[풀버전] 월가 내부자들 사이에서 도는 테슬라에 대한 핵폭탄급 소문 (강정수 박사)

Overview

Core thesis: The biggest drivers of Tesla’s 2026 valuation are robotaxis / FSD and robotics (Optimus), supported by explosive growth in energy (Megapack). If Tesla demonstrates factory-scale mass production of Optimus (3rd gen) or commercially deployable Cybercab robotaxis, markets may re-rate Tesla from “automaker” to “AI/robotics” company.

Key near-term valuation drivers:

• Progress and revenue recognition from FSD/robotaxi.
• Energy (Megapack) growth.
• Credible signs of Optimus factory production.

---

Key technological concepts and product/features

Robotaxi (Tesla Cybercab / Robotaxi)

• Cybercab: purpose-built vehicle with no steering wheel or pedals; currently produced in very limited numbers due to regulation.
• Commercialization constraints:
  • Requires U.S. federal and state legislative and regulatory approvals (House, Senate, President, plus state cooperation).
  • Early limited deployments (1–2k units) are likely; full-scale roll-out depends on regulatory timing and safety outcomes.
• Investor signal: FSD take-rate metrics (subscription/adoption) and initial autonomous-driving revenue.

Optimus humanoid robot

• Tesla aims to release a 3rd-generation Optimus and perform final tuning at the Texas Gigafactory.
• Initial units will likely be hand-built or produced in small numbers for data collection and testing.
• Mechanical constraints and challenges:
  • Degrees of freedom: Optimus ~21 DOF vs. human ~26–27.
  • Limitations include hand strength, heat generation, durability, and “last mile” reliability for safety-critical tasks.
• Physical AI vs mechanical body: both the robot “brain” and the mechanical body must reach very high reliability. Google DeepMind’s “Gemini Robotics” is cited as a physical-AI example (claimed ~40% reliability in a cited stage).

AI infrastructure and chips

• Nvidia:
  • Dominant hardware provider with a strong software/hardware ecosystem (CUDA).
  • High margins (~60–70%) and a large developer base (~3M CUDA developers) are major moats.
• Big tech response:
  • Google (TPU), Meta, Microsoft, Amazon develop/operate in-house chips and data centers to reduce Nvidia dependence.
  • Switching costs, software lock-in, and reliance on foundries (TSMC) are constraints.
• Tesla/Elon Musk:
  • Internal chip efforts are discussed as a theoretical route to vertical integration (chips + SpaceX economics), but practical challenges remain.

Data centers, energy & power constraints

• Data center buildout is a bottleneck due to grid capacity, land, and electricity availability.
• Examples and approaches:
  • Microsoft reportedly restarting an old nuclear plant to secure power.
  • Firms are considering on-site generation or other localized power solutions.
  • Space-based data centers (SpaceX/Tesla) are sci‑fi-level concepts; they face thermal physics and launch-capacity limits and are not near-term solutions.
• Risk: Power shortages could throttle AI deployments if electricity supply and grid upgrades don’t match contracted capacity.

Data scarcity for robotics

• Robots need abundant task-level, physical-interaction datasets.
• China is reportedly collecting large-scale robotic process datasets via human operators repeating tasks, giving a potential training advantage in physical-AI.

---

Analytical takes, risks, and market impacts

• Nvidia likely remains dominant in the near-to-medium term, though compression over time is possible; CUDA ecosystem and developer lock-in are significant defenses.
• Regulatory, safety, and public-perception risks are critical for robotaxi roll-out; a single major incident could materially affect stock and regulation.
• Elon Musk is both Tesla’s greatest asset and a source of unpredictable political/behavioral risk, contributing to stock volatility.
• Broader AI economy effects:
  • Productivity gains (e.g., accelerated developer output) can drive cost savings and shift corporate capex.
  • Overinvestment concerns persist among investors.

---

Practical guidance and recommendations

For investors

• Watch these primary 2026 catalysts:
  1. FSD take-rate metrics and FSD-related revenue.
  2. Robotaxi legislation progress.
  3. Optimus production signals (especially 3rd-gen demonstrations and factory-scale indications).
  4. Energy/Megapack sales growth.
• Monitor ancillary indicators:
  • Big-tech capex and Nvidia contract disclosures.
  • Chip supply metrics and pricing signals.
  • Regional power and data-center capacity availability.
• Prepare for volatility driven by technological, regulatory, and political risks (including founder behavior).

For individuals and professionals

• Learn fundamentals of coding and “computer-language thinking” to better specify and evaluate AI outputs (the speaker calls this “software vision”).
• Use AI to augment productivity while maintaining domain expertise—domain specialists who can direct AI remain valuable.
• For content and information gathering: watch and summarize foreign-language tech videos (English/German/Japanese) to inform investment and technical views.
• Be aware of algorithmic addiction risks (short-form video platforms); expect increasing regulatory attention to youth exposure.

Content and AI quality expectations

• Expect an initial flood of low-quality AI-generated content (“AI slop”), followed by market selection favoring higher-quality outputs and ongoing roles for human experts.
• Democratization of content and software creation is inevitable; professionals combining domain skills with software/AI fluency will stand out.

---

Notable products, technologies, and names mentioned

• Tesla: Cybercab (robotaxi), Optimus (humanoid), Megapack (energy)
• Nvidia: GPUs, CUDA, Blackwell, Vera Rubin (next-gen families referenced)
• Google/Alphabet: Gemini Robotics (DeepMind), TPU
• Microsoft, Amazon, Meta: cloud/data center investments and in-house chips
• TSMC, Samsung, SK Hynix: chip manufacturing and memory (HBM4 referenced)
• SpaceX: Starship / space data center concept
• Robotics firms: Boston Dynamics / Atlas; Hyundai, BMW (factory robotics trials)
• Blue AI: Dr. Kang’s AI startup (co‑founder role)

---

Actionable 2026 event checklist (what to watch)

• Legislation milestones for driverless vehicle commercialization (House and Senate votes; state regulations).
• Tesla public updates:
  • Cybercab production numbers.
  • FSD revenue and take-rate metrics.
  • Optimus 3rd-gen release and any evidence of factory-based production.
• Nvidia and big-tech announcements regarding contracts, chip roadmaps, and supply/pricing signals.
• Data center buildouts and power-capacity news (new generation projects, nuclear restarts, localized power solutions).
• Reports of large-scale robotics/data-collection centers, especially in China.

---

Main speakers / sources

• Interviewer: Kim Ya-hyun
• Interviewee / primary analyst: Dr. Kang Jeong‑su (co‑founder of Blue AI; PhD; technology–market analyst)