Zoom: The “Anti-Fragile” Asymmetrical Bet?

“The IT department of every company is going to be the HR department of AI agents in the future. Those digital employees are going to work with our biological ones, and that’s going to be the shape of our company in the future.” — Jensen Huang, CEO of NVIDIA (CES 2026)

If the Godfather of AI is right, the future of work isn’t just better video calls—it’s managing a hybrid human + digital workforce. While the market wrote Zoom off years ago, the company has quietly repositioned itself as the natural “HR Department” for those agents.

Investors are still psychologically scarred. Mention Zoom and the ticker probably triggers 2021 PTSD. Just the mention of the ticker probably feels like a personal attack. Even Cathie Wood fully liquidated her position in late 2023, but what if her investing thesis was correct, just early? While the market is staring in the rear-view mirror, they are missing a fortress balance sheet and a hidden AI stake that could soon rival the company’s entire current valuation.

The Cash Fortress (The Valuation Floor)

Forget the hype. Let’s look at the math. This is where the margin of safety lives:

Market Cap: ~$23 billion
Cash & Marketable Securities: $7.8 billion (Zero debt)
Enterprise Value: ~$15.2 billion
Free Cash Flow (FY26): $1.9 billion

You are buying a premium SaaS platform—a global brand with 140k+ enterprise seats and sticky workflows—at ~7.5x FCF. For context, boring hardware companies and legacy retailers trade at higher multiples. The market is pricing Zoom for a slow death, but the cash flow says it’s thriving.

This is the definition of “Dirt Cheap.” Even without a major catalyst, the downside is protected by a mountain of cash and a business that produces liquidity.

The Anthropic Windfall: A “free” Home Run

In 2023, Zoom quietly invested $51 million in Anthropic at a $4 billion valuation. Fast forward to February 12, 2026: Anthropic closed a $30 billion Series G at a $380 billion post-money valuation.

Analysts peg Zoom’s stake at $2.5–$4.5 billion today. If Anthropic IPOs at the rumored $750B+ range this year:

The Math: Zoom’s stake could hit $10 billion+.
The Proxy Play: Amazon owns more of Anthropic, but because AMZN is a $2.5T behemoth, the stake only moves their needle 3-5%.
The Impact: For a giant like Amazon, that’s a rounding error. For Zoom, it’s nearly 50% of its current market cap. When you buy Zoom today, you aren’t just buying a software company; you’re buying a massive, liquid stake in the leading “reasoning” AI- for almost nothing.

The Gen Z Factor & The “War Chest”

As Gen Z enters the C-suite, the “Microsoft-only” era is fading. 2026 data show that Gen Z and Millennials (who now make up over 60% of the workforce) prefer the low-friction and video-first nature of Zoom. They want a “System of Action,” not a “System of Record.”

The Competitive Advantage (vs. Salesforce/CRM): CRM is a giant, but it has virtually no cash compared to Zoom’s hoard. Zoom is lean, founder-led, and has a $7.8B war chest to acquire high-growth AI startups to force that return to 15% revenue growth.

The “Anti-Microsoft” Pivot: The Orchestrator of AI

The bear case is simple: “Microsoft Teams is free, so Zoom is dead.” But the data shows a “David vs. Goliath” moment is happening in the Enterprise:

Enterprise Revenue: Grew 7.1% last year (Q4 FY26), triple the rate of the small-biz segment.
The Fortune 10 Win: Zoom recently displaced Cisco in a 140,000-seat deal. Why? Because giants are tired of “Microsoft Lock-in.

The Federated AI Approach: While Microsoft locks you into OpenAI, Zoom’s AI Companion 3.0 is a “conductor.” It switches between Anthropic (Claude), OpenAI (GPT-5), and its own Small Language Models in 0.1 seconds. It’s “Best-of-Breed” vs. “Whatever Microsoft bundles.”

Metric	Salesforce (CRM)	Zoom (ZM)	Why This Favors Zoom
Market Cap	~$179.7B	~$25.3B	Lower Bar: Easier for a $25B company to 2x than a $180B giant.
Net Cash	$8.4B	$7.8B	Acquisition Firepower: Zoom’s cash is 30% of its market cap. CRM’s is only ~4%.
Enterprise Seats	~150,000 Companies	~220,000 Companies	Upsell Runway: Zoom has more “doors” to walk through to sell Phone/AI/Contact Center.
AI/Agent Displacement	High Risk	Lower Risk	Salesforce relies on “human” seats (Sales/Support). Zoom’s video/phone is “infrastructure.”
Debt Load	$7.6B	$0	Flexibility: Zoom has zero “interest rate heartburn.”

