This is Hamamoto from TIMEWELL.
NVIDIA GTC: Blackwell, Rubin, and 30x Inference
NVIDIA's GTC conference delivered the announcements the industry had been waiting for. Jensen Huang took the stage with the full Blackwell GPU architecture roadmap and the first detailed look at the next-generation Rubin platform.
The headline number: 30x improvement in AI inference performance relative to the previous generation. This is not a benchmark figure from a narrow test — it represents architectural improvements across memory bandwidth, interconnect speed, and compute density.
Key announcements:
- Blackwell Ultra: the next step in the current Blackwell generation, with improved HBM capacity and NVLink bandwidth
- Rubin architecture: the successor to Blackwell, designed with a fundamentally different approach to memory hierarchy and chip-to-chip communication
- NVLink 6: the interconnect standard that makes large-scale GPU clusters feasible at the bandwidth requirements of frontier AI training
For enterprises running AI inference workloads, the 30x figure is the one that matters most. Inference costs — serving AI responses to end users — are the dominant operational expense for AI products. A 30x improvement in efficiency translates directly into cost reduction or capability expansion at the same cost.
Tesla: 50% Stock Decline and the Musk Problem
Tesla's stock has fallen roughly 50% from its peak — a remarkable decline for a company that was, not long ago, the most valuable automaker in history.
The causes are layered:
Demand softness: EV demand growth has slowed across all markets. Tesla has responded with aggressive price cuts, which have sustained volume but compressed margins significantly.
Elon Musk's attention: Musk's involvement with DOGE (the Department of Government Efficiency) and his broader political activities have generated significant controversy. Some Tesla customers have publicly stated they are avoiding Tesla purchases because of Musk's political positions. The brand damage, while difficult to quantify, appears real.
Competition: BYD and other Chinese EV manufacturers have become credible competitors in markets outside the US. In Europe and Southeast Asia, Tesla's market share has declined.
In response to the stock decline and internal concerns, Musk held an all-hands meeting with Tesla employees. He addressed the demand concerns directly and reaffirmed Tesla's long-term roadmap: autonomous vehicles, robotaxis, and the Optimus humanoid robot as growth drivers. Whether those future bets can offset near-term pressure remains the central question for investors.
Looking for AI training and consulting?
Learn about WARP training programs and consulting services in our materials.
Nasdaq 100: Five Weeks of Decline
The Nasdaq 100 has declined for five consecutive weeks — its longest losing streak in several years. The index, heavily weighted toward large-cap technology companies, has given back a significant portion of the gains it made during the AI enthusiasm peak of 2024.
The decline is broad-based, affecting both AI infrastructure companies (chips, data centers) and AI application companies (software, consumer AI products). The compression suggests that investors are reassessing the timeline and magnitude of AI-driven revenue growth.
Nancy Curtin's perspective: Nancy Curtin, a widely-followed investment strategist, has argued that the current decline looks more like a correction than the beginning of a structural bear market. Her reasoning: corporate earnings remain strong, the labor market has not deteriorated significantly, and the underlying AI investment cycle is intact. Rate policy uncertainty and political risk are creating volatility, but the fundamental case for technology equities has not changed.
Not everyone agrees. Some analysts argue that AI revenue growth is not materializing fast enough to justify current valuations — that we are seeing the early stages of a multiple compression that could persist for several quarters.
Micron: HBM Market Leadership
Micron's stock has held up better than most technology peers, reflecting its strategic position in the HBM (high-bandwidth memory) market. As AI chip demand remains strong, HBM demand — which is tied directly to GPU shipments — continues to grow.
Micron's HBM3E product is certified by NVIDIA and shipping in volume. The company has been gaining market share at the expense of Samsung, which has struggled with quality issues in its HBM production.
Boeing F-47: Defense Technology Wins
Against the backdrop of technology sector weakness, Boeing received a significant contract award: the F-47 next-generation fighter jet program. The program is expected to be one of the largest defense contracts of the decade.
Defense technology is increasingly viewed as a portfolio diversifier — less correlated with the consumer sentiment and AI monetization timelines that are driving technology sector volatility.
Apple AI Leadership Shake-Up
Apple has made changes to its AI leadership team, reflecting internal pressure around the pace of Apple Intelligence development. The company has faced criticism that its AI features — launched with significant fanfare alongside iOS 18 — have not kept pace with competitors.
The leadership changes signal that Apple is taking the competitive threat seriously. Whether new leadership can accelerate Apple's AI roadmap meaningfully remains to be seen.
Vast Space: Commercial Space Station
Vast Space — a startup backed by significant private capital — is advancing plans for a commercial space station. The company represents the broader trend of private capital moving into infrastructure that was previously the exclusive domain of government space agencies.
The commercial space station market is still nascent, but the addressable market — microgravity research, manufacturing, and eventually tourism — is large if the cost of access continues to decline.
