Wednesday, June 10, 2026

C3Ai is a completely unloved stock, but, Tom Seibel is back! Turnaround story or, Value Trap!

C3.ai (NYSE: AI) – Business / Investment Report

Potential Turnaround Story or Value Trap?

Focus: The “Tom Siebel Effect”

Date: June 2026

1. Executive Summary

C3.ai represents one of the more controversial “fallen angel” AI stocks in the market today.

Once viewed as a premier enterprise AI platform and briefly trading above $170 after its IPO enthusiasm, the stock has collapsed due to execution failures, slowing growth, leadership instability, and investor skepticism. However, the return of founder Tom Siebel as CEO in May 2026 has materially changed the investment narrative. The question is no longer whether C3.ai is broken — it clearly was — but whether this is now a legitimate founder-led turnaround opportunity.

Investment conclusion:
C3.ai is not yet a confirmed turnaround, but it is now a

credible asymmetric turnaround candidate.

For a high-risk retail investor seeking AI exposure beyond obvious mega-caps, C3.ai may represent a classic “maximum pessimism” entry point, provided investors accept elevated volatility and execution risk.

2. The “Tom Siebel Effect” — Why This Matters

The central turnaround thesis revolves around one man:

Thomas Siebel

Siebel returned as CEO in May 2026 after stepping back due to serious health issues that materially disrupted sales execution and strategic oversight. Management itself acknowledged that performance deterioration accelerated while Siebel was less involved in day-to-day operations.

This matters because C3.ai is not a commodity SaaS company.

It is an enterprise AI sales organization, where:

relationships matter,
long sales cycles dominate,
government and Fortune 500 trust is essential,
executive selling often determines success.

Historically, Siebel has been one of Silicon Valley’s strongest enterprise sales operators, having previously built and sold Siebel Systems to Oracle for approximately $5.8 billion.

Why founder returns sometimes work

Turnaround history shows founder returns can be highly effective when:

✅ the founder remains deeply connected to customers
✅ execution problems (not product failure) caused deterioration
✅ balance sheet strength buys time
✅ organizational bloat gets reset

C3.ai arguably checks all four boxes.

The risk, however, is whether the business deterioration has gone too far.

3. Financials — Broken Business or Temporary Breakdown?

This is where the story becomes complicated.

Fiscal 2026 was ugly.

Quarterly revenue fell sharply to roughly $51.6 million, and bookings disappointed investors. Revenue contraction raised serious concerns about whether C3.ai had simply lost relevance in enterprise AI.

However, several important positives remain:

Strengths

1. Strong cash position

C3.ai still holds approximately $250M+ in annual revenue and substantial liquidity with minimal debt, meaning bankruptcy or forced dilution risk appears limited near term. This gives management time to execute a turnaround.

2. Aggressive restructuring already underway

Management implemented major workforce reductions and restructuring expected to deliver approximately $135 million in annualized cost savings.

This matters because many successful software turnarounds first go through a painful “reset” phase before operating leverage improves.

3. Guidance stabilizing

Despite weak recent performance, management guidance modestly exceeded Wall Street expectations for fiscal 2027, suggesting deterioration may be slowing.

Weaknesses

The biggest problem remains obvious:

Revenue is still shrinking.

Until growth stabilizes and reaccelerates, investors will remain skeptical.

For C3.ai, the key metric is not profitability yet.

It is:

Can they return to sustainable enterprise revenue growth?

4. Business Environment — Better Than It Looks?

Ironically, the macro environment may now favor C3.ai more than at any point in its history.

The enterprise world has moved from:

“Should we use AI?”

“How fast can we operationalize AI?”

This shift potentially benefits enterprise orchestration platforms.

C3.ai focuses on:

predictive maintenance
supply chain optimization
defense readiness
manufacturing intelligence
energy optimization
fraud detection
generative AI for enterprise workflows

These are real business applications — not chatbot hype.

The problem: brutal competition!

