AI’s Biggest Winners📉, Stop… | The VC Corner Summary

1. Key Themes

AI Is Most Valuable in Low-Margin Industries, Not High-Tech Ones

The conventional wisdom is that AI benefits knowledge workers and software companies most. The article flips this: low-margin industries are the biggest winners because efficiency gains flow directly to the bottom line.

"Low margin industries gain the most from AI because small operating savings flow directly into profit. The highest returns come from removing scheduling and coordination work inside existing systems, not adding new software employees must learn." — Daniel Kornum

The "Self-Driving Company" Is Already Here

AI agents are no longer a future concept — companies like Replit are running full internal operations through them today, replacing external software tools in the process.

"Replit uses internal agents across engineering, sales, marketing, support, and operations while keeping quality stable. Teams increasingly define goals while agents execute tasks, replacing several external software tools along the way." — Amjad Masad

VC Is Shifting From Markup Games to Real Capital Structures

The era of riding valuation multiples as a VC strategy is ending. The next generation of investors must understand infrastructure, energy, and manufacturing finance — not just software.

"Software investing is shifting as valuation growth alone no longer guarantees strong long term outcomes. Future investors will compete by designing capital structures for infrastructure, energy, chips, and manufacturing projects." — Hemant Taneja

Physical AI and Defense Tech Are the Fastest-Growing Investment Categories

Capital is flooding into robotics and defense at a pace that dwarfs prior years, signaling a major sectoral rotation away from pure software.

"Physical AI funding in the first half of 2026 surpassed the total invested across all of 2025. Large warehouse automation deals accelerated, with Sequoia leading investment activity and NVIDIA expanding its safety initiatives."

"Defense tech venture funding reached $35.4 billion year to date, led by Anduril's $5 billion Series H round. Exit value climbed to $1.8 trillion, driven primarily by SpaceX's public market listing."

AI Agents Must Be Managed Like Employees — and That's the Next Big Opportunity

The article frames AI agent management as an emerging discipline with its own operational rigor, and the companies that codify this will win.

"Managing AI agents requires the same discipline as managing people through context, evaluation, and clear objectives. The next major opportunity belongs to firms that encode business processes into reliable agent systems rather than selling generic tools." — George Sivulka

2. Contrarian Perspectives

The Biggest AI Winners Have the Lowest Margins

Against the consensus that AI supercharges high-margin software businesses, the argument here is the opposite: thin-margin industries (logistics, healthcare ops, food service) benefit most because every dollar saved goes straight to profit, and AI removes coordination work within existing workflows rather than adding new tooling.

"Low margin industries gain the most from AI because small operating savings flow directly into profit. The highest returns come from removing scheduling and coordination work inside existing systems, not adding new software employees must learn."

AI Shifts the Knowledge Advantage Against the Buyer

Most narratives frame AI as a tool that empowers buyers and users. The article surfaces a counterintuitive risk: buyers who use commercial AI models may actually be surrendering proprietary data and eroding their own informational edge.

"AI shifts the balance of knowledge as buyers reveal proprietary data while using commercial models. Companies need control over data, evaluations, orchestration, costs, and continuous learning to protect long term advantages." — Satya Nadella

Your Investor List Is a Liability, Not an Asset

Founders treat large investor databases as a volume advantage. The article argues the opposite — most names are irrelevant and the strategy should be radical qualification and focus, not broader outreach.

"Large investor databases rarely produce results because most names are not actively relevant to a fundraising round. Success comes from qualifying, scoring, and ranking a focused shortlist instead of sending more cold outreach."

3. Companies Identified

Replit

Description: AI-powered coding and development platform
Why mentioned: Live case study of a company operating as a near-autonomous "self-driving" organization using internal AI agents across every function
Quote: "Replit uses internal agents across engineering, sales, marketing, support, and operations while keeping quality stable. Teams increasingly define goals while agents execute tasks, replacing several external software tools along the way."

Thinking Machines

Description: AI company focused on human-centered, decentralized AI deployment
Why mentioned: Represents a philosophical counterposition to centralized AI — argues AI should augment local expertise, not replace it
Quote: "Thinking Machines argues AI should strengthen human judgment instead of replacing local expertise. Its strategy focuses on customizable models, better interfaces, and decentralized alignment shaped by each organization."

Anduril

Description: Defense technology company
Why mentioned: Anchor data point in the defense tech investment surge; closed a $5B Series H round
Quote: "Defense tech venture funding reached $35.4 billion year to date, led by Anduril's $5 billion Series H round."

Fireworks

Description: Enterprise AI infrastructure platform
Why mentioned: Largest deal of the week — raised $1.5B Series D at a $17B valuation
Quote: "Fireworks raised $1.5B in Series D funding at a $17B valuation to scale its enterprise AI infrastructure platform."

