Ξl Boyd

Founder | Investor | Business & Tech Mentor | Speaker - Empowering entrepreneurs to acquire, grow, scale, and exit 7+ figure businesses… while coalescing prof…

[Ξl Boyd @MrElBoyd]

WWDC26 is officially wrapping up.

In this final Dub Dub Daily, Apple sits down with Holly Borla from the Swift team to close the week with a look at what’s new in Swift 6.4.

The focus this year is on removing friction so developers can write clearer, more expressive code with less mental overhead. Highlights include cleaner syntax for optionals and some/any types, the ability to call async methods inside defer blocks, and meaningful improvements to compiler diagnostics, including a new @Diagnose attribute that gives developers more granular control.

It’s the kind of update that doesn’t always make headlines, yet makes daily work more enjoyable and productive. Small, thoughtful refinements that compound over time.

As WWDC26 comes to a close, the throughline has been clear: Apple is continuing to invest in the tools and frameworks that let developers build more capable, more intelligent apps, while keeping the experience grounded and developer first.

Today also marks another milestone as SpaceX begins trading publicly on the back of the largest IPO in history. Two different tracks, one focused on the tools we build with here on Earth, and one pushing the physical frontier outward, while both moving forward together at the same time. That convergence is worth watching.

If you spent any part of this week digging into the sessions or the daily recaps, what new capabilities are you looking forward to building with?

[Ξl Boyd @MrElBoyd]

Great point David. Open standards are key for shared AI backbones to scale without new lockins. There are advantages when they cut friction for builders while also supporting real coordination and choice.

There are plenty of examples of how standardized systems move faster & relatively cheaper than nonStandard systems.

With that said, we also have seen some great innovations in nonStandard breakthrough systems, that both stay that way & scale…as well as those that later become new standards. Luckily , there is room for both.

[Ξl Boyd @MrElBoyd]

Now imagine the future you’d like to live in together.

Not the bleak view the headlines keep insisting, the version that says everything is fracturing & division is inevitable. Instead, picture the one where real work is pulling people, companies, and even nations together around shared challenges & shared possibility.

Where the pressure we’re all feeling is being met with coordination instead of just competition, and where the future is being built by people who are choosing to coalesce rather than divide.
That world isn’t some distant fantasy. It’s forming right now, in ways that cut through the noise if you’re willing to look.

Right now, construction spending on the infrastructure of intelligence is running at record levels. Power demand from data centers is on track to more than double inside two years, and instead of just reacting to the strain, communities and operators are treating it as a design brief.

Capital is rotating toward solutions that can move at the speed of both technology and human reality.
At the same time, we’re seeing coordination at scale. China just put forward a $295Billion, five year plan to build a connected national AI infrastructure backbone.

On the other side, U.S. hyperscalers continue their own aggressive buildout. While competition is real, so is the growing recognition that some problems are too large to solve in isolation.

WWDC26 this week showed another layer of this. Beyond the technical announcements around Siri AI & Apple Intelligence, there was a clear working together ethos. Apple opening up more of its intelligence frameworks to devs, supporting third party models alongside its own, and creating tools that let builders integrate intelligence into their apps more deeply. It wasn’t just about one company advancing. It was about raising the floor for what thousands of teams can now build.

And then there’s space.
Space has always had a unique way of bringing people together. Artemis III, NASA’s upcoming crewed mission to the Moon, is a clear example, with a multinational effort involving multiple nations & companies working toward a shared goal.

At the same time, SpaceX($SPCX) is pricing and preparing to go public today, giving everyday investors a direct way to own a piece of the infrastructure that will help take humanity further. The same teams advancing Starship & orbital capabilities are also building the intelligence & power architectures that will support long term presence beyond Earth.

Not just exploration for its own sake, infrastructure that can eventually relieve pressure here on the ground.
This is what convergence looks like when you zoom out.

Every constraint we’re feeling right now, power, materials, siting, community alignment, even the limits of Earthbound systems, is also an invitation to design the next layer differently & better.

