Terminator, AI Jobs, and Financial Nudism

The episode opens with a discussion of bond market dynamics. Host Jack Howe explains that while bonds have underperformed over the past decade due to historically low yields, the current 5% yield on 10-year treasuries provides better return prospects going forward. He uses the concept of "financial nudism"—a stripped-down, minimalist investment approach—to frame the discussion, then addresses listener Tony's question about whether bonds require more active management now. Howe recommends four alternatives for bond investors: emerging market bonds (VWOB), fallen angel bonds (FALN), commodities (COMT), and collateralized loan obligations (JAAA). He explains that bond index weighting by issuance is counterintuitive, as it overweights the largest borrowers rather than highest-quality credits. Howe clarifies the concept of "inflating away debt" by using a historical example: homeowners with mortgages from the 1970s saw their fixed payments become trivial by 2000 due to inflation eroding the real value of debt. This would happen to government debt if policymakers pursue sustained inflation, but comes with costs including higher living expenses and erosion of bond investor trust. The second half features Ronnie Chatterjee, Chief Economist at OpenAI, discussing AI's economic implications. Chatterjee explains his role involves research on labor markets, enterprise impacts, and extreme AI scenarios; external communication; and internal learning at OpenAI. When asked about job market effects, he argues the unemployment rate is too coarse a metric—instead, work composition is changing as workers use AI for tasks outside their job descriptions, creating new roles. He advises college students to develop human skills in areas requiring accountability, leadership, and interpersonal connection (healthcare, education, legal roles), while not dismissing STEM fields. On forecasting AI's long-term impact, Chatterjee expresses caution about 10-15 year predictions, emphasizing they lack probabilities and thus limited practical utility. He points to emerging data suggesting productivity gains are appearing: Federal Reserve district studies showing companies reporting AI productivity benefits in earnings calls, case studies like Lowe's improving same-store sales, and widespread but shallow AI adoption (chatbots and calculators rather than "agentic coworkers"). He references the 1990s productivity paradox ("computers everywhere except in the productivity statistics") as a parallel, warning that measurement issues may obscure AI's actual productivity impact. Chatterjee advocates releasing granular data to identify "canaries in the coal mine" rather than making confident long-term forecasts.