THESIS

The AI Convergence is Here

We’re entering a new era where AI is no longer confined to screens or data centers — it's transforming the physical world, infrastructure systems, and the very foundations of trust. This thesis explores three converging frontiers:

AI × Robotics [→ Sensing & Automation]: Machines that perceive, decide, and act — reshaping how we move, build, and interact with the world.

AI × Infrastructure [→ Systems & Energy]: AI as both a stressor and a savior for energy, compute, and climate — ushering in intelligent infrastructure and sustainable acceleration.

AI × Security [→ Finance & Identity]: In a world of deepfakes and synthetic agents, AI-native defenses and programmable trust are essential for secure communication, commerce, and governance.

Together, these domains define the next decade of opportunity — where embodied intelligence, intelligent infrastructure, and secure identity systems will form the backbone of the AI-native world.

I) AI × Robotics [→ Sensing & Automation]

We are at the dawn of a new industrial era where AI escapes the confines of screens and enters the physical world. The convergence of artificial intelligence, robotics, automation, and sensing is enabling machines to perceive, decide, and act with increasing autonomy and precision. Like the PC and smartphone revolutions before it, this shift is being driven by compounding advances in hardware, neural networks, and real-time perception. As sensors get cheaper, AI models get faster, and interfaces get more intuitive, we are unlocking a wave of embodied intelligence that will transform logistics, manufacturing, infrastructure, space, and human-machine interaction itself.

Why Now: Software Already Ate Software — Now It’s Eating Atoms

The revolution in AI — driven by increasingly efficient neural networks and transformers — is no longer confined to the digital world. We’re entering the next phase: the convergence of AI, robotics, automation, and sensing. This wave will bridge the digital and physical worlds and reshape how we build, move, monitor, and interact with matter itself.

We are witnessing a rapid feedback loop:

• Better sensors → Better AI
• Better AI → Smarter robots
• Smarter robots → More adoption → Demand for better hardware and data

This virtuous cycle mirrors the PC boom of the 1980s and the smartphone explosion of the 2000s — but this time, it’s embodied intelligence. The AI-native robotics moment is upon us.

Three Big Shifts

1. Bridging Digital ⇋ Physical

Machine learning is enabling robots — both physical and digital — to understand, adapt, and act on the world with minimal human intervention. AI is no longer just informing action; it is executing it.

2. Zero Lag: Thought ⇋ Action

Transformers and neural nets are shrinking the delay between intelligence and execution. As decision-making moves closer to the edge, robots can act in near real-time — navigating, adjusting, improving. Automation becomes viable, cost-effective, and scalable.

3. High Bandwidth: Human ⇋ Machine

The interface between humans and machines is evolving. From telepathic wearables to vision-powered copilots, we’re enabling radically faster interaction loops. The result: freeing human intellect for higher-order creativity and problem-solving.

Why This Market Is Exploding

• Hardware costs are plummeting (sensors, compute, robotics platforms)
• AI models are now light enough to run at the edge
• Real-time sensor fusion is viable for industrial autonomy
• New interfaces (e.g., voice, gestures, wearables) are going from science fiction to shipping products

Portfolio Examples

II) AI × Infrastructure [→ Systems & Energy]

As AI scales, infrastructure becomes the bottleneck—and the opportunity. The convergence of AI with energy, compute, and climate systems is creating a new class of intelligent infrastructure. From energy-efficient chips to battery materials to smart grids and developer tools, we’re seeing breakthroughs that enable faster, cheaper, and more sustainable AI deployment. As transformers give way to specialized models and datacenter loads push the grid to its limits, the need for AI-native infrastructure is urgent—and the companies building it will define the next decade of system-level innovation.

Why Now: The Infrastructure Layer of Intelligence

The AI boom is placing unprecedented demands on our physical and digital infrastructure. Energy, compute, hardware, and systems optimization are no longer background considerations — they’re the foundation of progress. The convergence of AI with infrastructure and climate technologies is enabling a new generation of intelligent systems that are faster, cheaper, and more sustainable.

We’re moving beyond large transformer models running in hyperscale data centers. New architectures — both in hardware and software — are emerging to make AI more accessible, efficient, and deployable in real-world systems.

Three Macro Forces Colliding

1. Energy ↔ AI Feedback Loop

AI is both energy-hungry and energy-saving. On one hand, model training and inference demand efficient chips and better data center design. On the other, AI is optimizing everything from battery chemistry to grid flexibility to consumer energy decisions.

2. Infrastructure ↔ Intelligence

The “intelligence layer” is being embedded into core infrastructure: energy systems, hardware supply chains, mobility, and materials. Systems that were once static are now software-defined, AI-tuned, and adaptive in real time.

3. Post-Transformer Architectures

Transformers dominated the 2020s. The 2030s will be about specialized models and architectures tuned to edge workloads, vertical domains, and energy-aware compute. This opens up huge opportunities in new chip designs, ML compilers, and systems optimization.

Why This Market Is Exploding

• AI demand is stressing the grid → new opportunities in smart energy, optimization, and control
• New materials & chemistries needed for batteries, chips, cooling, etc.
• Vertical LLMs & domain-specific models are growing fast
• AI-native infrastructure tools are emerging for developers and operators
• Climate + compute is now a top-tier opportunity set, not a niche

Portfolio Examples

III) AI × Security [→ Finance & Identity]

As AI-generated content floods the digital world, trust becomes the new battleground. Deepfakes, synthetic identities, and autonomous agents are undermining traditional authentication methods, making identity, security, and financial infrastructure more critical than ever. This next wave demands AI-native defenses: real-time fraud detection, privacy-preserving biometrics, and programmable trust layers that can operate at global scale. In this environment, identity is not just a login—it's the cornerstone of every secure interaction, from payments to communication to governance. The companies building these systems will form the backbone of a secure, AI-native internet.

Why Now: When You Can’t Trust What You See or Hear

AI-generated content has reached a point where impersonation — of faces, voices, behavior — is trivial. In this emerging world of deepfakes and autonomous agents, identity is under siege. Traditional trust systems like passwords, 2FA, and even static biometrics are becoming obsolete. We need new primitives for proving personhood, ensuring secure communication, and enabling safe payments — all in real time and at scale.

At the same time, AI is also the most powerful tool for defending against this chaos. From biometric cryptography to intelligent AML systems to programmable finance infrastructure, we are entering an arms race where AI will be used on both sides of the security equation.

Three Tectonic Shifts

1. Deepfakes and the Collapse of Visual/Audio Trust

Anyone can now fake anyone. Real-time avatar generation, voice cloning, and generative video mean that "seeing is believing" is no longer true — especially in financial, legal, and interpersonal communications.

2. AI-Native Attacks Require AI-Native Defenses

Fraudsters are using LLMs, bots, and synthetic identities. Legacy infrastructure — rule-based systems, manual reviews — cannot keep up. The next wave of fintech must be secure-by-design and constantly adapting through machine learning.

3. The Rise of Verifiable Identity and Programmable Trust

From biometric cryptography to decentralized identity wallets, AI is enabling privacy-preserving, real-time, and cross-platform trust layers. These systems will be foundational not just for fintech, but for AI agents, marketplaces, and autonomous systems.

Why This Market Is Exploding

• Deepfake-driven fraud and impersonation are growing exponentially
• AI-native infrastructure is now essential for KYC, AML, and onboarding
• Payments and communication infrastructure are converging with identity
• The future of money requires embedded, real-time trust primitives
• Identity is the new perimeter in a multi-agent world

Portfolio Examples