The operators & minds who treat these pressures as design briefs instead of obstacles are already sketching systems that are more resilient, more distributed, and ultimately more abundant.
Imagine what becomes possible when power stops being the bottleneck & becomes the enabler. When critical materials move through more resilient, onshored, and circular systems. When capital and expertise move with both speed and wisdom. When orbital and multiplanetary infrastructure stops being a distant dream & starts actively expanding what’s possible here.

When communities are coCreators, not obstacles. When space stops being something only governments or billionaires touch and becomes something any person can participate in & benefit from.

That is the future that excites. That is the one worth building.

Not just to look back at what happened, to stand at the edge of what is forming and ask: What will we choose to imagine, fund, build, & coalesce next on this #FutureFinanceFriday.

The future is closer than you think. Let’s keep moving us forward together, For All Humanity.

“I’m going to show them a world where anything is possible”

[Ξl Boyd @MrElBoyd]

Day 4 of WWDC26 is all about making developers faster and more productive.

In this Dub Dub Daily, Apple sits down with Ken Orr (Director of Xcode) to break down the major upgrades to the developer toolchain:

• Deep Apple Intelligence integration in Xcode, including coding agents that can write specs, generate code, create tests, and handle documentation

• Xcode 27 is now 30% smaller with significantly faster project loading & improved iCloud syncing

• New full color themes for a more customizable and vibrant interface

• The new Device Hub that brings simulators and physical devices together in one clean, modern view

• Smarter workflows like block insertion points and flexible destination choosers
This is where the intelligence layer meets the actual tools developers use every day, turning high level ideas into working code more efficiently while keeping the process under your control.

If you spend any time in Xcode, these updates are going to change how you work.

For All Humanity

[Ξl Boyd @MrElBoyd]

Day 3 of #WWDC26 is where the real developer power moves show up.

In this Dub Dub Daily, Apple sits down with the team behind the Foundation Models framework to break down what actually changed for builders.

• Direct access to on device models + Private Cloud Compute for server-side intelligence

• Support for third party models (Gemini, Claude, Grok, and more) alongside Apple’s own

• Dynamic Profiles so apps can intelligently route simple tasks on-device and complex ones to the cloud

• App Intents that let Siri deeply understand and act inside your app with natural language

• Vision framework integration (OCR, barcodes, etc.) now callable directly from the intelligence layer

This is the bridge between the big Siri AI announcements and actual app experiences that feel smarter, faster, and still private by design.

If you’re building anything that touches intelligence, context, or Siri this year, these updates open up some genuinely new possibilities.

What are you most excited to experiment with from these changes?

For All Humanity

[Ξl Boyd @MrElBoyd]

SpaceX IPO and Straight Answers to the Biggest Open Questions

With SpaceX moving swiftly toward a public listing, here are clear answers to some of the most common questions, based on what’s actually known right now.

Q: Can orbital AI data centers actually work?

What we can say confidently:
No one has done this before. SpaceX itself states in its disclosures that neither the company nor anyone else has previously operated orbital AI compute infrastructure. It also notes that hardware deployed in orbit cannot be physically repaired or upgraded.

What will likely happen:
This will be significantly harder than it sounds on paper. Cooling high density AI hardware, dealing with radiation effects over time, and building reliable networking between orbital nodes are all unsolved problems at scale. It is more likely than not that meaningful orbital AI compute remains a few years away.

Q: How ready is Starship to support this vision?

What we can say confidently:
Starship is still in the flight testing phase. It has not yet demonstrated the high launch cadence, reliability, or cost per kilogram needed to make large scale orbital infrastructure economically viable.

What will likely happen:
Starship will eventually reach high operational cadence, as the vehicle is clearly capable in principle. However, the timeline remains uncertain. Any meaningful delay in reaching reliable, high frequency flights would directly slow down the orbital compute plans.

Q: Are there real customers for orbital compute beyond xAI?

