dingdongbell

$AMD(AMD.US)

Post 5 Ep. 24

Context: What caught my attention recently was AMD’s announced $8.2 billion acquisition of World Labs, the spatial AI company led by Fei-Fei Li. This is an interesting strategic shift. Rather than focusing solely on developing faster CPUs and GPUs, AMD is bringing AI model research expertise directly into its ecosystem, particularly in robotics, simulation and physical AI.

My trade: I increasingly see AMD as a company trying to influence how future AI workloads are built, rather than simply supplying the hardware to run them. Understanding how models evolve could help AMD design more efficient compute architectures and improve the relationship between its silicon, software and AI applications. The opportunity is compelling, although integrating such an ambitious acquisition brings execution and dilution risks.

Takeaway: The next competitive advantage in AI may come from understanding the workload before designing the hardware. AMD is making a bet that the future of computing belongs not just to those who build the fastest chips, but to those who understand what tomorrow’s intelligence will actually need.

@Captain's Treasure

$GLD SG(GSD.SG)

Post 4 Ep. 24

Context: What interests me about gold today is the changing relationship between government bonds and safe-haven demand. Traditionally, rising bond yields have been negative for gold. But when those yields rise because investors are increasingly concerned about government debt, fiscal sustainability and monetary credibility, the picture becomes much more complicated.

My trade: I continue to view my GLD SG position as exposure to that broader monetary uncertainty. The recent pullback despite strong global gold ETF inflows reminds me that short-term price action and longer-term investment demand don’t always move together. Rather than chasing every rally or reacting to every correction, I prefer maintaining a measured allocation while watching how these macro forces develop.

Takeaway: Gold’s appeal may increasingly extend beyond inflation protection or falling interest rates. When confidence in traditional safe-haven assets begins to weaken, an asset with no issuer and no sovereign credit risk becomes particularly interesting. Sometimes what gold doesn’t depend on is precisely what makes it valuable.

@Captain's Treasure

$Apple(AAPL.US)

Post 3 Episode 24

Context: What has become more interesting to me about Apple is how much of its AI strategy is being built into its own silicon. The new A20 Pro in the iPhone 18 Pro uses a 2nm process and delivers roughly double the AI processing capability of A19 Pro, while Apple is pushing the same direction with M6 across the Mac. Siri AI has now actually begun rolling out rather than remaining a future promise.

My trade: I therefore see Apple’s AI opportunity differently from the hyperscalers. Apple does not necessarily need to win the race for the largest foundation model. Its advantage may be controlling the chip, operating system, device and user experience together. If increasingly useful AI requires newer Apple silicon, intelligence itself could become another reason for users to upgrade their devices. That comes against an already strong backdrop: Apple’s latest quarter delivered a record $109.4 billion of revenue, up 16% YoY, while its active installed base reached another all-time high.

Takeaway: Apple may have arrived later to the headline AI race, but it is playing a different game. If AI becomes something people use continuously on their devices, owning the silicon, software and hardware together could turn intelligence into the next upgrade cycle — rather than simply another feature.

@Captain's Treasure

$NVIDIA(NVDA.US)

Post 2 Episode 24

Context: What I find increasingly important about NVIDIA is that the AI conversation is shifting from “Who can train the biggest model?” toward “Who can run intelligence most economically at scale?” NVIDIA’s latest Vera Rubin MLPerf results showed up to 3.7× higher inference throughput than GB300 NVL72, while management says Rubin can deliver roughly 30× higher throughput per megawatt and 35× lower token cost versus Grace Blackwell Ultra for relevant workloads.

My trade: That makes my NVIDIA thesis increasingly about AI economics rather than benchmark supremacy. As inference and agentic workloads become a larger share of AI spending, customers will care about tokens generated per dollar, per watt and per rack. If NVIDIA can keep lowering those costs through co-design across GPUs, CPUs, networking and software, each generation gives customers an economic reason to upgrade — not simply a technological one.

Takeaway: The long-term winner in AI may not simply be the company with the fastest chip. It may be the one that makes intelligence cheapest to produce at massive scale. NVIDIA increasingly looks like it is designing around exactly that objective. However, Nvidia is just a small percentage of my portfolio.

@Captain's Treasure

$Broadcom(AVGO.US)

Post 1 Episode 24

Context: What stands out to me about Broadcom today is not simply how fast its AI business is growing, but the quality of that growth. Q3 AI semiconductor revenue reached $16.7 billion, up 221% YoY, and management expects it to accelerate to $21.7 billion in Q4. At the same time, Broadcom generated $13.7 billion of free cash flow in a single quarter — equivalent to 46% of revenue.

My trade: That makes me view Broadcom differently from many AI beneficiaries. It is not merely spending heavily today in the hope of monetising AI years from now. Custom accelerators and networking are already scaling rapidly while the broader business continues to throw off substantial cash. Broadcom now expects total Q4 revenue of about $34.8 billion, up 93% YoY.

Takeaway: In a market full of AI promises, I increasingly value companies where the theme is already visible in the cash register. Broadcom isn’t just participating in the AI capex boom — it is converting that demand into enormous free cash flow. That makes the story much more than simply another AI trade.

@Captain's Treasure