David Autor 2.2 12 ideas

"You could also use that capability to say we're going to make healthcare more accessible and less error-prone or we're going to make education more affordable and more engaging." While the "bad" scenario is pure automation, the "good" scenario (which may be supported by government grants/policy) involves AI-augmented services. Companies positioning themselves as "AI Tutors" (Duolingo) or "AI-assisted Diagnostics" (Healthcare) align with the social/political desire to improve services rather than just cut costs. Long sectors where AI improves outcomes (Education/Health) rather than just replacing workers, as they face less regulatory tail-risk. Public sector funding or adoption speeds in healthcare/education are notoriously slow.

"If that supporting work is automated by AI... you're happy... It makes your expertise more valuable, allows you to specialize." Autor specifically cites lawyers: "Those that are left get paid more because the high expertise part of what they do is even more valuable." The "surviving" high-end professionals will rely heavily on proprietary data and AI tools to eliminate grunt work (drafting contracts). Companies that control the legal data moats and the AI tools to process them (Thomson Reuters, RELX) become essential infrastructure for the high-margin professional services industry. Long the "arms dealers" of professional expertise who sell the tools that allow lawyers/accountants to charge more for strategy while doing less drafting. Open-source models (LLMs) becoming good enough to bypass specialized proprietary databases.

"Think of how ride hailing software changed the taxi driving industry... it commodified the expertise of knowing your way around... barriers to entry fall." He also notes billions of dollars invested in "self-driving cars." Autor describes a mechanism where technology owners (platforms) capture value by commoditizing labor. While this is bad for the *wages* of the workers, it is incredibly bullish for the *platforms* (Uber, Tesla/Robotaxi) that can scale supply infinitely without relying on specialized human skills. Long the platforms that successfully use AI to lower barriers to entry for service provision. Regulatory backlash if wages fall too low or if "gigification" of the economy is halted by labor laws.

"I don't think the incentives of the private sector are ideally aligned... its objectives are much, you know, around reducing labor costs, right? People want to automate." Autor confirms that the "San Francisco Consensus" (tech leadership) is aggressively pursuing labor cost reduction. This implies sustained, high-volume spending on enterprise software that promises efficiency gains and headcount reduction (Microsoft Copilot, Salesforce, ServiceNow). Long the enterprise software vendors selling the "efficiency" narrative to CFOs. "AI disillusionment" if productivity gains do not materialize quickly enough to justify the software costs.

"There are many, many signs that we have long-term labor shortage. So, we're not going to run out of jobs." If there is a structural labor shortage, the cost of human labor will remain high or supply will be insufficient. This forces companies to invest in automation not just for efficiency, but for survival. This guarantees a long-term CAPEX cycle for robotics and AI automation hardware. Long automation providers as the solution to demographic decline. A deep recession could temporarily loosen the labor market, reducing the urgency for automation CAPEX.

Autor uses ride-hailing as the primary example of how technology commodifies expertise. He notes that GPS/Apps allowed anyone to drive, which "lowered wages of people in the taxi driving industry" but expanded the service. AI/Tech removes the "moat" of human expertise (knowing the streets), increasing the supply of labor. This suppresses labor costs (bullish for the platform's margins) while increasing service availability and volume (bullish for the platform's revenue). LONG the platform (UBER) that benefits from commoditized, cheaper labor inputs. Regulatory crackdowns on gig-worker classification.