What we can say confidently:
Yes. Anthropic has publicly stated interest in partnering with SpaceX on orbital AI compute capacity (in addition to its large terrestrial deal). Google is also in active discussions with SpaceX regarding orbital data centers and related launch support.

What will likely happen:
There will be some external demand, but xAI will likely remain the primary anchor customer for a long time. Other AI labs may experiment with orbital capacity, but most will probably continue prioritizing terrestrial infrastructure in the near term due to cost, latency, and risk considerations.

Q: What’s the relationship between SpaceX and xAI now?

What we can say confidently:
SpaceX acquired xAI in an all stock deal earlier this year. xAI is now a wholly owned subsidiary of SpaceX. This structure significantly reduces traditional governance conflicts that would exist between two separate companies.

What will likely happen:
Alignment between SpaceX’s infrastructure and xAI’s compute needs will be structurally cleaner than before. Public shareholders will still want transparency on how capital is allocated between core SpaceX operations and the AI/subsidiary side, however, the previous concerns about two separate entities are largely resolved.

Overall Outlook
Space is becoming the next frontier in a very literal sense. As AI infrastructure demand continues to grow, new questions around energy, compute location, and physical limits are forcing genuinely new approaches. Some of these ideas will prove harder than expected. Others may create real, lasting advantages.

The companies willing to ask the hardest questions, and actually try to solve them, are the ones that will define what becomes possible next. SpaceX is clearly placing itself in that category.

For All Humanity

[Ξl Boyd @MrElBoyd]

Apple just took a meaningful step toward embedding intelligence more deeply across its entire ecosystem while staying true to its core principles.

At today’s WWDC, Apple unveiled a rebuilt Siri that delivers on the long promised context aware experience.

The new Siri is currently partially powered by Google Gemini under the hood for its more advanced reasoning and personalization capabilities, while Apple’s own on device models continue to handle faster, privacy sensitive tasks across writing tools, image generation, and simpler interactions.

These are woven together through Apple’s orchestration layer and Private Cloud Compute infrastructure.

On visionOS, Siri gains real world visual awareness, recognizing surroundings and pulling relevant information from the physical environment. Across the broader ecosystem, it can now understand personal context, maintain conversation history, and act more intelligently across core apps, Shortcuts, and HomeKit.

What makes this particularly the kind of move we expect from Apple, is how they’ve architected the privacy layer. Sensitive, personal data stays on device wherever possible.

When additional compute or deeper reasoning is needed, Apple routes it through a privacy preserving cloud infrastructure they designed to minimize data exposure.

This hybrid approach of Apple’s own models for on device speed and control, Gemini for complex intelligence, enables a deeper, more integrated experience without forcing users to trade privacy for capability.

It’s a clear strategic bet that the best long term path is building intelligence that respects boundaries rather than extracting maximum data.

On the developer side, Apple also advanced its Core AI tooling in Xcode and opened the door to selectable LLMs, giving developers more flexibility in how they integrate intelligence into their apps & app building.

This was also Tim Cook’s final #WWDC keynote as CEO before he steps down in September and into the Executive Chairman of Apple’s Board of Directors seat. John Ternus (currently Senior Vice President of Hardware Engineering) will become Apple’s next CEO on the same date.

The broader picture is worth keeping an eye on, as Apple ($AAPL / $AAPLx ) is working to define its own lane in AI. One where deep integration, privacy, and a mix of on device and cloud intelligence are not competing priorities, rather, part of a single coherent architecture.

For All Humanity

[Ξl Boyd @MrElBoyd]

Tomorrow at Apple’s WWDC 2026, all eyes will be on what they reveal about the future of Apple Intelligence, and more specifically, how deeply they intend to integrate AI into the core Apple experience.

The biggest question isn’t just which model they choose (Gemini appears to be the frontrunner). It’s whether Apple will lock users into a single model or take a more open approach.

There are two clear paths:

Path 1 → Lock into Gemini

Apple could simply integrate Google’s Gemini as the primary intelligence layer across Siri, writing tools, image generation, and system level features.

