What Happened?
Overall Thesis
Morgan Stanley's weekend note on Tesla's unprecedented two-city same-day unsupervised robo-taxi launch validates the Q4 2025 rollout timeline and calls out the autonomy flywheel: more unsupervised miles → better model → higher FSD attach rates → auto margins → funds physical AI. Tesla is the only company making money selling EVs at volume anywhere; the bull thesis expands when unsupervised FSD rolls out to ~10M existing vehicles globally. Legacy auto (Farley at Ford, pivoting to 'beat China') has run out of coping strategies.
Narratives
Morgan Stanley's post-launch analyst note frames Dallas/Houston unsupervised launch as 'tangible progress' against growing skepticism — and lays out the autonomy flywheel: robo-taxi miles improve the model → faster unsupervised FSD for consumers → higher FSD attach rates → better auto margins → funds the physical-AI roadmap. Tesla is still the only EV maker making money at volume; the Model Y outsold all ICE + EV in China in March 2026 despite a 3-4x price premium. Once supervised FSD launches in China (pending), the demand pull is enormous.
Key Arguments
- Morgan Stanley: two-city same-day unsupervised launch = 'material evolution' from Austin; validates Q4 2025 rollout timeline
- Tesla on track for six more cities (Phoenix, Miami, Orlando, Tampa, Las Vegas + already-launched) in H1 2026
- Cybercab production ramping now at Giga Texas
- Model Y was #1 passenger vehicle in China in March 2026 (all EVs + ICE) despite 3-4x higher average selling price vs top 10
- Autonomy flywheel: unsupervised miles → better model → higher FSD attach → auto cash flow → funds physical AI
- Legacy auto (Ford's Jim Farley) has pivoted to 'beat China' — implicit acknowledgment they can't match Tesla
Hedges & Caveats
- Tesla has warned that timelines are 'not always right' — but the robo-taxi rollout is on track
- Cybercab still ramping from low volumes