Why Zoom Wins the “Agent” War Yuan Saw First

While Salesforce’s Marc Benioff spent 2025 screaming about “Agentforce,” Eric Yuan was already there in June 2024. In his Decoder podcast interview nearly two years ago, Yuan laid out a “prophetic” vision of “Digital Twins.” He envisioned AI agents that don’t just summarize meetings but attend them for you, negotiate contracts on your behalf, and make decisions based on your specific “First Principles.”

At the time, the market laughed it off as sci-fi. Today, it is the north star for the entire industry. Zoom wins here because:

The System of Action: AI agents don’t want to navigate 20 layers of a legacy CRM. They want to “join the meeting,” take notes, and execute tasks. Zoom is where the work actually happens.
The Switzerland of Tech: Zoom doesn’t force you into one LLM. Your digital twin can pick the best engine without vendor lock-in.
Zoom Phone just crossed 10 million paid seats. It’s the “wedge” that’s breaking open massive platform deals.

Rare Asymmetry with Minimal Capital Risk

This is a very clean, asymmetrical setup in tech. Zoom doesn’t need Microsoft or Salesforce to implode. It doesn’t need another pandemic. It just needs a founder-led company with a history of delivering during uncertain times to keep executing and achieving a very attainable 10-15% growth rate. Even a 20-30% growth, though wishful, is not delusional.

The Risks? This thesis will be tested. The primary risk here isn’t the total loss of capital—at 7.5x FCF, the floor is remarkably solid. Instead, the risk is opportunity cost. We don’t know how long it will take for the market to wake up and re-rate this “pandemic relic” into a “2026 AI powerhouse.”

In my view, the risk of losing principal is surprisingly low, but the patience required to see the thesis materialize is high.

Disclosure: I am not bearish on Microsoft; in fact, I own shares in MSFT. I am not bearish on CRM (Salesforce), though I currently own no shares. I simply believe Zoom is a mispriced anomaly hiding in plain sight.

The Duolingo “Lizard Brain” Liquidation: Why Retail Is Wrong Again

This is exactly why the average retail investor consistently underperforms the S&P 500: they are biologically wired to buy high and sell low.

We are currently witnessing a classic “Retail Doom Loop” with Duolingo (DUOL). If you’re staring at the price ticker instead of the fundamentals, you’ve already lost the game.

Your Biology Is Your Worst Enemy

The “lizard brain” (the amygdala) is a marvel of evolution designed to keep you alive, not to make you wealthy.

If your tribe started running 50,000 years ago, you didn’t pause to ask, “Is that a lion or just a technical correction?” You ran. In the wild, that is survival. In the markets, that is financial self-destruction.

The Tribal Sell: The stock drops 20% in a week → your lizard brain screams, “The tribe is fleeing! There must be a predator (AI) I can’t see!”
The Narrative Pivot: Loss aversion (we feel the sting of losses roughly 2x more than the joy of gains) forces us to rewrite history. We don’t say “the price is lower.” We say “the company is failing.” It’s emotional pain relief dressed up as “analysis.”

The Retail Loop of Doom

Retail investors rarely buy value; they buy social proof. Here is how the loop destroys portfolios:

The Validation Phase: Retail ignores the stock while it’s quietly forming a base. They only jump in after a +40% rally because “the price action proves it’s good.”
Conviction Evaporation: Because the thesis was built on a green line moving up and not fundamentals, the very first red candle triggers panic.
The Narrative Flip:
- Price Up: “Luis von Ahn is a genius; the owl is the future of AI.”
- Price Down: “The app is a fad; AI is disrupting them; the CEO is distracted.”

This is reflexivity at its most toxic: price shapes mood, mood shapes selling, and selling drives the price lower until the entire tribe has fled the cave.

Prisoners of the Moment

The current wave of hate toward Duolingo isn’t analysis—it’s emotional venting from people who bought near the $500+ peak in May 2025 and are now feeling the burn. They are prisoners of their own portfolio pain, not the reality of the business.