Neko Health

Description: AI-powered preventive healthcare platform
Why mentioned: Notable $700M Series C raise, signaling strong investor appetite for AI in healthcare
Quote: "Neko Health raised $700M in Series C funding to expand its AI-powered preventive healthcare platform globally."

Chai Discovery

Description: AI-driven drug discovery company
Why mentioned: $400M Series C signals continued momentum in AI biotech
Quote: "Chai Discovery raised $400M in Series C funding to advance AI-driven drug discovery."

Walden Robotics / LimX Dynamics

Description: Next-generation and humanoid robotics companies
Why mentioned: Both raised major rounds ($300M and $200M respectively), representing the physical AI investment wave
Quote: "Physical AI funding in the first half of 2026 surpassed the total invested across all of 2025."

Vanta

Description: Compliance automation platform (SOC 2, security certifications)
Why mentioned: Sponsor; highlighted as used by 16,000+ companies including Ramp, Cursor, and Harvey
Quote: "Used by over 16,000 companies like Ramp, Cursor, and Harvey, Vanta helps you get audit-ready quickly — and stay that way."

SpaceX

Description: Aerospace and space transportation company
Why mentioned: Primary driver of the $1.8T defense tech exit value figure via its public market listing
Quote: "Exit value climbed to $1.8 trillion, driven primarily by SpaceX's public market listing."

4. People Identified

Daniel Kornum

Description: Author/analyst covering AI's economic impact
Why mentioned: Source of the insight that low-margin industries are AI's biggest winners
Quote: "Low margin industries gain the most from AI because small operating savings flow directly into profit."

Satya Nadella

Description: CEO of Microsoft
Why mentioned: Cited for the "Reverse Information Paradox" — the risk that AI usage exposes buyer data and erodes competitive advantage
Quote: "AI shifts the balance of knowledge as buyers reveal proprietary data while using commercial models."

Hemant Taneja

Description: Managing Director at General Catalyst
Why mentioned: Argues VC must evolve beyond software markup games into complex capital structuring for hard assets
Quote: "Future investors will compete by designing capital structures for infrastructure, energy, chips, and manufacturing projects."

George Sivulka

Description: AI researcher/entrepreneur (likely founder context)
Why mentioned: Articulates the framework for treating AI agents with the same management rigor as human employees
Quote: "The next major opportunity belongs to firms that encode business processes into reliable agent systems rather than selling generic tools."

Amjad Masad

Description: CEO of Replit
Why mentioned: Real-world operator demonstrating what a fully agent-driven company looks like in practice
Quote: "Teams increasingly define goals while agents execute tasks, replacing several external software tools along the way."

Ruben Dominguez

Description: Author of The VC Corner newsletter
Why mentioned: Curator and author of this edition
Quote: "Another week, another pulse check on Tech."

5. Operating Insights

Build AI Agents in Stages — Prove Reliability Before Automating

The tactical framework for deploying AI agents is sequential, not simultaneous. Operators should resist the urge to automate everything at once and instead validate each layer before proceeding.

"The guide starts with manual workflows, then local operators, and finally scheduled automation after each stage proves reliable. Financial, legal, medical, and other high risk decisions should remain under human review before any action is taken."

Encode Your Business Processes Into Agent Systems — Don't Buy Generic Tools

The competitive moat isn't in which AI tool you use; it's in how deeply you embed your specific workflows into agent systems. Generic AI tools are commodities; proprietary process encoding is defensible.

"The next major opportunity belongs to firms that encode business processes into reliable agent systems rather than selling generic tools." — George Sivulka

Control Your Data Stack or Lose Your Competitive Edge

For any company using commercial AI models, data governance is now a strategic priority — not an IT concern. Ceding data control to third-party models is ceding long-term advantage.

"Companies need control over data, evaluations, orchestration, costs, and continuous learning to protect long term advantages." — Satya Nadella

6. Overlooked Insights

Deep Tech Is Now a $156.6B Asset Class — Quietly Rivaling Mainstream VC

While AI dominates headlines, the 2026 Deep Tech Report reveals a massive and underreported capital pool. This creates fertile ground for cross-sector investment theses that combine AI with hard science.

"It estimates Deep Tech startups raised a combined $156.6 billion across both regions [U.S. and Europe]." — Drumbeat Capital / Dealroom.co

AI Can Decentralize Alignment — Not Just Capability

Thinking Machines' strategy is quietly significant: rather than deploying one-size-fits-all AI, they advocate for AI alignment shaped by each individual organization. This is a distinct and underexplored design philosophy that could become a major enterprise differentiator.

"Its strategy focuses on customizable models, better interfaces, and decentralized alignment shaped by each organization."