ΣDX Performance Strategy
→ Where Data Becomes Performance
NSO Capital presents a disciplined investment strategy built to capture the economic value created by AI, real-time data exchange and automation.
Last Updated: May 7th 2026
NSO Capital | Asset Management | New York & Paris | contact@nso-capital.com | www.nso-capital.com

1

Strategic Overview

2

A Structural Transformation Is Underway
Digital infrastructure, accelerated computing, and artificial intelligence, powered by real-time data sets and advanced models (LLMs and LQMs), are becoming the planning and operating layer of the global economy.
Every major sector:
  • Information Technology
  • Financial services
  • Healthcare
  • Industrials
  • Energy
  • Retail (Platformization),
is being rebuilt around platforms powered by agentic software, semiconductors and automated reasoning allowing for BIS (Business Intelligence Services & Systems) to operate.
This is not a cycle, it's a secular transformation.
It is a multi-trillion-dollar capital investment program that is reshaping corporate earnings and market leadership for the next decade.

3

Two Forces Working Symbiotically
Policy and Industrial Strategy Shift
U.S. reshoring, semiconductor sovereignty, deregulation, energy incentives.
"Mar-a-Lago Accords" accelerate domestic tech dominance through fiscal and trade architecture. Technology and national security now aligned as a single policy directive.
AI Buildout Cycle
$8–10 trillion in projected global capex.
Largest technology infrastructure cycle since post-WW2 industrial expansion. Visibility into multi-year revenue and earnings growth for core enablers.
Together, these conditions create a rare environment where policy support, industrial business strategy, capital flows and secular growth move in the same direction.

4

Changing Fact Patterns Dictate Changes in Action
The facts are changing as computing power accelerates. Therefore, capital allocation must change, because the world has moved to 'real-time data' with accelerated compute.
At the turn of the millennium, the internet digitized information and reshaped the global economy. A similar inflection point is now underway.
As a result, several long-standing market assumptions no longer apply:
  • Earnings power shifts to companies that automate, not accumulate labor
  • Data is the currency of value creation
  • Platform economics replace linear revenue models
  • Automation, not human workflow, will drive productivity
  • Cost of capital is structurally different
  • Compute will become cheaper than labor

Real
Time
Data
with accelerated computing.

*According to MIT’s Iceberg Index, AI now has the demonstrated capacity to handle almost 12% of U.S. labor-market tasks.
This reflects technical exposure, not guaranteed displacement, but it underscores the scale of transformation underway.

5

What ΣDX Means
∑ Semiconductors + Agentic Software → DX (Data eXchange)
Σ = Semiconductors
The foundation of accelerated computing
Σ = Agentic Software
The intelligence layer
DX = Data eXchange
Real-time information flow
When semiconductors and agentic software enable vast data exchange at scale, productivity and valuation inflect. DX enables real-time analytics, reveals mispriced securities faster than traditional research, improves hedging and risk-adjusted returns, and accelerates competitive advantage.
ΣDX = where technology, information, and capital converge to create performance.

6

The Window of Opportunity:
Why the window is finite
Mass deployment of infrastructure, BIS and applications between now and the next decade.
  • Capex today → revenue flywheels tomorrow
  • Many winners not yet obvious, just as in 1998–2002

7

The Scale of Capital Commitment
$2.5T+
Committed
Capital committed to hyper-scaler infrastructure
$1T+
Training Market
AI training represents a trillion-dollar+ market
Even Larger
Inferencing & BIS Market
AI inferencing projected to exceed training market
Microsoft, Meta, and Google are reporting demand exceeding supply. The AI buildout is one of the largest coordinated capital-spending events in modern financial history.
"AI inferencing will change everything." — Larry Ellison

8

9

The Other Side of WACC
Debt Financing & Hyper-scaler Capital Structure
Rates continue to hover around the 4% line. Debt financing is becoming the critical variable in hyper-scaler economics.

Hyper-scaler CapEx accelerating to ~$700B in FY '26 — with an ever-growing percentage funded through debt and hybrid instruments for AI factories capable of inferencing, LLMs, LQMs, and agentic AI.
The Shift
  • From FCF-based funding
  • To risk-based capital structures
  • Debt, hybrid instruments, joint ventures, capital markets
The Implications
  • WACC assumptions change
  • Model sensitivity increases
  • New opportunities and risks emerge
  • Example: Oracle's debt-funded infrastructure play

As debt becomes the primary funding mechanism, the cost of capital becomes the critical variable in valuation models, not just earnings growth.

