How AI is Reshaping Labor Markets: A $Trillion-Dollar Opportunity Explained

Era 1 (1960s–1990s): Digitization of Physical Filing

• Software replaced physical filing cabinets (HR, medical, financial) with databases and front-end entry interfaces.
• Examples include Sabre (1959) for airline reservations, Quicken for finance, and PeopleSoft for HR.
• This era digitized information storage but did not automate the human labor required to manage or act on that data.
• The "gopher" role (staff retrieving files) was eliminated, but the headcount of knowledge workers remained static.

Era 2 (1998–2010): Cloud Migration

• Software moved from on-premise mainframes to the cloud, decoupling data from physical hardware and reducing IT overhead.
• Salesforce, NetSuite, and Zendesk pioneered cloud-based Customer Relationship Management (CRM) and support systems.
• This shift enabled financial services (payments, insurance) to bundle with software, creating viable markets for previously non-technical industries (e.g., Toast for restaurants, ServiceTitan for HVAC).

Era 3 (2024+): AI as Labor Substitution

• New software agents are now capable of performing the actual work that humans have done for 65 years (e.g., processing forms, making calls, analyzing data).
• Unlike previous eras, this shift compares software revenue directly against labor wages rather than just software licensing fees.
• The market potential expands drastically; for instance, the U.S. nursing wage market ($600B) vastly outpaces the dedicated software market ($600M).
• AI enables "input coffee, output code," allowing software to handle the actions previously required by humans on digitized records.

Pricing and Revenue Model Shifts

• Incumbents like Salesforce and Zendesk charge per seat, creating a revenue conflict where AI efficiency reduces the need for human seats.
• A "Copilot" model (productivity enhancer) risks cutting software spend by 90% if it increases individual worker output tenfold.
• An "Autopilot" model (full substitution) could reduce required seats to zero, eliminating seat-based revenue entirely.
• Future revenue models are expected to shift from seat-based licensing to charging for the "output of work" or cost savings, potentially increasing ACV by 2x to 10x if customers reallocate labor budgets to software.
• Customers may tolerate higher software spend if it replaces significantly higher labor costs (e.g., replacing $50M in labor with $1.4M in software spend).

Investment Thesis and Moats

• Wedge Strategy: Startups can enter via the "messy inbox problem," using AI to unstructured data (faxes, emails) before integrating downstream into systems of record (e.g., Tenor in healthcare).
• Defensibility: Differentiation via AI is temporary; long-term moats remain built on ownership of downstream workflows, network effects, and deep integration into the business ecosystem.
• Incumbent Risk: Major players like Salesforce face existential risk if they fail to evolve their pricing or features, as they hold the systems of record but rely on outdated pricing models.
• Market Expansion: Industries previously too small for software investment (e.g., compliance officers, manicurists, niche professional services) become viable targets when AI can substitute high labor costs.

Specific Sector Opportunities

• Financial Services & Insurance: Heavy reliance on outdated 30-year-old systems and massive labor budgets (e.g., compliance officers) presents immediate AI substitution opportunities.
• Legal Services: New models emerge where software replaces hourly billing with contingency models, automating case intake, medical chronology, and document drafting.
• Compliance: AI agents can address backlogs in transaction monitoring and bank account openings, solving the "no software, no people" bottleneck in banking.
• Obscure Industries: Founders are encouraged to target obscure verticals (mining, farming, specialized trade) where AI can solve specific, high-cost labor problems ignored by horizontal SaaS.

Economic and Employment Impact

• Technology acts as a deflationary force, driving costs down exponentially (e.g., computing power, translation services).
• Jobs requiring human relationship building (e.g., selling over golf) may see increased value as AI saturates digital communication channels.
• Most white-collar jobs will adopt "Copilot" roles, automating menial tasks and freeing workers for creative or high-value connection tasks.
• The historical pattern suggests displaced labor will transition to new, more productive roles rather than mass unemployment.

Evaluation Metrics for New Ventures

• Core investment metrics remain unchanged: customer count, retention, gross profit per customer, and overhead.
• Social media-era metrics like Daily Active Users (DAU) are still relevant for measuring engagement but must be paired with clear monetization pathways.
• The "smile curve" of usage (initial drop-off followed by plateau) is a key differentiator for successful platforms.
• New metric: The potential market size of a niche vertical expands significantly when AI can tap into latent labor budgets.

Future Outlook

• Full "autopilot" readiness is not yet universal; many complex workflows require human oversight, creating a risk of early over-hyping.
• Incumbents may fail to pivot due to pricing inertia, creating a window for new "AI-native" entrants to capture market share.
• Successful new companies will likely emerge from founders with deep domain expertise in obscure, high-labor industries.
• Generational shifts in software will occur in both verticals (niche workflows) and horizontals (sales, marketing, analytics) where AI can redefine the core product delivery.