Categories
AI Business Investing Technology

The Scarcity Portfolio: Navigating Sovereign Debt, Wafer Bottlenecks, and Orbital Compute

Today I was watching the interview of Gavin Baker by Patrick Oโ€™Shaughnessy on his Invest Like the Best podcast. Like prior conversations this was another fascinating excursion into the mind of a sophisticated and very successful tech venture investor.

During the conversation, Patrick asked Gavin what agents he was using that were especially helpful and he mentioned one which summarizes YouTube podcasts and videos for him. Like most of us Baker just doesnโ€™t have the time to watch or listen to them himself so good summaries are really helpful.

Turns out Iโ€™ve been working on a Google Gemini Gem that does this for me. When Baker mentioned his I fired up the new Gemini 3.5 Flash model and asked it to summarize the Baker interview.

Later in the conversation Baker used the term โ€œbattlefield AIโ€ which caused me to go back to Gemini again to learn more about that. The results were so interesting that I asked Gemini to create a syllabus for a semester class on these subjects. After that I asked it to convert our whole conversation into a Markdown file so I could share it. Youโ€™ll find it below.

I found this whole experience pretty stunning. I came away very impressed with Gemini 3.5 Flash both for the quality of the responses but also the sheer speed. Wow!

Anyway I hope you enjoy the following!


Categories
Business Creativity Space SpaceX

Test like you fly!

Thereโ€™s a phrase in the SpaceX documentary that keeps coming back to me: โ€œTest like you fly.โ€ It sounds like a slogan. The kind of thing that gets painted on a factory wall and eventually stops meaning anything. But the more I sit with it, the more I think itโ€™s actually a philosophy that reaches well beyond rocket engineering.

The video โ€” a 25-minute documentary SpaceX released last week โ€” is ostensibly about Starship Version 3. New ship, new booster, new engines, new pad, new test site. Everything rebuilt. And theyโ€™re not shy about framing it as a reset, not an upgrade. One description I read called it โ€œa quiet violence in progress.โ€ That phrase stopped me cold, because itโ€™s exactly right. Progress that looks violent from the outside โ€” all that fire and metal โ€” but is somehow quiet in its inevitability.

What moved me watching it wasnโ€™t the engines. It was the engineers. SpaceX put the people on camera: the ones running cryogenic pressure tests at 80 Kelvin, stress-testing tank structures at 70% proof, explaining their failures and their data with the flat affect of people who have made peace with how long hard things take. Thereโ€™s something almost monastic about it. You choose a problem that will not yield easily. You accept that the work will outlast any individual sprint of enthusiasm. You go back to it anyway.

I keep thinking about that in the context of what weโ€™re doing with AI โ€” the other enormous, fast-moving project that I spend so much of my mental energy on. The development arc is different: iterative releases, weeks not years between jumps, demos that blur into deployment. But the same principle is buried in there somewhere. The best AI teams I read about arenโ€™t the ones shipping the most polished demos. Theyโ€™re the ones building infrastructure for failure โ€” evals, red-teaming, structured feedback loops. Test like you fly.

The Raptor 3 engines now produce 280 metric tons of thrust each. Thirty-three of them on a Super Heavy booster means over 17 million pounds of liftoff force. I have no intuitive frame for that number. What I do have a frame for is what those numbers represent: three years of iteration on top of five years before that, on top of a theoretical foundation laid by people who didnโ€™t live to see any of this. Thereโ€™s a compounding in that which I find genuinely moving. Nobody built the Raptor 3 in isolation. It came from everything that broke before it.

The hardest part of the documentary isnโ€™t the engineering. Itโ€™s the implicit acknowledgment of how much remains undone. No Starship has yet achieved full orbital velocity with both stages intact. In-space refueling is still untested. The thermal protection systems need more work. And yet โ€” SpaceX talks about unmanned cargo missions to Mars before the end of this year like itโ€™s on the roadmap, not the wish list. That sentence used to sound like marketing. Watching the footage, it doesnโ€™t anymore.

Iโ€™m not sure what to do with that feeling exactly. Itโ€™s something between awe and vertigo. Weโ€™re living in a moment when the audacious has started to have quarterly milestones. When the impossible keeps showing up on timelines and then โ€” bewilderingly, uncomfortably โ€” meeting them.

Test like you fly. Fail with rigor. Build the thing you actually need, not the thing you could more easily explain.

I keep turning that over. Thereโ€™s a post in there somewhere about writing, too โ€” about the drafts nobody sees, the structural tests that fail, the versions that taught you the one that worked. But thatโ€™s for another day.

For now Iโ€™m just sitting with the footage of those 33 engines lighting up, and the quiet weight of how much went wrong before they could do that.

Categories
AI Business

The Gravity of Compute

We are currently witnessing the single largest deployment of capital in human history. The “Hyperscalers”โ€”the titans of our digital ageโ€”are pouring hundreds of billions of dollars into the ground, turning cash into concrete, copper, and silicon.

The prevailing narrative is one of unceasing, exponential growth: bigger models require bigger clusters, which require more power plants, which require more land. It relies on the assumption that the demand for centralized intelligence is insatiable and that the current architecture is the only way to feed it.

But history suggests that technology rarely moves in a straight line; it swings like a pendulum. Two forces are emerging from the periphery that could impact the ROI of this massive infrastructure build-out. One is hiding in your pocket, and the other is waiting in the sky.

A recent conversation with Gavin Baker outlines a potential “bear case” for datacenter compute demand: the rise of Edge AI.

We often assume we need the “God models”โ€”the omniscient, trillion-parameter giants hosted in the cloudโ€”for every interaction. But do we?