10

11

12

Mar-a-Lago Accords: State-Directed Capitalism
Current US policy shifts are moving US industrial strategy back on shore via a form of state-directed capitalism focused on strengthening strategic sectors: semiconductors, AI infrastructure, energy, and advanced manufacturing.
Drive Rates Lower
Keep interest rates materially lower and maintain them
Redirect Capital
Channel investment into AI, semiconductors, & energy infrastructure
Support Dollar Dominance
Maintain U.S. dollar strength via technology and manufacturing, not trade
This is state-directed capitalism; not to replace markets but to accelerate them. Capital, incentives, and deregulation are working in the same direction: toward U.S. technological supremacy. All the while, as the US sovereign benefit from reducing fiscal drag.

13

Policy Supportive of Lower Rates
Treasury Can Refinance at Low Rates
The Treasury can refinance at low rates for the next decade. AI infrastructure needs cheap capital to scale. Disinflationary forces from automation reduce labor costs and price pressure.
Investor Implications
  • UST10 yields remain around 4%
  • Mortgage normalize to the 5.5% level
  • Credit spreads are likely to widen due to issuer crowding out

When the policy objective is technological supremacy, capital needs to be abundant and inexpensive.

14

Inflation & Labor: Second-Order Effects
1
Worker Displacement
Larger than forecast as AI systems replace repetitive human work
2
Inflation Stabilizes
Over 3 to 5 years (if not sooner) as productivity gains offset cost pressures
3
Transformational Opportunity
A decade of change across labor, supply chains, and capital markets
Productivity rise → margin expansion → valuation support.
This is not a hype cycle; it's a structural reset.

15

The Investment Question
How do you allocate into the largest technology buildout in modern history without chasing hype, peak valuations, or excessive volatility?
The answer → Through ΣDX Performance
A strategy built to capture value through the prism of accelerated and enhanced data flows, high-performance BIS and applications, companies where data becomes monetization, and capital structures that reveal mispriced risk and emerging winners.
AI is not only transforming products, it is transforming how information is created, transmitted, priced, and traded in real time.

16

ΣDX Performance: Value Identification Framework
1
Buy Infrastructure Before Monetization Is Visible
Identify structural winners in the build phase
2
Exploit Cross-Capital-Structure Mispricings
Find value across equity, credit, and hybrid securities
3
Capture Upside with Controlled Risk
Maintain liquidity and disciplined hedging
4
Participate in Exponential Value Creation
Avoid speculative excess while capturing secular growth
The trade is not "buy AI." The trade is to invest where data becomes performance.

17

Market Environment

18

Monetary Debasement: The Macro Backdrop
Structural Reality
Developed economies face persistent fiscal deficits and shrinking foreign demand for sovereign debt, increasingly relying on domestic buyers, including stablecoins backed by U.S. Treasuries.
Probability of Monetary Debasement
Lower real rates, aggressive industrial policy, and fiscal dominance converge to push policy toward currency dilution, prioritizing growth over austerity.
AI’s Role in the Equation
AI is inherently deflationary, enabling productivity gains that offset the inflationary impact of expanding debt and making the multi-trillion-dollar buildout cheaper to finance.
Market Signals
Gold is at record highs, reflecting rising strategic demand for rare earths, and continued outperformance in AI infrastructure.
Investment Implication
Hold the beneficiaries of debasement: AI infrastructure, scarce resources, and assets tied to real productivity gains, poised for long-term value creation.