If you’re “hating” the company today, ask yourself: Were you investing, or were you gambling with leverage and money you couldn’t afford to lose? Your anger isn’t with the green owl. It’s with your own risk management.

The Irrelevance of Anchoring

“I bought at $500, so I need it to get back there to be ‘right.’”

Your entry price is 100% irrelevant. The past transaction cannot be undone. All that matters is what the business is worth today versus its future potential. If you overpaid for a great house in a bidding war, the house didn’t suddenly become “bad”—you just paid a premium for a premium asset.

The market is currently offering Duolingo shares at a 70–80% discount to the peak. I’m betting this is normal growing pains, not an irreparable decline. No great company avoids “off years.”

I very much could be wrong but my cost basis and holding period is unique to me. I am comfortable with the risk, and that’s all that matters.

Pivot from Strength: The Agility Advantage

Ignore the stock price for a minute. Duolingo is acting from a position of extreme strength. Compared to the “supertankers” of digital media, Duolingo is a nimble speedboat.

Company	Market Cap	Employees	The “Agility” Factor
Duolingo	~$4.7B	~850	Speedboat: Can pivot the entire roadmap in a month.
Spotify	~$106B	~7,300	Tanker: Massive scale, but harder to maneuver.
Netflix	~$406B	~16,000	Supertanker: Incredible reach, but heavy overhead.

From first principles, it is infinitely easier for Duolingo to pivot its roadmap to AI-native learning than for these behemoths to turn their ships. Focusing on user growth over immediate monetization is the correct long-term move, even if it causes short-term “earnings heartburn.”

Planting vs. Harvesting

Wall Street is throwing a tantrum because Duolingo is “foregoing” more than $50 million in bookings to remove ad friction.

The contrarian view? They aren’t losing money: they’re reinvesting in distribution. They are betting that a network effect of 100 million daily active users (DAUs) will be worth far more than squeezing a few extra bucks out of frustrated free users today. They are trading short-term harvesting for long-term dominance.

The AI “Barrier to Success”

In a world where everyone has GPT-4, the winner isn’t the one with the “best” AI, it’s the one with distribution, brand, and proprietary data. Duolingo has billions of data points on how humans learn and a owl brand that is a global cultural icon. They aren’t being disrupted; they are the disruptors. They are using their $1.04 billion cash pile and a new $400 million buyback to repurchase shares while the lizard-brain crowd hands them over at a massive discount.

Final Thoughts:

I’ll concede this: if you signed up for a quiet, slow-and-steady compounder, DUOL isn’t for you. Luis von Ahn is in “Full Founder Mode,” and that makes some shareholders uncomfortable.

But if you have an appetite for volatility and believe in the long game, this is a prime investment candidate for a Roth IRA. The potential gains are astronomical—and entirely tax-free.

Stop confusing a high-volatility stock with a risky company. They are not the same thing. If they hit 100M DAUs by 2028, buying DUOL near a $4.7B valuation today will look a lot like buying Netflix in 2011. Forget your entry price and the stock price. Focus on the compounder.

Duolingo: The Fake Sentiment Crisis

The market currently treats Duolingo like a “dead app walking,” terrified that AI will turn language learning into a relic. They are playing the math of a utility company while ignoring the psychology and network effects of a platform. The argument that Duolingo is headed to zero is dumb. The argument that AI language tools make learning a new language obsolete is even dumber.

Part 1: The Blackjack Blunder (Panicking Over a 6)

In Blackjack, if the dealer is showing a 6—the statistically weakest card in the deck—and you are sitting on a Soft 18 or a 7, you don’t surrender. You double down.

The Table Panic: Right now, the market is “surrendering” because they are convinced the dealer has a King or an Ace hidden under the cards. They are sweating over a ghost.
The Reality: The dealer (the bears) has a weak hand built on vague fears about AI. Meanwhile, your hand is a 40% revenue grower with $1.1 billion in net cash and zero debt. The market is surrendering a winning hand because it’s spooked by a bust card. When a stock is down 70% from its high, but the business is still growing revenue at 41%, the market is no longer pricing the business on facts—it’s pricing it on fear.

