The godfather of AI predicts mass unemployment is on its way. This CEO warns even a 10% reduction ‘will feel like a depression’

Rapid advances in AI are raising near-term labor-market risks: a 2025 MIT study cited in the article estimates AI could replace nearly 12% of the U.S. workforce, and Salman Khan warns even a 10% contraction in white-collar work would feel like a depression. Signs of corporate adoption already include roughly 55,000 AI-linked layoffs last year and Salesforce cutting 4,000 customer-service roles, while driverless vehicles threaten over 1 million rideshare jobs. Khan proposes a private-sector funded national reskilling collective (a 1% commitment of personnel costs or profits) to mitigate displacement; policymakers and investors should view this as a structural risk to consumer demand and sectoral earnings rather than an immediate market shock.

Analysis

Market structure: AI accelerates winner-take-most economics — cloud/AI infrastructure (GPUs, data centers, enterprise AI stack) and automation hardware gain pricing power while labor-heavy services (contact centers, rideshare drivers, some white‑collar roles) face margin compression. The MIT 12% replacement estimate and ~1M potentially affected rideshare roles imply durable demand shift into capex (servers, GPUs) and reskilling services, compressing unit labor costs but raising upfront capital intensity over 1–3 years.

Risk assessment: Tail risks include a near-term AGI “shock” or major model failure triggering swift regulation (AI tax, strict liability, hiring freezes) within 12–36 months that could crash valuations of high‑multiple AI plays. Hidden dependencies: corporate willingness to reallocate payroll to capex and the pace of successful retraining; catalysts to watch in next 30–90 days are large enterprise AI rollouts, major layoffs guidance, and AI policy legislation.

Trade implications: Favor long positions in AI infrastructure (GPUs, IBM enterprise stack, select cloud exposure) and shorts/insurance on labor‑heavy service providers (CRM/customer‑service platforms, gig platforms exposed to autonomous driving). Use multi‑month LEAPS on winners and put‑spreads on losers to control cost; rotate out of discretionary/consumer names if unemployment trends worsen over the next two quarterly prints.

Contrarian angles: The consensus underprices the growth in reskilling/education and healthcare staffing demand — firms offering scalable certification and staffing (edtech, healthcare staffing tech) could see multi‑year revenue tailwinds. Also, shorting broad consumer tech (AMZN) is risky: AWS durability may re-rate AMZN even as retail cuts jobs, so prefer calibrated, pair trades rather than blanket shorts.

Key Decisions for Investors

• Establish a 2–3% portfolio long in AI infrastructure exposure: buy 12‑18 month LEAPS on NVDA or, if constrained, 12‑18 month calls on IBM (target +30–50% in 12 months) to capture GPU/data‑center re‑rating driven by enterprise AI capex.
• Open a tactical 1.5–2% short position on CRM (Salesforce): implement a 3‑month put‑spread (buy 1x 10% OTM put, sell 1x 20% OTM put) to profit from near‑term margin pressure from AI replacing customer‑service roles and to limit premium outlay.
• Set a pair trade: long AMZN AWS exposure via a 6‑12 month bull call spread (~10%–30% OTM) representing 2% portfolio weight, while short 1–2% of UBER equity or buy 3–6 month puts (5–10% OTM) to hedge autonomous‑driven displacement risk in rideshare.
• Reduce sector exposure: trim consumer discretionary/transportation exposure by 25–50% within 30 days if national unemployment prints rise >0.2ppt over two months; monitor EU/U.S. AI regulation milestones in the next 60–90 days as stop/re‑evaluate triggers.