C3.ai now competes with giants including:

Unlike earlier years, C3.ai is no longer a first mover.

Execution now matters far more.

5. Customers, Contracts & Existing Relationships

This is where the bull case becomes more compelling.

C3.ai already serves meaningful enterprise and government customers.

Notable historical and ongoing customers/relationships include:

Baker Hughes
United States Air Force
United States Department of Defense
Shell
3M
Bank of America
Cargill
Koch Industries

Key contract: U.S. Air Force

One of the most important developments was expansion of C3.ai’s U.S. Air Force relationship.

In 2025, the contract ceiling increased to $450 million through 2029, focused on predictive maintenance and readiness analytics across military aircraft fleets. This is highly relevant because defense AI spending is growing rapidly.

For someone with our interest in NATO and defense modernization, this is one of the stronger parts of the thesis.

Baker Hughes relationship

The multi-year renewal with Baker Hughes through 2028 remains strategically important because it embeds C3.ai into energy-sector digital transformation.

This partnership gives C3.ai credibility and a distribution mechanism into:

oil & gas
chemicals
industrial infrastructure

6. Potential Future Customers & Growth Areas

If the turnaround works, growth likely comes from six areas:

1. Defense & NATO modernization

Military predictive maintenance, logistics, battlefield readiness, fleet optimization.

2. Utilities & power grids

AI optimization of increasingly strained power systems.

3. Manufacturing

Industrial AI remains underpenetrated.

4. Energy sector

Oil, gas, LNG, chemicals, carbon optimization.

5. Financial fraud detection

Banks increasingly require AI risk systems.

6. Government agencies

Federal AI modernization remains in early innings.

In other words:

C3.ai participates in many of the same long-duration themes you already like:
AI + defense + industrial modernization + infrastructure.

7. Bull / Base / Bear Scenarios

Scenario	What Happens	Possible Stock Outcome
Bull Case (30%)	Siebel fixes execution, revenue reaccelerates, defense + enterprise wins expand	2x–4x+ upside
Base Case (40%)	Slow stabilization, moderate growth	Limited but respectable upside
Bear Case (30%)	Revenue keeps deteriorating, hyperscalers dominate	Value trap / further downside

The market is currently pricing something closer to the bear case.

That is why speculative investors are interested.

Final Investment View

C3.ai today resembles a high-risk founder-led turnaround, not a broken meme stock.

The biggest reason to consider it is simple:

Tom Siebel is back, and the stock is deeply unloved.

That combination has historically created opportunities.

But this is not yet investable as a “core AI position” like your existing AI tollbooth thesis (MRVL, CRDO, QCOM, etc.).

Instead, I would view it as:

A speculative optionality bet on a founder-led turnaround

For a Canadian retail investor:

TFSA approach: small position sizing, gradual accumulation, and only if willing to tolerate major volatility.

The single most important metric to watch:

Quarterly revenue stabilization and reacceleration.

If revenue turns upward while sentiment remains negative, that is when C3.ai could rerate quickly.

NOTE: This weeks "Shell" news may be critical for an eventual turnaround story!

this is actually more important than the headline first suggests.

C3.ai announced an expanded multi-year agreement with Shell this week (June 4) to scale AI-powered reliability and predictive maintenance across Shell’s global asset operations. Importantly, this is not a pilot project or “proof of concept.” It is an expansion of an existing long-term relationship that began in 2018, which is exactly the type of evidence turnaround investors want to see.

Here is why I think this matters:

1. This validates that Shell is getting real economic value

Shell is not experimenting here.

C3.ai says the existing deployment already monitors 13,000+ pieces of industrial equipment and has generated “hundreds of millions of dollars” of economic value through reduced downtime and improved reliability. Shell is now expanding the relationship instead of shrinking it.

In enterprise software, especially industrial AI:

Renewals and expansions are often more important than flashy new logos.

If Shell were unhappy, they would not deepen the relationship.