Part 2: Dirt Cheap—The Valuation Gap

The quants are missing the forest for the trees. Let’s look at the actual damage:

The “Gift” Multiple: Duolingo is trading at an EV/Sales of ~4.2x. For a 40% grower, that is an absolute anomaly. Most software companies growing this fast trade at 8x to 12x revenue.
The Cash Rebate: With $22 per share in net cash, you’re buying a world-class growth engine at a “distressed asset” price.

Part 3: The Platform Moat (Robinhood, Spotify, and Bad Bunny)

Comparing Duolingo to titans like Meta or Netflix might be a stretch in scale, but the logic is identical: platforms that own the “habit” win.

Engagement is a Moat: In Q3 2025, Duolingo crossed the massive milestone of 50 million Daily Active Users (DAUs), growing 36% YoY. People aren’t just downloading the app; they are addicted to it. A translation tool is a utility; Duolingo is a habit.
The Spotify/Robinhood Parallel: Remember 2022? The “smart money” said Spotify was dead because of Apple Music. They said Robinhood was a “zero” because of Fidelity. They were wrong. They ignored the UX and the habit. Same with edtech rivals like Babbel or Rosetta Stone. Duolingo’s network effects (social sharing, leaderboards) make it the iPhone of apps: sticky in a way pure utilities aren’t. AI might commoditize translation, but it can’t commoditize community. Duolingo has 100M+ total users sharing streaks; that’s a flywheel no chatbot can spin alone.
The 35% Spike: Look at the “Bad Bunny Effect.” Following his Super Bowl LX appearance, Duolingo saw a 35% week-over-week spike in Spanish learners. People don’t just want to “understand” culture; they want to participate in it.

Part 4: The “Utility Fallacy” (Why AI Can’t Kill Art)

The argument that AI translation makes learning a language obsolete is the single dumbest take in modern tech. It treats language as a “data transfer” problem, ignoring the soul of human behavior.

Language = Social Status: Treating language like a pure utility is an enormous misreading of human nature. Just like food is more than just fuel, and the clothes you wear go beyond just functionality, language is a signal of effort, intelligence, and respect.
Soul Can’t Be Translated: People don’t listen to Bad Bunny because they need to translate his lyrics into a technical document. They listen because they like the music, the rhythm, and the slang. You don’t learn a language to “get the data”—you learn it to inhabit the soul of the culture.
The Attraction Factor: A survey found that 79% of adults find bilingualism more attractive than monolingualism. Using a translator app to impress someone you’re attracted to has zero “oomph.” It’s the difference between a “fancy restaurant” and “buying ingredients at the store.”
The Restaurant Logic: Saying AI kills language apps is like saying people will stop eating at restaurants because groceries are cheaper. We don’t eat out for the calories; we eat out for the experience and the connection.

Let me be clear: Those arguing that AI will turn language into a utility are engaging in one of the silliest “intellectual” debates I have ever heard. It is shockingly ridiculous. It assumes humans are robots that only care about the shortest path from Point A to Point B. In the real world, we take the scenic route because that’s where the value is. And here’s the kicker, bears forget: Duolingo is AI. They’re building the moat higher while rivals play catch-up. Their Duolingo Max tier (powered by GPT-4) rolled out AI-driven “Video Call” and “Roleplay” features last year, letting users practice real conversations with an AI that adapts on the fly. It’s not replacing the habit; it’s supercharging it.

Part 5: The 5-10x Opportunity (The Bottom Line)

Investing is most rewarding during difficult times when sentiment is at its lowest. Even if you don’t believe Duolingo returns to its peak 45x revenue multiples, this is an opportunity worth taking. With earnings dropping on February 26 (just two weeks out), hint at it as the “next card flip.”

Asymmetric Risk: the downside is protected by a billion-dollar fortress.
The Path to $1,000: For Duolingo to reach $1,000, it needs to become a $47 billion company. In the tech world, that’s just a “successful mid-cap”—the size of Workday or Lululemon. Given its AI-powered “Personal Tutor” narrative, this is a very plausible outcome over the next 5-10 years.
I started a position last week. An opportunity based on a SaaS apocalypse, which feels irrational given that Duolingo is not even a SaaS company. I will look like a visionary or an investor in complete denial.

The Verdict: You’re being offered a “Royal Flush” opportunity. By the time the math proves Duolingo is a $50 billion company, the stock will already be at all-time highs. Betting against the Owl is betting against human nature. I’ll take the other side of that trade every time.

Invest in Myself

artificial-intelligence