automate workflows and augment workers, and 3) Leverage industry-specific data to fine-tune the model over time.

2. What are some examples of AI startups disrupting various services industries? The author provides examples of AI startups in the legal, healthcare, education, and banking industries that are using language-based AI to automate workflows and augment workers.

[06] The Seurat playbook for vertical AI

1. What is the Seurat playbook for vertical AI startups? The Seurat playbook involves: 1) Identifying the most salient pain-point for customers and solving it, 2) Using that wedge to win over customers, and 3) Layering in more products over time to cross-sell and up-sell, improving customer lifetime value and becoming more defensible.

2. How does the author relate the Seurat painting technique to vertical SaaS and AI startups? The author draws a parallel between Seurat's pointillism technique, where individual dots come together to form a rich, expansive work, and how vertical SaaS and AI startups can start with a seemingly niche feature but over time layer in more products to become a one-stop-shop for their customers.

[07] The state of the AI revolution

1. How does the author describe the current stage of the AI revolution? The author suggests that we are still in the "Irruption Phase" of the AI revolution, where early products are emerging and we are seeing fast innovation, but the true winners are not yet clear.

2. What is the author's perspective on the longevity of early AI movers? The author cautions that, like the dotcom bubble, early movers in the AI space may not necessarily be the ones that ultimately win the race, and it will take time to fully understand how LLMs should best automate workflows and augment workers.