That is a meaningful signal.

2. This is moving beyond “AI monitoring” into Agentic AI

The new agreement reportedly adds:

AI-agent root cause analysis
diagnostic automation
remediation recommendations

In simple terms:

Old system:

“Something is wrong with compressor #14.”

New system:

“Compressor #14 is likely failing because vibration + heat + pressure trends resemble three prior failures. Recommended intervention: X.”

This is a much more valuable product category because it moves from detection → diagnosis → action.

Given our broader thesis around Agentic AI, this part is important.

C3.ai may actually have an underappreciated niche in industrial agentic AI, especially for:

energy
utilities
chemicals
defense logistics
heavy manufacturing

3. Shell could become a “reference customer” for the energy industry

This may be the most underrated aspect.

Energy companies tend to copy proven deployments.

If Shell demonstrates strong ROI, it increases the probability of:

ExxonMobil
Chevron
BP
TotalEnergies
LNG operators
utilities
industrial process companies

expanding industrial AI budgets.

C3.ai already has credibility in energy through both Shell and Baker Hughes, which creates an ecosystem effect. The long-running relationship with Baker Hughes was also expanded in 2025 to continue AI deployment in energy and industrial markets.

4. Why this matters to the turnaround thesis

For me, this is incrementally bullish, but not thesis-changing by itself.

What it does prove:

✅ Major customers are staying
✅ At least one flagship customer is expanding spend
✅ The product appears to deliver measurable ROI
✅ C3.ai still has enterprise relevance
✅ Siebel’s “industrial AI” thesis may not be broken

What it does NOT yet prove:

❌ Revenue reacceleration across the company
❌ Broad customer momentum
❌ Sustainable growth recovery

In other words:

The Shell news is evidence that C3.ai may still have a strong product in certain verticals.

The open question remains:

Can Tom Siebel turn isolated successes into company-wide execution again?

My interpretation for an investor

If I were building the turnaround case, I would put this development in the “important confirming evidence” bucket.

Not a reason alone to buy.

But if over the next 2–3 quarters we also see:

more defense wins,
additional industrial expansions,
stabilization in revenue,

then this Shell expansion starts to look like...

the first sign of a real turnaround rather than random good news.

Monday, June 8, 2026

June 8th, This week in our Retire fund portfolio! Antennae up!

Signals this week are mixed-to-cautiously bullish with elevated correction risk, especially because the Nasdaq and S&P are trading near historically stretched multiples while macro risks are reappearing. This week, I would characterize the setup as “uptrend intact, but fragile.”

The Bull Case (Why markets may still move higher this week)

Despite expensive valuations, three forces continue to support North American equities:

AI capex and earnings momentum remain very strong
Large-cap technology and semiconductor spending are still accelerating. Institutions continue to treat AI as a multi-year infrastructure cycle rather than a short-term hype phase. That has kept flows into the Nasdaq despite high multiples.
Corporate earnings are still outrunning recession fears
Wall Street strategists remain broadly constructive on 2026 because earnings growth expectations have held up better than feared. Goldman recently raised its S&P target, arguing earnings growth is offsetting valuation concerns.
Rate-cut expectations still matter (but are wobbling)
Markets still expect eventual easing, which supports high-multiple growth stocks. However, stronger economic data has recently pushed bond yields higher, complicating the “multiple expansion” story.

The Bear Case (Why this week could turn volatile)

This is where I think investors need to pay close attention:

1. Valuations are stretched

The S&P 500 and Nasdaq are near record highs with multiples that leave little room for disappointment. Historically, when markets get this expensive, good news is priced in quickly, but bad news hits hard.

2. Bond yields are rising again

One of the biggest risks to high-growth stocks is rising yields. When yields climb, future earnings get discounted more heavily — and richly valued tech names feel it first. This matters especially for AI leaders and the “second derivative” names you follow.

