Understanding Economic Bubbles

In 1636, Dutch people paid $25,000 for a single tulip. 150x a craftsman's yearly salary. It seemed like prices could only go up — until Tulips' value fell 90% in 3 days. The story behind "the Bitcoin of the 1600s": The Dutch Golden Age was booming. Global trade routes, financial innovation, and agricultural advancements made them unstoppable. Until something seemingly insignificant happened: In 1593, the first tulip bulb was planted in the Netherlands. These exotic flowers from Central Asia quickly became a status symbol among the elite. Gardens filled with tulips became symbols of power and wealth. Until something remarkable happened: A virus infected some tulips, creating vivid flame-like streaks of color. These "broken tulips" were impossible to replicate. The rarest variety? The Semper Augustus - white petals with dramatic red streaks. Here's where the chaos starts: In 1624, a single Semper Augustus bulb sold for 1,000 guilders. By 1633? 5,500 guilders. By 1637? 10,000 guilders. For perspective: A skilled craftsman earned 300 guilders per year. People started getting reckless: A futures market emerged in taverns called "colleges." People traded tulip contracts without ever seeing the actual bulbs. No money down. No collateral needed. Just promises to pay astronomical sums for flowers they'd never touched. February 3, 1637: A routine tulip auction in Haarlem. 70 bulbs were up for sale. Not a single bulb sold. Three forces collided to trigger the crash: • Spring bulbs emerged, revealing massive supply • No new buyers willing to pay astronomical prices • Traders realized profits only existed on paper Traders had bought bulbs not to plant them, but hoping to sell them to someone else at a higher price. When they ran out of "greater fools" willing to buy, the entire system imploded. It was about human psychology: FOMO drove people to speculate. Limited supply created artificial scarcity. And social pressure made rational people act irrationally. Sound familiar? The same patterns repeat in every bubble: • 1720: South Sea Company • 1990s: Dot-com • 2008: Housing • 2021: Crypto Different assets, same human behavior. Here's the crucial lesson: The tulip markets of 1637 echo eerily in today's digital landscape. Information overload. Endless distractions. Viral trends that spike and fade. In this environment, attention has become the scarcest resource. It's impossible not to get caught up in the FOMO. The antidote? History shows us that during bubbles, the people who thrive are rarely the speculators. They're the ones building something authentic amid the chaos. The ones who stay true to a pure vision. This principle remains unchanged in our digital world: Clear signals outperform loud noise. Authentic expertise outshines viral trends. Short-term trends & FOMO are what's causing increasing distrust in today's world. But a consistent brand creates clarity in confusion. Trust amid skepticism.

Day 3 of 3 for what the Dot Com Bubble can teach us about the AI one. There's a joke or a rule of thumb - depending on who you ask - in commercial real estate that says a new building needs to go bankrupt twice before it turns a profit. The person who builds it loses money, the next guy does, then finally the third person to pick it up for a fraction of the cost can actually get a return out of it. Something similar happened in the Dot Com era. All the money going into infrastructure built out fiber optic networks and Internet backbone well in excess of what anyone could figure out a business case for. One interesting story was 360networks. They had a method for laying cable from rail cars along rail right of way. They hit a $13B market cap on $234M revenue and filed for bankruptcy 7 months later. Companies like that built the modern Internet infrastructure then sold it for pennies on the dollar. This stuff got so cheap that I used to watch Boise State football games when I was in college in the early 2000's. A Boise company streamed the video from the Jumbotron along with the radio announcers... for free. It was insanity. The explosion of e-commerce, video streaming, and telecom from 2001-2011 was built on the carcass of the Dot Com Boom. People keep ranting about how there can never be enough revenue from AI to pay for the tens of billions of dollars being spent on Nvidia chips and data centers. I just laugh. First, you never know what people are going to be able to charge for. I attached a figure from a recent WSJ article showing rocket launches in Florida. (Note that this graphic is terrible because the 2024 line is only half the year so it is still increasing exponentially.) I attended the ISDC private space conference in 2006 when it was just weirdos and talks on the space elevator. I heard a solid talk about how the Space Shuttle proved reusability didn't make sense because the launch volume was so far below where it broke even to do it. And when Musk started, we all KNEW there wasn't enough launch demand to pencil out, either. But look at that chart. Supply does sometimes create its own demand. Even if the AI data centers getting built today don't make money, that's not a bad thing for the rest of us that aren't paying for them. Because we'll have the chance to buy up that capacity for pennies on the dollar. Just like JibJab, College Humor, Strong Bad, and a startup called YouTube couldn't have afforded to build the infrastructure they needed to operate, so will there be a zillion new ideas built on buying Nvidia chips by the pound. The current AI moment itself was started using excess capacity left over from the crypto mining mania. Don't worry that you can't imagine how anyone will ever turn a profit after buying all those Nvidia cards. Maybe they won't. But the price of AI compute is going to drop fast. Anything that's marginal today will be able to turn a profit after the bust. That's the biggest lesson from 2000.