This would be the simpler, more controlled route. It gives Apple a powerful model quickly without having to build one at frontier level themselves.

Pros: Faster time to market, simpler user experience, strong performance out of the gate.

Cons: Reduces user choice, creates long term dependency on Google, and weakens Apple’s positioning as a privacy and control focused company. It also hands Google significant influence inside Apple’s ecosystem.

Path 2 → Build the Control Layer, Not the Model

Apple could instead focus on building a strong privacy, security, and orchestration layer within Apple Intelligence. Then allow users to connect to the model of their choice (Grok, Gemini, Claude, or future options).

This would mean Apple handles the onDevice processing, data protection, and system integration, while the heavy intelligence work happens through user selected models.

Pros: Gives users real choice, strengthens Apple’s privacy brand, reduces dependency on any single company, and future proofs the system as better models emerge.

Cons: More technically complex to execute well, potentially slower rollout, and requires Apple to maintain strong relationships (and technical integrations) with multiple AI providers.

From a firstPrinciples standpoint, Path 2 aligns much more closely with Apple’s historical strengths and stated values around privacy and user control. It would also be the smarter long term strategic move in a world where model dominance is still in flux and likely to change over time.

That said, there is the matter of execution risk. Building a clean, secure, & seamless multiModel experience is significantly harder than plugging into one strong model.

We’ll all be watching closely tomorrow, June 8, 2026, at 10:00 AM PT/1:00 PM ET to see which direction Apple actually chooses.

If you are in the  ecosystem, which path are you hoping for on this #ScienceSunday?

[Ξl Boyd @MrElBoyd]

Tomorrow at Apple’s WWDC 2026, all eyes will be on what they reveal about the future of Apple Intelligence, and more specifically, how deeply they intend to integrate AI into the core Apple experience.

The biggest question isn’t just which model they choose (Gemini appears to be the frontrunner). It’s whether Apple will lock users into a single model or take a more open approach.

There are two clear paths:

Path 1 → Lock into Gemini

Apple could simply integrate Google’s Gemini as the primary intelligence layer across Siri, writing tools, image generation, and system level features.

This would be the simpler, more controlled route. It gives Apple a powerful model quickly without having to build one at frontier level themselves.

Pros: Faster time to market, simpler user experience, strong performance out of the gate.

Cons: Reduces user choice, creates long term dependency on Google, and weakens Apple’s positioning as a privacy and control focused company. It also hands Google significant influence inside Apple’s ecosystem.

Path 2 → Build the Control Layer, Not the Model

Apple could instead focus on building a strong privacy, security, and orchestration layer within Apple Intelligence. Then allow users to connect to the model of their choice (Grok, Gemini, Claude, or future options).

This would mean Apple handles the onDevice processing, data protection, and system integration, while the heavy intelligence work happens through user selected models.

Pros: Gives users real choice, strengthens Apple’s privacy brand, reduces dependency on any single company, and future proofs the system as better models emerge.

Cons: More technically complex to execute well, potentially slower rollout, and requires Apple to maintain strong relationships (and technical integrations) with multiple AI providers.

From a firstPrinciples standpoint, Path 2 aligns much more closely with Apple’s historical strengths and stated values around privacy and user control. It would also be the smarter long term strategic move in a world where model dominance is still in flux and likely to change over time.

That said, there is the matter of execution risk. Building a clean, secure, & seamless multiModel experience is significantly harder than plugging into one strong model.

We’ll all be watching closely tomorrow, June 8, 2026, at 10:00 AM PT/1:00 PM ET to see which direction Apple actually chooses.

If you are in the  ecosystem, which path are you hoping for on this #ScienceSunday?

[Ξl Boyd @MrElBoyd]

While the video lays out Grok’s core directive, Let’s dive into why that core actually matters & why the moves happening now position Grok to become the strongest overall AI in the years ahead.

Grok was built with a different foundation than the others. From the start, the prime directive has been to accelerate humanity’s understanding of the universe through maximum truth seeking.