3. Geopolitical and inflation risks are back

Oil volatility, Middle East tensions, tariff uncertainty, and sticky inflation are resurfacing as risks. Reuters noted that stronger jobs data and renewed inflation concerns have already pressured tech sentiment heading into this week.

4. Narrow leadership = warning sign

A lot of the gains remain concentrated in a relatively small group of AI-related winners. When breadth narrows too much, markets often become more vulnerable to pullbacks.

My Base Case for This Week (June 8 week)

Probability-weighted view:

Scenario	Probability	What it looks like
Range-bound / mild pullback	~45%	2–5% weakness in Nasdaq; profit-taking in AI leaders
Continued melt-up	~35%	Markets shrug off valuation concerns and grind higher
Sharp correction	~20%	Inflation/yields or geopolitics trigger 5–10% selloff

Given the setup, I would expect higher volatility and sector rotation rather than a market crash. The most likely outcome is choppiness with selective weakness in expensive AI names while industrials, defense, energy, financials or value rotate in and out.

For someone with our thesis (AI + quantum + defense + silver/critical materials), I would be more inclined to:

Trim extended winners only if position sizing has become outsized
Keep dry powder for forced selloffs in quality AI infrastructure names
Expect silver, defense and energy-adjacent names to potentially act as partial hedges if inflation/geopolitics rise again
Focus on second-tier picks-and-shovels rather than only the mega-caps at peak multiples

One metric I would watch closely this week: the U.S. 10-year Treasury yield. If yields keep climbing while the Nasdaq stays expensive, the probability of a meaningful pullback rises materially. Conversely, if yields settle, the AI rally can continue longer than most expect.

A fair way to summarize the market right now is:

“Fundamentals still support higher prices, but valuations mean the market is increasingly unforgiving.”

Given our existing themes (AI infrastructure, quantum, defense/NATO, silver/critical materials, biotech) and the current setup of high multiples + rising yield risk, I would frame this week as a “barbell market”: investors may continue chasing AI winners while simultaneously rotating into hard assets, defense, and cash-generating businesses.

Scenario A: Market Continues Higher This Week (“Melt-Up”)

Most likely winners (ranked):

1. AI Infrastructure / Picks-and-Shovels (highest probability of alpha)

This remains the strongest momentum trade if yields stabilize.

Why: Institutions are still underweight relative to the size of the AI buildout. Spending on networking, memory, optics, power, and cooling continues regardless of short-term macro noise.

Best-positioned categories:

Networking / interconnect
- Marvell Technology
- Credo Technology Group
- Broadcom
Memory / HBM
- Micron Technology
Cooling / power
- Vertiv Holdings
- Eaton

What tends to outperform in melt-ups:
The second-tier AI names (our preferred hunting ground) often outperform the Magnificent Seven because they are less crowded and still rerate upward.

2. Defense / NATO Buildout

If geopolitical headlines intensify, defense could outperform even during a broad rally.

Canadian names we already favor:

Kraken Robotics
Volatus Aerospace

U.S./Europe anchors:

Palantir Technologies
Rheinmetall
RTX Corporation
Equinor

Why this week:
Defense increasingly behaves like a structural growth sector, not just a recession hedge.

3. Quantum (High Beta)

If risk appetite remains strong, speculative capital may flow back into quantum names.

Highest-beta public proxies:

IonQ
D-Wave Quantum
Rigetti Computing

But: these are highly rate-sensitive. Rising yields can reverse momentum quickly.

Scenario B: 2–5% Pullback / Correction This Week

If yields rise or inflation fears intensify, I would expect this rotation:

1. Precious Metals & Silver (best hedge in our framework)

This aligns closely with our thesis.

Why silver may outperform in a wobble:

Safe-haven demand
Industrial AI/datacenter/robotics demand
Persistent supply tightness

our favorites remain strong:

First Majestic Silver
Endeavour Silver

Also strong:

Pan American Silver
Wheaton Precious Metals

ETF/holding hedge:

Sprott Physical Silver Trust

2. Energy / Utilities / Power Infrastructure

If inflation reaccelerates, power infrastructure may quietly outperform.