19

The Gold Trade: Safe Haven in a New Monetary Regime
1
Why Gold Matters Now
  • Safe-haven demand surging as fiscal positions deteriorate across Western economies
  • Global central banks & SWFs buying record amounts as a hedge against inflation and diminished central bank appetite for UST
2
The Fiscal + Treasury Demand Link
  • U.S. deficit: $1.8T → with traditional foreign buyers (China, Japan, oil states) pulling back
  • Funding gap emerging → gold becomes a hedge against long-term deficit monetisation
  • Gold's rise parallels declining global appetite for long-duration U.S. debt
3
Why This Supports Gold
  • Gold = non-sovereign monetary anchor in a digital-currency tech-driven world
  • Hedge against diminished foreign Treasury demand for USTs.
  • Central bank accumulation + real-rate decline + deficit stress → structural tailwind

20

Gold Holdings by Country
Central banks and governments are buying gold, not Bitcoin. The widening spread indicates preference for regulated, sovereign store-of-value assets and governments preparing for long-term fiscal strain.

21

UST Bond Market
The U.S. Treasury market is not behaving according to traditional rate-cycle models. Elevated issuance in bond markets is expected to continue due to fiscal demands and the funding of the AI buildout.
  • The Federal Reserve rate-cutting cycle is expected to deliver 1-2 more rate cuts for 2026
  • The Central Bank has entered a renewed period of modest QE
  • The rate of change in deflation is expected to accelerate due to automation and productivity gains
  • Kevin Warsh has been nominated to head the Federal Reserve in May 2026 pending Senate confirmation
Rate volatility is an opportunity, not a danger. The trajectory of interest rates should be structurally lower once policy alignment is completed.

22

Structural Drivers: The Next Decade
The next decade is defined by forces that push inflation down, expand margins, and reward technology-enabled businesses. These are structural, not cyclical.
Productivity Rise
Advances in automation becomes cheaper than hiring, shifting capex priorities.
AI Dislocation of Labor
Systems replace repetitive human work, reducing wage pressure and expanding profit margins.
Demographics
Aging population and fewer births create structural labor constraints.
Restrictive Immigration
Policy changes accelerate the need for automation solutions.
Growth shifts from labor to computing. Earnings accrue to companies that automate.

23

Capital Flight to U.S. Tech
Global capital is moving toward the United States at a scale not seen in decades. The driver is not interest rates—it is technology dominance.
Why Capital Is Rotating to the U.S.
  • Deepest, most liquid capital markets in the world
  • U.S. largely controls the global AI stack (semiconductors, compute, cloud, software), with China a close second
  • Sovereign wealth funds, pensions, insurers and endowments are increasing U.S. tech exposure in both private and public capital markets
  • Portfolio managers prefer dollar assets in geopolitical uncertainty
What the Data Shows
  • Multinational corporations repatriating supply chains into U.S. jurisdiction
  • Record buying of U.S capital market assets
  • SWFs driving large allocations into semiconductors, AI & defense-related sectors
  • Private capital markets driving the demand for many upstart data centre chip and inferencing platforms

24

Capital Waiting on the Sidelines
$15T
State Capitalism
SWFs collectively manage $15T in
AUM as of mid-2025
90%
Wealth Concentration
Top 10% of U.S. households own
90% of public markets
$15-30T
Wealth Transfer
Baby-boomer wealth transfer creating capital rotation
When rates fall, that cash has to leave the sidelines and chase performance. This represents a massive structural tailwind for equity markets.

25

AI Factory Scaling:
Chronic Supply Shortage
The bottleneck is supply, not demand. That's a structural earnings tailwind.
Structural Reality
GPU capacity shortages across every hyper-scaler. Lead times measured in quarters, not weeks. Demand is accelerating, not slowing.
Order Books Are Full
AI factory constructors sign multi-year, non-cancellable commitments with 20–30% deposits on average.
In the Event of Slower Capex
Backlog maintains multi-year revenue visibility with customers that have investment-grade balance sheets. Players locked into long-term supply contracts. Earnings growth becomes durable, not cyclical.