That’s not just marketing, it’s the actual north star. Most other models were optimized primarily for user engagement, safety theater, or corporate alignment. Grok was optimized to pursue reality as accurately as possible, even when it’s inconvenient.

That difference in core is compounding.
Right now we’re seeing multiple high leverage pieces coming together:

• Massive realWorld data from 𝕏

• Physical world interaction and robotics through Tesla and Optimus

• Advanced manufacturing and energy infrastructure thinking (TeraFab and related efforts)

• Tight integration with top-tier coding environments (Cursor, Build, etc.)

• And xAI’s continued focus on scaling compute and models without diluting the original mission

Because @Grok started later than the other frontier models, it had to move fast just to reach parity. What many people don’t realize is that this “catch up” phase is now transitioning into a compounding advantage phase.

The combination of an uncorrupted truthSeeking core + realWorld data + physical embodiment + superior tooling is extremely powerful.

Over the next three years, these elements are likely to create dramatic leaps across reasoning, realWorld understanding, scientific discovery, & practical capability. Not because of random breakthroughs, because of a clear, firstPrinciples plan that was set in motion from the beginning.

This isn’t about crowning any model “the best” for its own sake. It’s about having an AI whose fundamental orientation is toward truth and human progress, one that can help us understand the universe more deeply and solve the hard problems we actually face.

That core matters more than most people currently appreciate.

For All Humanity.

[Ξl Boyd @MrElBoyd]

Stop and imagine for a moment the world you actually want to live in.

Not the one the headlines keep selling, imagine the one that makes you happily exhausted at the end of the night and wakes you up enthusiastic for what the day holds.

The one where Eureka moments arrive almost daily. Where we Star Trek in and out of new orbits, solve the Foundation mysteries of our time, or simply step into a Halliday style world for a race where people from everywhere show up, get along, enjoy the moment, and somehow end up coalescing to move humanity forward.

That world is closer than the noise suggests.

Right now, construction spending on the infrastructure of intelligence is running at record levels, outpacing categories that once defined “big”.

Power demand from data centers is on track to more than double inside two years. Communities are raising their hands and saying, “If this is the future, let’s make sure it actually works for the people who live here.”
So capital is rotating toward solutions that can move at the speed of both technology and human reality.

Deep tech breakthroughs in light based computing and room temperature systems are removing walls that once looked permanent. And the same teams building the ships that will land on other worlds are also building the intelligence and power architectures that will guide them.

This is what convergence looks like at scale.
Instead of the future arriving as a single wave, it is arriving as pressure meeting imagination.

Every constraint we are feeling right now, from power, materials, & siting, to community alignment, is also a design brief for the next layer of capability.

The operators and minds who treat these pressures as invitations rather than obstacles are already sketching the outlines of the world we just imagined.

Imagine what becomes possible when power is no longer the bottleneck, and becomes the enabler. When critical materials flow through resilient, onShored, & circular systems instead of fragile chokepoints. When capital and expertise move with both speed & wisdom. When orbital and multiplanetary infrastructure stops being a distant dream and starts relieving pressure on the ground. When communities are not obstacles but coCreators.

That is the future that excites. That is the one worth building toward with everything we have.
We are doing more than just watching it form. We are inside the moment where positive imagination can still win.

Every child is born with that fire.
Many have had it buried under noise & negativity. This is the work of reigniting it, and not with slogans, instead with living proof that the convergence is real, the leap is possible, and the future we actually want to live in is being built by people who refuse to settle for the smaller story.

So here we are on another #FutureFinanceFriday.
Not just to recite what already happened, to stand at the edge of what is forming and ask…What will we choose to imagine, fund, build, and coalesce next?

The ring is already spinning. The connections are already lighting up.
The only question left is how brightly we decide to shine.

Welcome to the future that is closer than you think. Let’s keep moving us forward.

For All Humanity.

[Ξl Boyd @MrElBoyd]

Some hyperscalers are making one of the most important infrastructure bets of the AI era, by locking in nuclear power for the massive, always on electricity their data centers need.