I had coffee ☕️ on Friday with friend Tommy H. of the Fed. We caught up on building and regs but the convo turned quickly to Tommy’s passion for the intersection of economics and neuroscience. I was blown away. Over the last few days I have found myself reading up on human behavior and I thought what better phenomenon to look at than memecoins. Coins like Dogecoin or tokens inspired by celebrities or political movements, have seen massive swings in valuation. While some analysts view these assets as speculative fads, their persistence reveals deeper insights into human behavior. By unpacking the neural and psychological roots of risk-taking, reward-seeking, and social validation, we can gain a better understanding of why memecoins thrive and what risks they present. TIL that at the core of memecoin trading is dopamine-driven behavior. Brain imaging studies have shown that anticipation of reward, rather than the reward itself, triggers spikes in dopamine. Memecoins capitalize on this by: ✔️ Offering the illusion of asymmetric upside (i.e., the next Dogecoin) ✔️ Rewarding rapid price movements with instant feedback loops (price jumps = dopamine hits) ✔️ Reinforcing behavior with intermittent, unpredictable rewards — a hallmark of operant conditioning, similar to gambling Human decision-making is profoundly social. The mirror neuron system, which allows us to simulate and internalize the actions of others, makes us especially vulnerable to herd behavior. In the context of crypto: ✔️ Viral memes, influencers, and FOMO amplify perceived value ✔️ Platforms like Reddit, TikTok, and X create echo chambers that validate risky choices ✔️ Social validation — likes, retweets, influencer endorsements — activates reward pathways akin to those triggered by monetary gain Behavioral neuroscience intersects with cognitive psychology in explaining how biases distort rational decision-making: ✔️ Availability bias: Traders overvalue tokens with frequent media exposure ✔️ Loss aversion: Investors hold losing positions longer to avoid realizing losses ✔️ Overconfidence bias: Retail traders overestimate their skill, especially in bull markets ✔️ Recency bias: Recent pump-and-dump wins encourage riskier bets, despite long-term loss rates These biases are amplified in high-volatility, high-narrative markets like those driven by memes Understanding the neural basis of speculative behavior can inform smarter regulation: ✔️ Disclosure rules may not work if investors are acting on emotion and reward-seeking, not logic ✔️ Real-time nudges (e.g., warnings about volatility or concentration) could leverage behavioral insights ✔️ Policy responses should acknowledge that memecoin trading is part financial decision, part neurological compulsion The memecoin phenomenon is a revealing lens into the intersection of money, mind, and modern digital culture. What a cool coffee!