Baker suggests that within three years, our phones will possess the DRAM and battery density to run pruned versions of advanced models (like a Gemini 5 or Grok 4) locally. He paints a picture of a device capable of delivering 30 to 60 tokens per second at an “IQ of 115.”

“If that happens, if like 30 to 60 tokens atโ€ฆ a 115 IQ is good enough. I think that’s a bear case.” โ€” Gavin Baker

Consider the implications of that specific number. An IQ of 115 isn’t omniscient, but it is competent. It is capable, nuanced, and helpful.

If Appleโ€™s strategy succeedsโ€”making the phone the primary distributor of privacy-safe, free, local intelligenceโ€”the vast majority of our daily queries will never leave the device. We will only reach for the cloudโ€™s “God models” when we are truly stumped, much like we might consult a specialist only after our general practitioner has reached their limit. If 80% of inference happens on the edge for free, the economic model of the trillion-dollar data center begins to look fragile.

Then there is the second threat, one that attacks the terrestrial constraints of the data center itself: the Orbital Data Center. Elon Musk and SpaceX – along with Google’s Project Suncatcher – envision a future where the heavy lifting isn’t done on land, but in orbit. Space offers two things that are scarce and expensive on Earth: unlimited solar energy and an infinite heat sink for radiative cooling. If Starship can reliably loft “server racks” into orbit, the terrestrial moat of land and power grid accessโ€”currently the Hyperscalers’ greatest defensive assetโ€”evaporates.

We are left with a fascinating juxtaposition. On one hand, we have the “Edge,” pulling intelligence down from the clouds and putting it into our hands, making it personal, private, and free. On the other, we have “Orbit,” threatening to lift the remaining heavy compute off the planet entirely to bypass the energy bottleneck.

There are hundreds of billions of dollars betting on a future of heavy, centralized gravity. But if the edge gets smart enough, and the orbit gets cheap enough, the gravity may have shifted.

Categories
AI Business SpaceX

Overcoming Limiting Factors: Orbital Data Centers & The Optimus Era

One of my favorite persons to follow on X is @pbeisel (Phil Beisel). Heโ€™s quite active sharing his thoughts about many of the same topics Iโ€™m interested in: technology, AI, robotics, computing, etc. Phil’s written a series of great articles about Tesla Full Self Driving, Optimus, etc. that are well worth spending time with.

On Saturdays, he get together on YouTube with Randy Kirk and they talk about whatโ€™s interesting from the last week – often thatโ€™s got something to do with various aspects of the โ€œMusk-conomyโ€ – the various companies of Elon Musk.

This weekโ€™s edition reviews Philโ€™s distillation of the Cheeky Pint interview with Elon published earlier this week. As usual, Philโ€™s comments add additional insights into the topic.

When I begin viewing a long YouTube video, I also like an accompanying summary that I can follow along. YouTube now has the ability to generate these summaries but Iโ€™ve got a custom Gem prompt that I prefer to use instead which tailors the results a bit more to my liking.

Below, for example, is the summary of this weekโ€™s conversation between Phil and Randy that was generated by Gemini Pro 3:

Executive Summary: The Musk “Musconomy” Convergence

The central thesis of the discussion is that Elon Musk is moving toward a total vertical integration of his companies (Tesla, SpaceX, and xAI) to overcome terrestrial “limiting factors” and dominate both the physical and digital manifestation of AI.


1. The “Limiting Factor” Philosophy [11:20]

  • Problem-Solving Framework: Musk focuses personal time and resources strictly on the “limiting factor” of any given projectโ€”currently identified as compute power and energy.
  • Vertical Integration: To bypass supply chain bottlenecks (e.g., turbine blades for power plants), Musk is opting to manufacture raw materials and components in-house rather than relying on external catalogs [18:18].

2. Orbital Data Centers: The Space “Escape Hatch” [24:19]

  • Energy Constraints: Terrestrial data centers are hitting a wall due to slow public utilities and permitting [15:26].
  • The Vision: Moving inference-based data centers to orbit using a constellation of satellites connected by optical laser links.
  • Economic Viability: Musk projects economic viability for space-based data centers within 30โ€“36 months, with reusability of the Starship being the primary hurdle [25:03].
  • Strategic Advantage: Unlike Google or Meta, Musk owns the “kilogram-to-space” delivery mechanism, potentially forcing competitors to rent capacity from SpaceX [32:19].

3. Optimus and the “Abundance Engine” [39:00]

  • Physical Dexterity: Musk is prioritizing high-dexterity actuators designed in-house to achieve human-level utility [40:30].
  • Training Scale: Tesla is moving toward training Optimus in “gymnasiums” using 10,000โ€“30,000 bots working 24/7 to develop “composable” skills (basic movements) and “decomposable” skills (complex tasks) [55:13].
  • Impact: Optimus is viewed as a paradigm-shifting product that will redefine global GDP by decoupling labor from human constraints [54:56].

4. xAI: The Digital Control Plane [56:19]

  • The “Brain” Portability: xAI is viewed as the “orchestration AI” for the entire fleet of Muskโ€™s physical assets (Starships, Teslas, and Optimus) [59:01].
  • Unified Interface: The vision includes a seamless “digital personality” or movable brain that follows the user from their phone to their car to their home robot [01:00:15].

Key Projections & Timelines

Objective Target/Detail Timestamp SpaceX IPO Likely to happen before a Tesla merger to attract cheap capital [03:31] Solar Scaling Aiming for a 300x increase (100 gigawatts/year) [22:21] Starship Reusability remains the “unlock” for space-based AI economics [25:51]

Conclusion: The “Musconomy” is transitioning from separate ventures into a singular entity where SpaceX provides infrastructure, Tesla provides the physical bodies, and xAI provides the intelligence.