Interesting names:

GE Vernova
Siemens Energy
BWX Technologies

3. Profitable Cash-Flow AI Enablers

If markets wobble, speculative AI often sells off first while profitable tollbooth names hold better.

Examples:

Nasdaq (Verafin thesis)
Qualcomm
International Business Machines

What Usually Gets Hit First in a Correction

These are the categories I’d expect to struggle first:

Unprofitable high-beta AI stories
Small-cap speculative quantum
Long-duration biotech (especially pre-revenue)
Overextended semis trading at extreme multiples

That means names like smaller quantum/speculative biotech can become opportunities, but often after the first flush lower, (which occurred last week)

My Ranking of “This Week” Opportunity Buckets

If market stays strong:

AI infrastructure
Defense/NATO
Quantum
Silver miners
Biotech

If market weakens:

Silver / precious metals
Defense
Power infrastructure
Profitable AI tollbooths
High-beta AI after selloff

The One Thing I Would Watch Daily This Week

If the 10-year U.S. Treasury yield rises sharply and the Nasdaq still rallies, that divergence usually breaks one way or another — and often violently.

Rule of thumb this week:

Yield down / stable → risk-on
Yield up sharply → expect rotation or pullback
Oil spike + yield spike → silver & defense likely outperform

For a Canadian retail investor in our position, this looks less like a week to “go all in” and more like a week to prepare buy lists and scale into weakness selectively rather than chase.

Thursday, June 4, 2026

The case for owning silver stocks/ETFs at a time of severe shortages in this precious/technology metal

SILVER 2026–2030

Investment/Business Report Including Optimal TFSA Weighting Strategy

Executive Summary

Silver has evolved into one of the world’s most strategically important commodities.

Historically viewed as a precious metal, silver is now increasingly essential to:

AI infrastructure
Robotics
Data-center electrification
Aerospace
EV systems
Grid modernization
Defense technologies
Renewable energy

At the same time, silver still functions as:

a hard asset
inflation hedge
monetary protection
geopolitical safe haven

This creates a rare dual-demand dynamic:

Silver benefits when technology booms AND when monetary systems weaken.

That combination is unusual.

Gold tends to benefit primarily from fear.

Silver can benefit from growth + fear simultaneously.

Why Silver Matters in the AI/Robotics Era

Why our core thesis is increasingly being validated.

AI is not merely software.

It is an industrial and electrical buildout.

The world is now constructing:

hyperscale data centers
robotics factories
autonomous systems
power networks
electrical switching systems
advanced semiconductors
defense electronics

Silver is deeply embedded throughout this infrastructure because it has:

The highest electrical conductivity on Earth

The highest thermal conductivity

Exceptional reliability in high-performance electronics

This is why silver appears in:

servers
semiconductors
connectors
relays
robotics
EVs
aerospace electronics
precision military systems

The AI revolution is therefore partly a metals story.

And silver is increasingly one of its hidden beneficiaries.

The Structural Silver Deficit

This is perhaps the strongest pillar of the thesis.

Silver has entered repeated annual deficits where:

demand > supply

And the market cannot easily fix it.

Why?

Because most silver is not mined intentionally.

Roughly 70%+ comes as a by-product of:

copper mining
zinc mining
lead mining
gold mining

Meaning:

Even much higher silver prices may not rapidly increase supply.

This is different from gold.

The result:

prolonged shortages become possible.

That is one reason many institutional investors increasingly

view silver as a strategic scarcity asset.

Analysis of our Four Holdings

1. Endeavour Silver Corp.

Role: Mid-Cap Silver Torque

EDR gives us leveraged exposure to rising silver prices.

Silver miners often move 2–5x faster than silver itself in bull markets because margins expand dramatically.