Had a great time speaking at Reuters Next this week on the topic of whether or not we're in an AI bubble. Some of my takes: 1. AI will revolutionize bottom lines even more than toplines. Our portfolio company Klarna improved revenue per employee by 73% just last year--and as AI transforms functional areas such as sales, product/eng, marketing, and recruiting, this will become the new norm. (As a result, companies might raise less money, but VCs will benefit from investments that get valued off their rich EBITDA multiples). 2. Market corrections follow when hype outpaces value creation. Despite its recent resurgence, the crypto bubble was a real bubble because of the lack of widespread application--we'd all struggle to name 20 crypto companies with $20M in ARR. But AI is different. Companies like Abridge, Cursor, Glean, Perplexity, Hebbia, HeyGen, Cohere, Harvey, ElevenLabs, Synethesia, not to mention OpenAI and Anthropic are not only buzzworthy; they're generating cold, hard cash in the tens or hundreds of millions. Perhaps the internet bubble is the most appropriate comparison--while initial exuberance may create some degree of correction, the underlying technology *will* reshape the global economic landscape (P.S. while NVIDIA trades at 40x forward earnings, Cisco peaked at over 100x in the dotcom bubble) 3. Much like a medieval castle, dead bodies pile up when you don’t have a moat. And thanks to AI-assisted code generation, tech or "integrations" as a moat won't cut it for application software today. We've seen AI companies differentiate from others building on the same foundational models via vertical specificity, economies of scale, human-centered workflows, brand, and adjacent services. Such a joy to do this with Krystal Hu and George Sivulka (who I first met when he was living in a literal closet, complete with hangers) cc Northzone

How the GenAI Bubble is and is not like the Dotcom Bubble In late 1996, Fed chairman Alan Greenspan warned that the frenzied investment in any startup that had a .com in its name was “irrational exuberance.” But excitement was such that the warning, even coming from such a creditable source, was ignored. Between 1995 and March 2000, investments in the NASDAQ rose by 800% - But the bubble bust and by October 2002 all those gains were lost. When ChatGPT was released in November 2022 people were dazzled by its ability to generate text that appeared indistinguishable from human writing. It appeared a great advance in AI had been achieved and a new AI millennium was proclaimed. Intelligent machines would replace expensive human workers resulting in unprecedented productivity, boosting corporate profits sky high. Since then, we have seen a rally that has added almost $16 trillion to the S&P 500. But today creditable sources like Jim Covello, Head of Global Equity Research at Goldman Sachs, are warning that we are again in an investment bubble. He said: “Most technology transitions in history, particularly the ones that have been transformational, have seen us replace very expensive solutions with very cheap solutions, potentially replacing jobs with tremendously costly technology is basically the polar opposite.” That LLMs are vastly expensive, and from an energy standpoint even unsustainable, is now obvious. But can they deliver on the touted productivity gains? If not, it is not for lack of trying. Research by The Upwork Research Institute reveals that 39% of C-suite leaders are mandating the use of genAI tools, with an additional 46% encouraging their use. But the results are not encouraging, the same study found that nearly half (47%) of employees using genAI say they have no idea how to achieve the productivity gains their employers expect, and 77% say these tools have actually decreased their productivity and added to their workload. The Internet (eventually) actually did “replace very expensive solutions with very cheap solutions” but the dot-com startup investment bubble was irrational because there were no barriers to entry. After the bust, most failed right out of the gate and it would take decades for the real winners to emerge. LLMs startups also have no barriers to entry, the technology will always be vastly expensive, and in the end, it just doesn’t deliver. When this bubble bursts it could be ugly indeed. There may be no long-term winners, at least not big ones, this time around.