26

Why AI Is Reshaping Every Sector
"AI training is a multi-trillion-dollar market—but the inferencing market will be much, much larger. AI will run factories, finance, medicine, legal systems… and even write the agents that automate every business function."
— Larry Ellison
AI is writing the software that automates industries. AI Agents will run:
  • Factories
  • Financial markets
  • Banking & legal systems
  • Healthcare & drug simulation
  • Sales & Operations
  • Agentic Customer service
  • Supply chains
  • Transport & Electric Vehicles

27

Corporate Adoption: From Pilots to Budgets
AI spending is shifting from pilot projects to committed budgets. Companies are buying AI services now, with productivity gains built into forecasts.
Committed Budgets
CFOs allocating for automation and cost reduction
Integrating Automation
Apps can only scale once real inferencing capacity exists.
Revenue Visibility
Multi-year revenue visibility for AI suppliers

28

Structural Inflection in AI Adoption
Observed Market Conditions
  • The ROIC-AI debate is real
  • Infrastructure demand far exceeds supply
  • Adoption is accelerating
  • Capital deployment is massive
  • Regulatory conditions are favorable

Structural Constraints in Traditional Frameworks
  • Valuation models lag the economics of AI and inferencing
  • Capex-to-revenue timing distorts near-term valuation signals
  • Short-term price volatility is emphasized over long-term earnings durability
  • Valuation approaches often overlook balance-sheet strength and upside optionality
Caution around valuation can delay participation during the infrastructure buildout but market corrections can create valid entry points.
ΣDX Performance is designed to navigate these structural gaps through disciplined valuation and risk management.

29

Investment Framework

30

The ΣDX Performance Strategy
A disciplined investment strategy built to capture the economic value created by AI, automation, and real-time data exchange.
1
Σ (Semiconductors + Software)
The engine of compute and application scalability
2
DX (Data eXchange)
Real-time information integration for faster, more accurate decision-making
3
Intrinsic Value Discipline
Price meets long-term economic value
4
Cross-Capital Structure Analysis
Identify mispriced risk and emerging winners
5
Dynamic Risk Management
Asymmetric upside with controlled volatility

Objective: Deliver 1.5x NASDAQ Composite per annum return over a 3–5 year horizon.

31

The Logic of ΣDX
AI adoption is dictated by compute capacity, software capability, and the velocity of data exchange.
Why Σ (Semiconductors + Agentic Software)
  • Semiconductors enable compute
  • Software directs compute
  • Together determine the speed and scale of AI adoption
  • Hardware bottlenecks create pricing power
  • Software drives monetization and recurring revenue
Why DX (Data eXchange)
  • Real-time integration of financial + alternative data
  • Faster predictive analytics and risk assessment
  • Earlier identification of dislocations and mispricing
  • Information advantage → performance advantage
Σ + DX = the structural engine of the modern digital economy.

32

How ΣDX Generates Value
ΣDX focuses on the economic beneficiaries of AI; not just the headline names.
Information Moves Fast
Positions where data flows accelerate decision-making
Prices Inferencing Economy
Not just the training layer, but the application layer
Targets Beneficiaries
Companies powering and benefiting from AI adoption
Multi-Sector Opportunity
Semis, cloud, cybersecurity, data centers, analytics, automation
Identifies Winners
Structural winners, not short-term themes

33

Intrinsic Value & Cross-Capital Structure Analysis
We value companies across their entire capital structure to detect opportunity before consensus.
Intrinsic Value Discipline
Fundamental valuation across equity, preferreds, unsecured, secured, hybrids. Assess survivability, cash generation, and margin expansion. Identify price/value intersections where probability-adjusted returns are superior.
Cross-Capital Structure Analysis
Covenant stress, liquidity runway, refinancing risk. Equity optionality and cash-flow durability. Detect stress before price collapses. Detect opportunity before the market recognizes it.
Real alpha comes from seeing what others miss; identifying undervalued growth before consensus catches up.

34

Information Advantage
Speed, accuracy, and integration of information determine investment outcomes in the AI era.
Real-Time Data Feeds
Alternative data and financial data integration
AI-Powered Screening
Scenario modeling and natural-language processing
Signal Extraction
Rapid interpretation of insider activity and short-interest mapping
Earlier Price Recognition
Faster interpretation leads to earlier positioning

35

Dynamic Risk Management
Risk is treated dynamically, not passively, to protect capital while preserving upside.
Non-Traditional Hedging
Across credit, equity, CDS, and options
Volatility Oversight
To dampen drawdowns
Cross-Asset Correlation
Monitoring and adjustment
Balance-Sheet Risk
Analysis and liquidity-aware sizing
The objective: asymmetric upside, controlled downside, consistent compounding.