Why we own it

✔ Production leverage

✔ Expansion optionality

✔ Re-rating potential

✔ Strong upside in silver squeeze scenarios

Risks

✘ Mexico operational risk

✘ Execution risk

✘ High volatility

Role in TFSA

Growth engine

2. First Majestic Silver Corp.

Role: High-Beta Silver Conviction Play

AG has historically behaved like a high-octane silver vehicle.

Few silver miners react as aggressively to sentiment and metal price appreciation.

Why we own it

✔ High silver sensitivity

✔ Strong retail following

✔ Brand power in silver investing

✔ Potential upside in squeeze environments

Risks

✘ Extremely volatile

✘ Can fall hard in corrections

✘ Emotionally difficult to hold

Role in TFSA

Alpha accelerator

3. XGD

Role: Precious Metals Shock Absorber

This stabilizes the portfolio.

While silver miners may move violently, XGD offers:

larger miners
diversified precious metals exposure
gold downside protection

Gold tends to outperform during:

recessions
financial stress
liquidity crises

Why own it

✔ Lower volatility

✔ Diversification

✔ Crisis hedge

✔ Better drawdown control

Role in TFSA

Emotional stabilizer

4. Sprott Physical Silver Trust

Role: The “Real Silver” Core

This is your pure bullion exposure.

No mine failures.

No cost overruns.

No political risk.

Just silver.

Why own it

✔ Direct silver ownership

✔ Physical backing

✔ No mining risk

✔ Long-term monetary hedge

Role in TFSA

"Foundation asset"!

The Weighting Question

Maximum Alpha While Still Sleeping at Night

Asking the right question.

The answer depends on balancing:

upside potential

against

psychological survivability

three models.

OPTION 1 — “Sleep at Night / High Conviction”

Best balance for most investors

Holding	Weight
PSLV	35%
XGD	30%
AG	20%
EDR	15%

Why this works

You still participate strongly if silver runs.

But drawdowns become manageable.

If miners crash temporarily, your bullion + gold exposure softens the blow.

Expected personality fit

8/10 confidence for your TFSA

OPTION 2 — “Maximum Alpha but Still Rational”

My preferred fit I would advise for family members

Holding	Weight
PSLV	30%
AG	30%
EDR	25%
XGD	15%

Why I like this for style

This aligns closely with:

our AI infrastructure thesis
robotics conviction
silver scarcity belief
willingness to own volatility

Yet:

45% remains defensive

(PSLV + XGD)

while

55% is torque

(AG + EDR)

This could materially outperform if silver enters a true secular bull market.

Why this may be ideal

You are still able to:

“sleep at night”

without sacrificing meaningful upside.

Expected personality fit

9/10 fit for you

OPTION 3 — “Aggressive Silver Supercycle”

Maximum upside / hardest emotionally

Holding	Weight
AG	40%
EDR	35%
PSLV	20%
XGD	5%

Reality check

This could massively outperform.

But:

You must emotionally tolerate:

-40% to -50% drawdowns

even if the long-term thesis remains intact.

Most investors fail psychologically here.

Expected personality fit

6/10 for your TFSA

(too emotionally demanding)

My Preferred Recommendation

If I were optimizing specifically for:

TFSA tax efficiency

Silver supercycle exposure

AI/robotics tailwinds

High alpha potential

Ability to hold through volatility

I would lean toward:

30% PSLV / 30% AG / 25% EDR / 15% XGD

Why?

Because it accomplishes four things:

1. Maintains direct silver exposure

(PSLV)

2. Captures explosive upside

(AG + EDR)

3. Avoids becoming emotionally unmanageable

(XGD stabilizer)

4. Preserves TFSA compounding

Remember:

Inside a TFSA:

5–10x winners become extraordinarily powerful because gains are tax free.

That favors selectively embracing volatility.

One Additional Suggestion

Given our conviction level:

I would also consider a dynamic weighting model.