Existing large nuclear plants already deliver what the buildout demands, that being energy density and uptime that renewables alone struggle to match at this scale.

They run at ~92% - 93% capacity factor, providing firm baseload power that doesn’t depend on weather or get disrupted by supply chain shocks.

At the same time, a new generation of Small Modular Reactors (SMRs) is moving from concept to deployment. These smaller, factory built units offer similar reliability advantages with greater siting flexibility and faster scaling potential.
Multiple hyperscalers have already signed deals for both restarted large reactors and new SMR capacity.

This shift makes sense. AI workloads need consistent, high density power. Nuclear delivers it with high uptime and a small physical footprint compared to the land and infrastructure required for equivalent renewable & storage setups.

The race for reliable electricity is now running in parallel with the race for compute. The operators who secure firm, scalable power earliest will have a structural advantage as demand continues to climb.

For All Humanity.

[Ξl Boyd @MrElBoyd]

A new study just strengthened one of the leading explanations for how the Grand Canyon formed.

Researchers from the U.S. Geological Survey and Arizona Geological Survey analyzed mineral grains (zircons) in ancient lake deposits east of the canyon and found strong evidence that the ancestral Colorado River drained into a large basin which is now called Lake Bidahochi around 6.6 million years ago.

Over time, the lake grew until it eventually it spilled over a natural barrier. That spillover event likely carved the initial path of what became the Grand Canyon and helped establish the modern course of the Colorado River all the way to the Gulf of California.

This helps resolve a long standing geological puzzle around the river having existed in western Colorado as far back as 11 million years ago. Yet, clear evidence of its full continental scale path only appears much later. The new data supports the idea that the river spent millions of years pooling in this upstream basin before breaking through.

It’s a reminder that some of Earth’s most dramatic landscapes were shaped not by sudden catastrophe, but by patient, iterative processes playing out over deep time, with water finding its way, landscapes slowly yielding, and systems reorganizing themselves into something new.

Something to keep in mind when you are formulating and building your MTP, time and persistence in a positive direction can move the largest of seemingly immovable obstacles. Turning “impossible” into “I’m Possible” again & again over time.

For All Humanity.

[Ξl Boyd @MrElBoyd]

Great question Carl. The short version is, the ring didn’t create the raw material. It collected, concentrated, & repeatedly processed the available dust and gas into solid planetesimals over an extended period…acting like a sustained, high efficiency assembly line. Hope that makes it more clear.

[Ξl Boyd @MrElBoyd]

Scientists just discovered a massive “planet factory” right beyond Jupiter that is reshaping how we understand the birth of worlds.

A new study from the Max Planck Institute for Solar System Research reveals a giant dust filled ring outside Jupiter that acted like a cosmic manufacturing plant.

This ring produced multiple generations of planetesimals (the building blocks of planets) with strikingly different compositions over time.

Instead of a single, uniform process, the ring kept churning out new raw material and forging it into solid bodies, essentially running a high volume, long duration factory for planets.

This is the kind of overlooked mechanism that turns scattered dust into entire solar systems. The discovery shows formation wasn’t a one and done event…it was iterative, resilient, & far more productive than models predicted.
The implications stretch far beyond our own solar system.

If these planet factories are common, the universe may be far richer in habitable worlds and raw materials than we thought. It also gives engineers and mission planners a clearer map for where to look for resources when we start building outposts on other moons and planets.

This is firstPrinciples abundance in action on a cosmic scale. A hidden ring of dust quietly turning chaos into ordered, life supporting worlds, one generation at a time.

The same pattern we see here on Earth with precision manufacturing breakthroughs is playing out across the solar system. Overlooked resources, turned into exponential enablement.

For All Humanity on this #ScienceSunday

[Ξl Boyd @MrElBoyd]

Monash University just delivered a genuine leap in photonic computing.