36

Style, Structure & Time Horizon
Style & Structure
  • Long/short
  • Event-driven
  • Cross-capital structure
  • Credit & equity-linked
  • Opportunistic hedging
Time Horizon (portfolio construction)
  • Short-term: Dislocations, earnings volatility, market overreactions
  • Medium-term: 6–18 month price normalization
  • Long-term: AI compounding over 3–5 years
The strategy blends trading agility with long-horizon secular exposure to the AI buildout.

37

Benchmark-Driven Allocation Constraints
Fear of Volatility
Reluctance to own emerging winners during buildout phase
Underestimating Inferencing Impact
Automating decisions, workflows and operations
Misunderstanding the Capex-to-Revenue Flywheel
Today's infrastructure spending becomes tomorrow's earnings
Ignoring Economic & Industrial Policy
Policy objective → technological supremacy = capital must be inexpensive
Overreliance on Past-Cycle Models
Traditional frameworks designed for the past don't capture AI dynamics
Treating AI Like a Theme
Rather than a restructuring cycle
Where perception and reality diverge; opportunity emerges.

38

Why Timing Matters
In infrastructure cycles, returns accrue before the economic impact is fully visible.
1
Inferencing >> Training
Market size differential creates long-term opportunity
2
Infrastructure Takes Years
Build phase creates multi-year visibility
3
Demand Exceeds Supply
Across compute, power, and data centers
4
Revenue Lags Capex
Creating multi-year earnings uplift
5
Step-Function Reactions
Price reactions occur in jumps, not smooth curves

39

The Guiding Principle
"Technology is no longer a sector; it is the mechanism through which every sector creates and unlocks value."
— Noel Misthopoulos
This is the lens through which the next decade of investing must be viewed.

40

PerformanceΣDX

41

The PerformanceΣDX Thesis
PerformanceΣDX is a structural map of where AI creates economic value across the global economy.
Each letter represents a sector undergoing AI-driven transformation. Capital rotates long or short based on intrinsic value, earnings power, and disruption risk.

This is a blueprint for allocating into the AI economy.

42

P — Platforms (SaaS, Cloud, Ecosystems)
Platforms capture data, scale through software, and monetize through subscriptions.
Characteristics
  • Deep moats via data lock-in and switching costs
  • High-margin, recurring subscription revenue
  • Operating leverage expands EPS as usage scales
  • Mission-critical integration across enterprise workflows
Examples
  • Microsoft (Azure, GitHub, Copilot)
  • Google (Gemini ecosystem)
  • Meta
  • Amazon
  • Shopify
Why it matters: The global economy is being platformized; platforms are the toll roads of digital activity.

43

E — Energy (Nuclear, Natural Gas)
Power is the hidden bottleneck, and energy providers become indirect beneficiaries of AI.
Natural Gas
Transition bridge to nuclear with regulatory compliance
Nuclear Power
Baseload power for data centers requiring uninterrupted energy
AI buildout = explosive electricity demand + need for stable baseload power. Power availability becomes a constraint for AI infrastructure. State-capitalism and industrial policy accelerate investment.

44

R — Retail
Retail is now a data platform, in addition to being a store, using AI to optimize every operational layer.
Real-Time Inventory
SKU-level analytics and automated replenishment
Dynamic Pricing
Precision-pricing and personalized recommendations
Automated Logistics
Supply-chain routing and optimization
Margin Expansion
Significant cost efficiencies through AI
Retail winners are the companies with the best data loops—not the most stores.

45

F — Fintech Solutions
Fintech continues to displace legacy banking through data, software, and lower cost to serve.
Real-Time Credit Risk Scoring
Instant onboarding and credit decisions
Platform-Based Deposits
Lower infrastructure and compliance costs
Banking Disruption
Structural challenge to traditional models
The future of banking is software, not branches.

46

O — Omniverse Technologies (AI + Spatial Computing)
The Omniverse blends AI, simulation, and spatial computing to create autonomous digital environments.

Use Cases
Simulation replaces trial-and-error, reducing cost, risk, and time to market.

47

R — Real Estate (AI Factories)
Data Centers
Compute space and power-adjacent real estate
Network Infrastructure
Fiber, towers, and connectivity backbone

Real estate is transforming into compute space, driven by AI infrastructure demands.

