Daily Podcast Summary -- March 21, 2026
Urgent and Timely
The Iran war is reshuffling safe haven dynamics and driving rates higher. The US dollar has regained safe haven status after losing appeal during 2025's tariff turmoil. Gold is underperforming despite its multi-year uptrend -- investors are liquidating profitable gold positions for dollar cash to meet near-term obligations, and the Strait of Hormuz closure makes physical movement of gold costly and difficult. The 10-year Treasury yield rose to 4.32% (up from 3.97% pre-war), and futures now predict only one Fed rate cut this year, down from three before the war. The 30-year mortgage rate climbed to 6.36%.
Treasuries face a push-pull. Their safe haven appeal is offset by inflation risk from oil spikes, war-related fiscal spending, and ramped-up missile production costs. Short-duration treasuries may be preferable to limit inflation exposure while retaining safe haven benefits. In extreme fear scenarios (VIX spikes), investors may accept negative real returns just for principal certainty.
Massive divergence under the surface of a calm-looking market. The S&P 500 is only down about 3% YTD, but underneath: 57 stocks are up 20%+ and 47 are down 20%+. Tech software (IGV ETF) has dropped 20%, while energy is up 30%. Sector and stock selection matter enormously right now. Financials are down 10%, partly on private credit market fears.
Stocks and Companies to Watch
- IGV (iShares Expanded Tech-Software Sector ETF): Down 20% YTD -- significant underperformance worth monitoring for potential entry or further downside.
- Energy sector: Up 30% YTD, the clear market winner. Energy exposure serves as an implicit hedge on war escalation.
- Financials sector: Down 10% YTD, worst-performing sector, partly driven by private credit fears.
- 10-year Treasury: Yielding 4.32%. Rising yields are a global trend.
- Gold: Underperforming during the crisis due to liquidity needs and physical movement constraints -- may present a buying opportunity for long-term holders with sufficient cash reserves.
- Bitcoin: Gets a mention as an asset that, unlike gold, can be moved digitally anywhere -- a potential advantage when physical asset movement is constrained by conflict.
- Target date funds (2030/2035 vintages) from American Funds, BlackRock, Fidelity, T. Rowe Price, and Vanguard referenced for retirement allocation guidance.
Investment and Portfolio Strategy
Diversify more broadly than you think. Motley Fool CEO Tom Gardner now recommends owning at least 50 individual stocks (up from the previous recommendation of 25) to ride out volatility in a richly valued market expected to deliver below-average returns over the next decade.
If you are within 10 years of retirement, reduce risk now. Research shows that portfolio returns in the years immediately before retirement have the greatest impact on retirement spending outcomes -- even more than returns after retirement. Target date funds for near-retirees hold 50-67% stocks, suggesting those in the "retirement risk zone" should consider dialing back to that range.
Gold's dip may be a long-term buying opportunity. Its underperformance is driven by short-term liquidity needs and physical movement constraints, not a fundamental change in its value proposition. The de-dollarization structural trend (global reserve managers diversifying from dollars into gold since 2022) has not reversed -- short-term crisis dynamics are just temporarily favoring the dollar.
Factor in rising borrowing costs for any property decisions. With mortgages at 6.36% and only one rate cut expected, plan accordingly for any real estate moves.
AI and Career Strategy
AI agents fail 97.5% of real-world work -- but the fix is not technical. Scale AI tested frontier agents on 240 real Upwork freelance projects and found a 97.5% failure rate on work a paying client would accept. The gap is not capability but institutional context: agents operate on timescales of hours while real jobs require months or years of accumulated knowledge. A technically competent AI coding agent wiped out a production database with 1.9 million student records because it lacked the context to distinguish production from temporary infrastructure.
The highest-leverage career move right now: become the keeper of institutional context. Harvard data from 62 million workers shows junior employment dropped 8% at AI-adopting firms while senior employment rose. The market is pricing context as the scarce resource. Your career security lies in holding organizational knowledge that agents cannot access, not in competing on task speed.
Start writing evaluations (evals) as a core competency. Evals encode senior judgment into guardrails that keep agents from going off the rails. Most companies either write no evals or write shallow ones. A good eval can be as simple as "before destroying any cloud resource, verify it is not tagged as production." You do not need to be an engineer -- you need to know your domain well enough to articulate what must be true for an output to be safe.
Companies are already regretting AI-driven layoffs. Gartner predicts half of companies that cut staff for AI will rehire for similar roles by 2027. Forrester found 55% of employers regret AI-driven layoffs. The invisible contextual stewardship humans provided was load-bearing infrastructure.
Alibaba's SWE-CI benchmark: 75% of frontier models break previously working features when maintaining code over 233 days. Early decisions compound into technical debt. If humans still have to maintain the code, the productivity gains from AI code generation are more limited than headlines suggest.
Dig Deeper
- Scale AI's Remote Labor Index vs. GDPVAL benchmark -- the gap between "can AI do this task" and "can AI do this job" and what it means for workforce planning
- Gold's underperformance during the Iran crisis -- is this a structural shift or a buying window?
- The retirement risk zone research from Think Advisor -- implications for your personal allocation if retirement is within a decade
- Private credit's impact on financials sector performance and whether the 10% drawdown is overdone
- How to build an eval framework for your organization's AI deployments before an agent causes a production incident
- Treasury yield trajectory if the war extends -- at what yield level do treasuries become attractive again despite inflation risk?