Behavioral Science in Financial Decision-Making Behavioral science provides a fascinating lens to understand how psychological factors influence financial decision-making. Traditional finance theories assume that individuals act rationally, weighing risks and rewards to maximize profits. However, behavioral finance reveals that human behavior often deviates from this ideal, shaped by cognitive biases and emotions. The attached study on Behavioral Finance Biases in Investment Decision-Making highlights how these biases affect our financial choices: Prospect Theory and Loss Aversion People experience losses more intensely than equivalent gains, leading to risk-averse behavior in gains but risk-seeking in losses. For example, an investor might hold onto losing stocks longer, hoping for a rebound, rather than cutting losses. Herding Behavior Fear of missing out and the belief that others have better information often lead investors to follow the crowd. This can inflate market bubbles or deepen crashes, as individual decisions become influenced by group behavior rather than analysis. Overconfidence Bias Many investors overestimate their knowledge and decision-making skills, leading to excessive trading and ignoring risks. This bias is particularly prevalent in new investors, who may overtrade or ignore market fundamentals. Anchoring Bias Investors tend to fixate on initial information—such as a stock's historical price—regardless of its relevance to current decisions. This can result in suboptimal choices, like refusing to sell an overvalued stock due to its past performance. Why This Matters: Understanding these biases is crucial for investors, financial advisors, and policymakers. It helps them design strategies to mitigate irrational behaviors, such as: Promoting financial education to recognize biases. Implementing default options like auto-enrollment in retirement savings plans. Leveraging framing techniques to encourage better decision-making, such as presenting options in terms of long-term outcomes rather than short-term fluctuations. Behavioral finance highlights a vital truth: financial decisions are not just about numbers—they’re deeply human. By applying behavioral insights, we can create systems that guide individuals toward better financial outcomes. Have you noticed biases in financial decision-making, either personally or professionally? I’ve noticed that I go with the herd for a while and then distrust herd-like behavior. I turn to experts to see whether they question the ongoing herd-like behavior and tend to then favor the experts. I’m not sure, however, what makes me question the herd behavior after a while. Is it my own cautiousness or some thing else? #BehavioralScience #BehavioralFinance #DecisionMaking #Psychology #FinancialInclusion #BehaviorChange #BehavioralInsights #Finance #InvestmentTips

Billions Burned on Fluency Without Understanding-The Coming Tragedy of the AI Economy The AI whiplash currently building, is validating what one researcher and critic, Gary Marcus has been saying for years, exposing the unraveling of an AI economy built more on hype than substance. Here are the multitude of ways the hype has far surpassed reality: 1. Inflated Valuations, Hollow Returns Despite billions poured in, the AI industry has delivered little beyond dazzling predictions. Market caps have ballooned far beyond real revenue streams. Critics have long warned, the economic fundamentals never matched the feverish valuations—making today’s backlash feel inevitable. 2. Failures Piled High From the underwhelming launch of GPT-5 (supposed arrival of AGI) to multiple Gen AI projects failing, the cracks are becoming undeniable. AI’s hype machine promised miracles; the real world is served mirages. 3. Dreams of AGI—Now Deferred Leaders who once predicted imminent artificial general intelligence are backpedaling. Sam Altman is saying that the term AGI "isn't helpful" and admitting that there "may" be a bubble? Even Eric Schmidt concedes uncertainty. The consensus didn’t just fade—it imploded under the weight of reality. 4. A Bubble Built on Illusion The tragedy isn’t just economic—it’s human. We anthropomorphized machines, convincing ourselves they “think” or “learn.” This collective lack of critical analysis of the technology created a mirage economy, where speculation thrived on illusions of intelligence. 5. Misplaced Fortunes, Risky Bets The capital poured into data centers and AI infrastructure has surpassed the excesses of past bubbles. Economists warn of unsustainable burn rates and asset-heavy investments with little grounding in true productivity gains. History teaches: industries bet big on infrastructure—then a single innovation flips the game. Will AI be different? 6. A Cycle History Has Seen Before Financial historians remind us that bubbles often precede breakthroughs. Yet here, the tragedy cuts deeper: we’ve mistaken statistical pattern-matching for cognition—and invested as if it were civilization’s next tulip bulb/railroad /dot-com website. Why it’s tragic: The AI economy isn’t just a bubble—it’s also a mis-allocation of talent, capital, and imagination. Instead of building systems with genuine reasoning and judgment, we’ve glorified autocomplete and called it intelligence. Billions are being burned chasing illusions, while the harder, slower path to real AI remains underfunded. The hype machine may have sparked the fire, but the collapse will leave behind the ashes of billions of dollars incinerated—alongside a sobering reminder of how easily humanity can be tricked into mistaking fluency for intelligence. #AI #AIEconomy #AIBubble #AIHype #AIBacklash #AIRealityCheck #AICollapse #AIWhiplash #AITragedy #AIOverhyped #AIInvestmentCrash #AIillusion #AIvsReality #AIReckoning #HouseOfCards #AIMarketBubble #HistoryRepeats