48

M — Media & Entertainment
AI is reshaping how content is produced, personalized, distributed, and monetized.
Lower Production Costs
Automated editing, localization, and rendering
Personalized Feeds
Dynamic advertising and content recommendations
Enhanced IP Monetization
AI-driven analytics optimize content value
Entertainment becomes a data-driven, algorithmically optimized engine.

49

A — Artificial Intelligence
AI is the core engine of the buildout—from enterprise deployment to agentic automation.
LLM/SLM/QLM Evolution
Advancing model architectures
Agentic AI
Replacing task-based workflows
Enterprise Automation
Deployment across every vertical
Margin Expansion
From labor displacement
AI is not a sector—it is infrastructure for the modern economy.

50

N — Nanotechnologies
Nanotech underpins semiconductors, advanced materials, and next-gen compute.
Advanced Semiconductor Fabrication
Enabling smaller, more powerful chips
Materials Engineering
At atomic scale for superior performance
Quantum Computing Foundations
Building blocks for next-generation compute
Precision Medicine
Biotech manufacturing at molecular level
Nanotechnology is the physical foundation of the AI revolution.

51

C — Cybersecurity & Defense
As AI accelerates automation, digital attack surfaces expand exponentially.
AI-Powered Threat Detection
Real-time anomaly identification using machine learning models
Cloud & Endpoint Security
Protecting distributed infrastructure and identity
National Security
Infrastructure protection becomes non-discretionary
Cybersecurity becomes non-discretionary as attack surfaces multiply.

52

E — EVs & eVTOLs
Transportation becomes software-defined, powered by AI and real-time data.
Urban & Air Mobility
Autonomous aviation and eVTOLs
Battery Optimization
Performance and longevity
Predictive Maintenance
AI-driven vehicle health monitoring
Fleet Automation
Real-time routing and optimization
Mobility evolves from vehicles to intelligent networks.

53

Σ — Semiconductors & Agentic Software (Core Enablers)
The most critical layer. Semiconductors and software determine the velocity of global AI adoption.
GPUs, TPUs, AI Accelerators
Power compute at unprecedented scale
Software Enables Monetization
Automation and application layer
Hardware Bottlenecks
Create pricing power for suppliers
Entire AI Stack Depends on Semiconductor capacity
The constraint is physical, not financial
Semiconductors are the new oil. Software is the new factory. Together, they define economic velocity.

54

DX — Data eXchange Layer
DX is the real-time intelligence layer connecting all sectors of the AI economy.
Rapid Data Sharing
Cross-referencing at machine speed
Dataset Vectorization
Private enterprise data optimization
Enterprise Inferencing
At scale across organizations
Predictive Analytics
Risk detection and scenario modeling
Data exchange is the "nervous system" of the new economy.

55

Trade Mechanics & Selection

56

Security Selection Process
Every position must combine value, visibility, liquidity, and asymmetry.
1
Opportunity Identification
Identifying market inefficiencies
2
Thesis & Trade Setup
Formulating investment hypothesis
3
Catalyst Identification
Pinpointing market-moving events
4
Alternative Data Integration
Leveraging unconventional insights
5
Liquidity Analysis
Evaluating market depth
6
Timing Analysis
Optimal entry and exit points
7
Fundamental Valuation
Analyzing intrinsic value
8
Capital Structure Positioning
Assessing risk and reward
9
Real-Time Monitoring
Tracking market dynamics
Selection begins with information and ends with asymmetry.

57

Key Considerations
We size trades based on what a security is worth across multiple forward scenarios.
Signals Considered
  • Margin expansion potential
  • Ability to maintain and control pricing power
  • Evolution of cost of capital
  • Regulatory shifts and tailwinds
  • Capex-to-revenue conversion
  • Competitive moat durability
We act only when market price intersects intrinsic value.

