2 Seemingly Unstoppable Artificial Intelligence (AI) Stocks That Can Plunge Up to 96% in 2026, According to Select Wall Street Analysts

RBC analyst Rishi Jaluria warns Palantir (PLTR) could fall to $50 by 2026 (≈72% downside from a ~$178 close on Dec. 4), citing an extreme trailing-12-month P/S of 117 and scaling challenges for its Foundry platform. GLJ Research founder Gordon Johnson projects Tesla (TSLA) could drop to $19.05 (≈96% downside), pointing to a forward P/E near 200, limited monetization from FSD and Optimus, and possible 2025 sales declines despite ~1.8M annual EV deliveries and five years of GAAP profitability. Both calls highlight valuation-driven downside risk for two widely held AI/EV names and could prompt investors to reassess positioning on growth/valuation exposure.

Analysis

Market structure: The article highlights stretched valuations in marquee AI/EV names (PLTR trailing P/S ~117; TSLA forward P/E ~200) that create clear winners — providers of scalable AI compute and enterprise recurring‑revenue software (e.g., NVDA, MSFT, ORCL) — and losers: single‑product, high‑customization platforms and hardware‑heavy EV plays that cannot convert hype into margin expansion. A re‑rating (20–60% multiple compression) would reallocate capital from thematic growth into cash‑flowing software and semiconductors over 6–24 months, compressing financing for speculative small‑caps immediately.

Risk assessment: Tail risks include a regulatory/autonomy liability shock for Tesla or a major contract cancellation/shift in US defense procurement for Palantir; both are low‑probability but >10% portfolio‑impact events within 12–24 months. Near term (days–weeks) expect elevated IV and headline‑driven swings; medium term (quarters) catalysts are FY26 guidance, backlog disclosures, and FSD/Optimus deliveries; long term (years) is actual AI adoption vs. personalized implementation limits for Foundry.

Trade implications: Favored executions are hedged, asymmetric short exposures: buy‑put spreads and pair trades (long NVDA/AI infrastructure, short PLTR/TSLA) to capture rotation while capping downside. Cross‑asset: expect rising equity IV, modest widening in high‑yield/BBB spreads on a 20–30% tech reset, and USD safe‑haven flows; hedge equity shorts with buying IG protection only if systemic contagion emerges.

Contrarian angles: Consensus underestimates durability of government lock‑ins for PLTR and Tesla's cash generation — shorts should be sized and time‑limited; conversely, consensus overprices perpetual growth for both. Historical parallels: dot‑com multiple busts show 40–80% drawdowns on narrative names but also multi‑year recoveries for durable business models; structure trades to monetize reversion without betting on total collapse.

Key Decisions for Investors

• Establish a tactical short on PLTR sized to 1.5% of fund NAV using a cost‑limited put spread: buy PLTR Jan 2026 60/40 put spread (long 60, short 40) and allocate max premium = 1.5% NAV; add if futures news shows >10% downward revision to FY26 guidance within 90 days.
• Protect/express bearish view on TSLA with a tail hedge: buy TSLA Jan 2026 100/30 put spread sized to 1.0% NAV (limits premium) and trim if TSLA shares fall >40% or if FSD/Optimus tangible revenue >$2B consensus in next 12 months.
• Implement a relative‑value pair: go long NVDA equal‑delta to a 2.0% NAV position and short TSLA 1.5% NAV (delta‑neutralize) over 3–9 months to capture AI capex re‑allocation; scale into NVDA on any pullback ≥10% within 90 days.
• Reallocate portfolio weight: reduce exposure to single‑product/high P/S AI names (e.g., trim PLTR positions by 50% if P/S>50) and redeploy proceeds into enterprise SaaS (MSFT +2% NAV, ORCL +1% NAV) with target holding period 12–36 months to capture recurring revenue re‑rating.