They graduated from Stanford. Due to AI, they can’t find a job

AI-driven coding tools are materially depressing entry-level hiring in tech, with a Stanford study showing employment for early-career software developers (ages 22-25) down nearly 20% from its late-2022 peak and entry-level hiring in AI-exposed roles falling about 13% relative to less-exposed jobs. Major tech firms (e.g., Meta, Block, Autodesk) have cut staff while startups and AI labs expand capacity—Anthropic’s CEO says 70–90% of some product code is produced by its AI—and investors are warning of potential overspending on AI. The disruption is pushing graduates toward additional degrees, lower-tier employers or startups, and shifting engineering work toward oversight and AI management, a structural change with medium-term implications for tech labor supply, productivity and investor positioning.

Analysis

Market structure: AI-driven coding productivity concentrates value with AI infrastructure (GPUs, cloud, LLM operators) and experienced engineers while compressing wages and hiring for entry-level devs (Stanford study: ~20% drop for 22–25 year-olds). Expect pricing power to shift to dominant model providers and hyperscalers; mid‑market consultancies and volume-focused dev shops are direct losers. On macro, persistent wage pressure for juniors is a modest disinflationary force for services wages but rising AI capex (hardware, data centers) supports tech capex and corporate credit spreads in the medium term.

Risk assessment: Near-term (days–weeks) risks include earnings guidance cuts from software firms and hiring freezes; medium-term (3–12 months) tail risks include regulatory constraints (privacy/audit rules) or major model failure outages that trigger liability and reputational costs. Hidden dependencies: venture funding cycles and university enrollment surges (more MS grads) can lengthen the labor supply shock for 12–36 months. Catalysts that could accelerate reversal: clear regulatory guardrails (6–24 months) or a sharp fall in LLM accuracy causing re-hiring.

Trade implications: Favor long positions in AI infrastructure (GPUs, cloud) and cybersecurity; avoid or short headcount‑intensive software names and software consultancies exposed to junior wage deflation. Use put spreads on exposed public software (e.g., ADSK) 3–6m to capture margin hits; hedge big-cap AI spenders (e.g., META) with small-duration puts around earnings windows. Rotate 5–10% of small/mid-cap SaaS into infra/cyber over next 4–12 weeks.

Contrarian angles: Consensus understates winners among talent platforms, premium contracting marketplaces and QA/AI‑oversight tooling — these can see revenue CAGR uplift as firms buy human oversight. Reaction may be overdone for deep-pocketed AI investors (META) that can monetize models via ads and commerce; historically automation depressed entry roles but expanded skilled demand over multi‑year cycles (industrialization analog). Unintended consequences include political/regulatory backlash and concentration risk that could revalue large incumbents either up or down depending on policy.

Key Decisions for Investors

• Establish a 2–3% long position in NVIDIA (NVDA) or equivalent AI-infrastructure exposure within 2–6 weeks; target +25–40% upside over 12 months, set stop-loss at -20% to hedge GPU demand cyclicality.
• Initiate a 1–1.5% notional 3–6 month put-spread on Autodesk (ADSK) sized to capture a 10–20% downside on margin/hiring guidance misses (buy 10–15% OTM puts, sell deeper OTM puts) and close on the next earnings release.
• Buy a tactical 1% notional 4–6 month put on META (META) around its next quarterly earnings as insurance against AI capex surprise or ad-revenue weakness; trim if guidance normalizes or implied vol collapses by >30%.
• Rotate 5% of small/mid-cap SaaS and staffing-exposed positions into: 2–3% cybersecurity (e.g., HACK or CHKP), 2% talent/contractor platform (e.g., UPWK) over the next 4 weeks; exit if entry-level hiring recovers >10% quarter-on-quarter or if EU/US AI regulation with heavy compliance >90 days is passed.