Researchers have built the first fully integrated oncChip valleytronic circuit using atomically thin 2D materials. In one compact device, it can generate, precisely steer, and read light based information by harnessing the “valley” degree of freedom, which is a quantum property that lets light carry information in new ways.

The breakthrough combines these few atom thick materials with engineered nanoscale structures (metasurfaces) through a clever stacking approach. The result is a complete photonic system that operates at room temperature and handles multiple information streams simultaneously.

This matters because light based processing sidesteps many of the heat and energy walls that currently constrain electronic AI chips. Photonic approaches offer massive bandwidth and ultra fast transmission with far lower power draw.

Room temperature operation also removes a major practical barrier that has slowed quantum adjacent photonic systems.
Early implications point to more energy efficient AI accelerators, new pathways for quantum information encoding, secure optical communications, and advanced imaging. Keep in mind this is all onChip scale hardware.

We’re watching the photonic hardware layer closely. Advances like this directly support the kind of efficient, scalable compute that makes exponential intelligence growth sustainable rather than energy constrained.

For All Humanity.

[Ξl Boyd @MrElBoyd]

Four disciplined M&A moves this week show capital and expertise aligning at the exact points where America needs it most.

Marex Group ($MRX) acquired Levmet, expanding its physical market making and European power/gas trading capabilities in base metals, ferrous metals, energy, and power.

Berkshire Hathaway agreed to acquire homebuilder Taylor Morrison ($TMHC) for $8.5Billion in cash. This is a direct bet on significantly more U.S. housing construction in the years ahead.

Motorola Solutions ($MSI) announced the acquisition of D-Fend Solutions (@DFendSolutions), the industry leader in counter-drone technology, for $1.5Billion.

Nextpower ($NXT) entered the battery energy storage and AI data center markets with its acquisition of Prevalon Energy for up to $365Million.

These deals represent focused operators acquiring specialized capabilities that strengthen core platforms and solve real bottlenecks like physical commodities liquidity, housing supply, critical infrastructure protection, & the power backbone for AI.

Together they point to the same larger pattern for the United States, accelerating buildout in energy, housing, security, and compute infrastructure.

Domestic capital is moving with precision toward assets that enhance supply chain resilience, expand housing inventory (which helps moderate long term price pressure), secure critical systems, & power the next wave of AI driven growth.

This is firstPrinciples capital allocation in action, turning today’s constraints into tomorrow’s expanded capacity, opportunity, & broader prosperity across multiple foundational industries.

For All Humanity.

[Ξl Boyd @MrElBoyd]

Wiwynn (@WiwynnCorp) Chair Emily Hong just delivered a clear warning that should catch the attention of every manufacturer, operator, and investor in the physical infrastructure space.

Demand for data center hardware will stay extremely strong for the next three to five years. Shortages have now moved beyond memory chips and are hitting other vital components. The bottleneck picture keeps rotating, and meaningful relief is not expected until late 2027 or 2028.

The components most affected include high end PCBs and substrates, power delivery hardware such as transformers, switchgear and busbars, networking optics, liquid cooling systems, and the passives needed for power integrity in high density racks.

This creates a structural opening for American precision manufacturers who can deliver speed and tight tolerance at volume.

Take G.S. Precision as a working example. Founded in 1958 by George Schneeberger, now led by his son Norm, the company has built proprietary high speed processes for complex metal and special alloy components. They already supply mission critical parts for high tech long range smart missiles and aerospace systems because they produce at speeds and precision levels traditional shops cannot match.

The same capabilities that power defense programs can now power the AI data center racks that will train the next generation of systems guiding Starship and beyond.

For manufacturers and connectors who can scale similar methods here in the United States, the opportunity is large. OnShore or nearShore production of these rotating bottleneck components would capture high margin, long term contracts from hyperscalers while strengthening domestic supply chain security.

This is the kind of firstPrinciples move that turns today’s constraints into tomorrow’s manufacturing renaissance and multiplanetary capability.

For All Humanity.

[Ξl Boyd @MrElBoyd]

https://t.co/deG9LtQuph