Goldman shares latest updates on Tesla’s Optimus project By Investing.com

Goldman Sachs reports steady progress at Tesla on the Optimus humanoid robot, with meaningful advances in complex hardware (hands and forearms) and continued in-house design to protect IP. AI training and translating human motion into robotic action remain central technical challenges; Tesla intends to deploy Optimus internally first and pursue gradual production before scaling later in the decade. Commercialization and sizable market upside hinge on improvements in software validation, hardware costs and real-world reliability, so material revenue impact is likely long-term and uncertain.

Analysis

Physical-AI commercialization (humanoid robots + autonomous systems) will reallocate margin and capex across the automation stack: winners are suppliers of high-torque, power-dense actuators, compact power electronics and energy storage that scale linearly with unit volumes; losers are low-margin contract manufacturers and small design shops that can’t meet repeatability or certification requirements. Tesla's choice to keep critical design in-house raises the bar for suppliers—expect a bifurcation where a small set of vertically-integrated suppliers capture most of the economic upside within 12–36 months while the rest get squeezed or forced into take-private/roll-up scenarios.

The near-term value-creation path is inside-out: internal deployments and factory use cases will determine commercial viability before any significant external revenue shows up, so tangible revenue/cost inflection points are 6–24 months away (supplier contracts, pilot deployments, published reliability metrics). Key tail risks that would reverse the thesis are regulatory incidents or a demonstrable mismatch between energy density/actuation cost and required TCO for common industrial tasks—either can push a multi-year commercialization timeline out by 3–5 years.

For portfolio positioning, favor exposure to the compute and industrial incumbents that benefit from incremental automation while avoiding or shorting speculative pure-play robotics names with high burn and low order visibility. Use asymmetric option structures (long-dated call spreads, short near-term premium) to express multi-year optionality while monetizing short-term hype; size handheld tactical trades small (1–3% NAV) and reserve 3–5% NAV for multi-year optionality if a clear supplier contract or production-signature appears.

Key Decisions for Investors

• Buy NVDA 18–30 month call spread (LEAP buy/sell) to capture secular demand for AI training infrastructure: express long GPU exposure with capped cost. Size 1–2% NAV; target >2.5x payoff if AI training intensity remains elevated over 24 months; stop-loss: 50% of premium if NVDA misses data-center guide or macro weakens.
• Overweight industrial automation incumbents (ABB, FANUY) for 6–12 months — tactical 2–3% NAV position. Thesis: if internal deployments scale, order flow into established industrial robotics/electromechanical suppliers should re-rate; upside target +25–35%, downside -12–15% if adoption stalls.
• Pair trade (9–12 months): Long FANUY (or ABB) / Short ARKQ (equal dollar). Rationale: capture rotation from speculative robotics ETF names into blue-chip automation suppliers as commercialization risk falls. Risk/reward: asymmetry if spread widens (30%+ target); downside if risk-on inflates ARKQ (limit position to 1–2% NAV and hedge with put protection).
• Options tactic on TSLA — sell 1–3 month OTM call spreads (covered if you own shares) to harvest premium against near-term hype, while buying a 18–36 month TSLA call spread to retain long-term optionality. Keep short-dated premium exposure <1% NAV; long-dated spread 1–2% NAV. This creates positive carry and asymmetric upside if commercialization milestones materialize but protects against hype-driven short-term spikes.