During silver pullbacks:

add to EDR/AG

During euphoric silver spikes:

trim miners slightly

Move gains into PSLV

This slowly converts:

speculative upside → hard-asset security

over time.

That is often how long-term precious-metals investors quietly compound wealth through cycles.

My overall view:

This four-position setup is actually quite sophisticated for a retail investor because it combines:

physical silver + torque + diversification + monetary protection

rather than betting entirely on one outcome.

ED Note: Regarding compression (see above) silver appears now to be at an inflection point for a push much higher.

Saturday, May 23, 2026

As Anthropic and OpenAi begin the IPO dance, we look at some second tier plays that shoukd return more alpha

The Year of Mega IPOs

Why Second-Tier Infrastructure Companies Could Produce the Greatest Alpha

A Retail Investment Thesis Built Around MRVL + CRDO

Executive Summary

Many retail investors will instinctively try to buy the coming AI IPOs:

Anthropic
OpenAI
potentially future agentic AI leaders and infrastructure platforms

That instinct may be wrong.

Historically, the largest wealth creation in platform revolutions often came not from the headline companies, but from the second-tier tollbooths enabling the ecosystem.

Think:

Internet → Cisco, Qualcomm, Broadcom
Smartphones → TSMC, Qualcomm, ASML
Cloud → Nvidia, Arista, Equinix
EVs → semiconductor and battery suppliers

The argument here is:

The largest risk-adjusted AI alpha from 2026–2029 may not come from buying Anthropic or OpenAI at trillion-dollar valuations. It may come from owning the infrastructure companies required to make them function.

That is where the MRVL + CRDO thesis becomes compelling.

Anthropic and OpenAI are both increasingly expected to pursue IPOs in 2026, amid extraordinary investor enthusiasm around frontier AI. Recent reporting suggests OpenAI and Anthropic could be among the largest IPOs in history, with valuations approaching the trillion-dollar range.

Part 1: Why 2026 Could Be “The Year of AI IPOs”

The market is entering what could become:

The public monetization phase of the AI revolution

We are moving from:

Phase 1 (2023–2025)

GPU scarcity / model training

Winner:

NVIDIA

Phase 2 (2025–2027)

Agentic AI deployment

Winners:

Anthropic
OpenAI
enterprise AI ecosystems

Phase 3 (2026–2029)

Infrastructure scaling

Likely winners:

networking
optics
interconnect
memory movement
AI compute orchestration

This shift matters enormously.

The market is beginning to realize:

AI does not scale linearly.

Every leap in intelligence requires:

exponentially more bandwidth,
lower latency,
greater memory movement,
more energy efficiency,
larger AI clusters.

Anthropic’s rapid growth and massive compute commitments illustrate the scale of infrastructure required.

Recent reports indicate Anthropic has committed to extraordinary compute spending and is scaling aggressively to support Claude and future agentic systems.

Part 2: Why Buying Anthropic/OpenAI IPOs May Not Produce the Best Alpha

This may sound counterintuitive.

But by IPO:

OpenAI and Anthropic may already be priced for perfection.

Potential issues:

1. Massive valuations

Reports now discuss valuations:

OpenAI: ~$850B–$1T
Anthropic: hundreds of billions approaching $1T

At those levels:

future upside becomes mathematically harder.

A stock at a $900B valuation doubling to $1.8T is possible—but far harder than a $60–$100B infrastructure supplier tripling.

2. Capital intensity risk

AI model companies burn extraordinary capital.

Anthropic reportedly spends billions on compute and infrastructure to maintain frontier capability.

Retail investors may discover:

Owning the “brains” is expensive.

Sometimes:

owning the shovels is better!

3. Commoditization risk

Over time:

Claude, GPT, Gemini, xAI, and others may compete aggressively.

Margins could compress.

But:

the infrastructure still gets paid.

Whether OpenAI wins or Anthropic wins:

"Data still moves no matter who wins or how systems eventually commoditize".

Part 3: The Real Bottleneck = Moving Intelligence

This is the core thesis.