58

Capital Structure Signal Framework
Capital structure reveals information that the market often misprices. Balance sheets tell the truth before the income statement does.
1
Who Gets Paid
And in what order
2
Survivability
How long a company can last
3
Refinancing Benefits
Who wins from deleveraging
4
Equity Optionality
When equity becomes valuable
5
Mispriced Credit
Where value hides
6
Hidden Equity Upside
Buried beneath negative sentiment

59

Long & Short Playbooks
Longs capture structural value; shorts protect capital and generate alpha.
Longs Target
  • Mispriced secular growth
  • Earnings inflections
  • Capex → revenue lag periods
  • Margin acceleration
  • Under-monetized platforms
  • Emerging monopolies
Shorts Target
  • Legacy incumbents losing share
  • Companies priced for perfection
  • Unsustainable capital structures
  • Inferior technology paths
  • Margin compression
  • Regulatory vulnerability

60

Discipline & Performance Attribution
Exits are rule-based, never emotional. Attribution ensures discipline; outperformance is engineered, not hoped for.
Exit Triggers
  • Intrinsic value reached
  • Thesis violation
  • Accelerated risk signals
  • Catalyst priced in
  • Better use of capital emerges
Performance Attribution
  • Sector contribution
  • Strategy bucket
  • Trade duration
  • Alpha vs beta contribution

61

Risk Framework

62

Risk Management Philosophy
Risk isn't volatility; risk is permanent impairment.
How We Define Risk
  • Liquidity failure
  • Capital impairment
  • Thesis violation
  • Regulatory or macro shock
  • Mispricing across capital structure
Core Approach
Survivability and capital preservation come first. Returns come from disciplined asymmetry, not constant exposure.
A 25-year career in high yield teaches one principle: you live to fight another day.

63

Allocating Into a Once-in-a-Generation Buildout
The opportunity cost of under-allocation is increasing as capital is deployed into the largest infrastructure buildout in modern history, reshaping global economic activity.
1
Largest Economic Cycle
AI buildout is the largest since post-WW2
2
Finite Window
3–7 year opportunity window
3
Structural Advantage
Most investors are structurally late
4
Credit + Tech Expertise
Unique combination of skillsets
5
Intrinsic Value Discipline
Reduces unrewarded risk
6
High-Yield Mindset
Applied to AI equities creates asymmetric trades
7
Bond Market Intuition
Superior timing on rates, spreads, dislocations
"Technology is no longer a sector; it is the mechanism through which every sector creates and unlocks value."
— Noel Misthopoulos

64

Glossary of Key Terms
Agentic AI / Agentic Software — AI systems capable of autonomous decision-making and task execution without continuous human supervision. Represents "Version 3.0" of AI deployment.
AI Factory — Large-scale data center infrastructure purpose-built for AI training and inferencing workloads.
BIS (Business Intelligence Services & Systems) — Enterprise platforms and applications that leverage AI and real-time data to drive operational decision-making.
Digital Twin — A virtual replica of a physical system used for simulation, testing, and optimization.
DX (Data eXchange) — Real-time information flow across systems, markets, and enterprises; the intelligence layer of the strategy.
eVTOL (Electric Vertical Take-Off and Landing) — Aircraft designed for urban air mobility, powered by electric propulsion.
Inferencing — The process by which a trained AI model applies its learning to new data to generate outputs, predictions, or decisions. Distinct from training.
LLM (Large Language Model) — AI models trained on vast text datasets to generate and understand natural language (e.g., GPT, Gemini, Claude).
LQM (Large Quantitative Model) — AI models designed for quantitative reasoning, financial modeling, and data-driven decision-making.
Omniverse — A platform for building and operating real-time 3D simulation environments, blending AI with spatial computing.
PerformanceΣDX — The sector-mapping framework of the strategy, where each letter represents a sector undergoing AI-driven transformation (Platforms, Energy, Retail, Fintech, Omniverse, Real Estate, Media, AI, Nanotech, Cybersecurity, EVs + ΣDX).
Platformization — The shift of traditional industries toward platform-based business models with network effects, data capture, and recurring revenue.
Σ (Sigma) — Semiconductors; the foundation of accelerated computing and the core hardware enabler of AI.
SLM (Small Language Model) — Smaller, more efficient AI models optimized for specific tasks or edge deployment.

65

NSO Capital
  • contact@nso-capital.com
  • www.nso-capital.com
  • +203-904-4442
Asset Management | New York | Paris
Where data and disciplined capital analysis drive performance.
© 2026 NSO Capital. All Rights Reserved.
This material is provided for informational purposes only and does not constitute an offer or solicitation to invest.

66