The current wave of artificial intelligence (AI) innovation brings to mind the lessons of the dot-com bubble. Back then, hype and capital fueled a flood of startups, but many chased ideas without clear value or sustainable business models. Take Pets.com, which gained massive attention but collapsed under the weight of unproven demand and unsustainable spending. Similarly, Webvan promised to disrupt grocery delivery but failed due to over-expansion and poor operational execution. In contrast, companies like Amazon and eBay thrived because they solved real problems and built models that could scale. Today, the AI space is showing similar signs. Many startups are focusing on use cases that lack clear value or trying to solve problems that don’t truly exist. While the potential of AI is undeniable, the rush to innovate often sacrifices practicality and impact. Innovation for its own sake is not enough. If AI startups are to avoid the fate of many dot-com casualties, they need to focus on creating solutions that address genuine, pressing needs. The winners in this era will be those who build sustainable models, deliver measurable value, and stay grounded in the realities of their markets. Let’s remember: technology isn’t about looking futuristic—it’s about making a meaningful difference. Are we learning from the past, or are we destined to repeat it? #artificialintelligence #technology #startups #strategy #entrepreneurship

A Bubble Within A Bubble? Bubble 1) AI software solutions being churned out at a record pace Funding is still surging. - VC put $49.2B into gen-AI in the first half of 2025; - Software/AI now ≈ 45% of all VC...with the first half 2025 already beating 2024’s total. Company count keeps exploding. - 10,095 AI startups across top countries - - 5,509 in the U.S. alone... But production value lags. - 71% of firms use gen-AI somewhere, yet >80% report no material EBIT impact; value remains localized to specific units. - Separately, 69% of leaders say most AI projects never make it to live ops. Innovation cadence is slowing. - Major model upgrades are arriving later with less value...good for stability, but it lengthens payback for fast follower apps. Take: - Lots of apps + cheap capital + thin differentiation + slow enterprise rollout = heightened software froth risk. - The market will reward proofs of ROI and punish “feature wrappers.” Bubble 2) Is AI infrastructure “overbuilt” and slow to deploy? Investment tsunami: - Big Tech’s AI build-out: Amazon, Alphabet, Microsoft, Meta together are pacing $344–$364B of 2025 spend tied largely to AI infrastructure, with 2024–26 totals near $1T. Massive build queue: - 10 GW of new DC projects expected to break ground in 2025; 7 GW to complete ≈ $170B in assets. - - 156 GW of large-load data centers sit in the queue. Data Center Vacancy is ultra-low & pre-leased. - North American vacancy ~2.3–2.6%; - 73% of new builds pre-leased - This may be a good sign but only if demand remains at levels never seen before in any other industry for this extended time frame. Analysts still flag risk. - Moody’s and IEEFA warn localized overbuild/stranded-asset Take: There’s execution risk, not classic oversupply...power and interconnect delays slow monetization, keeping many sites capacity-tight. Overbuild risk is highly regional, where forecasts outrun real demand. AI is the most profound technological revolution in human history...but if we let hysteria and unchecked capitalism drive the agenda, we risk turning it into a mid-term boondoggle. The long-term impact will still be world-changing, but only deliberate, disciplined action ensures it delivers on its promise. #ai #economy #money