Most investors still think:

AI = chips.

That is increasingly incomplete.

The next bottleneck appears to be:

data movement

Meaning:

Compute cannot function without:

Networking
Interconnect
Optical systems
Memory fabrics
Low-power transmission

This framework is becoming increasingly correct:

GPU boom → networking boom → photonics boom

Part 4: Why MRVL Matters

Marvell Technology = The “AI Infrastructure Backbone”

Marvell sits at the intersection of:

custom AI silicon
networking
optical interconnect
cloud AI scaling
hyperscaler architecture

Importantly:

Marvell is deeply tied to Amazon Trainium, which is highly relevant because Anthropic increasingly depends on AWS infrastructure.

Amazon and Anthropic expanded their collaboration in 2026 around Trainium compute and large-scale cloud commitments.

Why MRVL could outperform expectations

Marvell is selling:

"The roads AI travels on"!

Whether:

Anthropic wins,
OpenAI wins,
xAI wins,
or all of them win,

Marvell still benefits.

That diversification matters.

Strengths

✔ Lower risk than smaller AI names
✔ Multiple hyperscaler exposure
✔ AWS/Trainium leverage
✔ AI networking leadership
✔ Strong institutional ownership

Weakness

❌ Already well discovered by Wall Street

Part 5: Why CRDO Matters

Credo Technology Group = The Hidden AI Bottleneck

This is the higher-alpha piece.

Credo focuses on:

high-speed connectivity
optical DSPs
Active Electrical Cables (AECs)
ultra-efficient interconnect

As AI clusters become larger:

bandwidth becomes everything.

Credo increasingly positions itself as a connectivity-at-scale company for hyperscaler AI environments, with major pushes into optical solutions for AI fabrics.

Recent growth has been explosive, driven by hyperscaler demand and AI networking expansion.

Why CRDO could become a multi-bagger

Because investors may still underestimate:

how much data movement Agentic AI requires.

Agentic systems are not simple chatbots.

They reason.

They call tools.

They chain models.

They coordinate across systems.

That creates:

massively larger networking demand.

Part 6: The Combined Thesis

Why MRVL + CRDO together makes sense

Building an

AI Tollbooth Portfolio

MRVL = stability + platform exposure
CRDO = asymmetric upside + networking torque

Why this pairing works

Factor	MRVL	CRDO
Risk	Lower	Higher
Upside	Strong	Very High
Anthropic relevance	High	Indirect but meaningful
Agentic AI leverage	High	Extremely high
Valuation risk	Moderate	Higher
Hyperscaler exposure	Broad	Concentrated

The combination reduces risk while preserving upside.

Suggested Retail Allocation

For a retail investor seeking:

alpha without excessive concentration risk

I currently favor:

60% MRVL / 40% CRDO

Why?

Because:

MRVL acts as the anchor, while CRDO provides the torque.

In portfolio construction terms:

MRVL lowers the probability of catastrophic disappointment.

CRDO raises the probability of outsized returns.

Risks to the Thesis

1. AI capex slowdown

If hyperscalers pause spending:

Both stocks may correct sharply.

2. IPO disappointment

If OpenAI/Anthropic IPOs underperform:

AI sentiment could temporarily weaken.

3. Valuation compression

Especially for CRDO.

4. Networking commoditization

Competition from:

Broadcom
Nvidia
internal hyperscaler solutions

Bottom Line

The smartest way for a retail investor to play the Year of AI IPOs may not be buying the IPOs themselves.

Instead:

buy the companies that must win regardless of which AI lab dominates.

Among second-tier infrastructure companies:

MRVL + CRDO is one of the strongest two-stock AI infrastructure theses I currently see for 2026–2029

because it aligns directly with what I believe becomes the next great bottleneck:

"The movement of intelligence itself"!

Ed Note:

I have no current shares of either MRVL or CRDO at present, but have placed them on our watch list for now!