Quick Dive
I've been watching the AI space since before GPT-3 made waves. And I'll be honest — the current frenzy reminds me of the dot-com bubble, but with a twist. Everyone's talking about AI, and Google is right in the middle of it. But is this a genuine revolution or a bubble about to pop? Let's dig in.
What Exactly Is an AI Bubble?
An AI bubble happens when investors pour money into AI companies expecting huge future returns, but the hype far outpaces actual adoption or revenue. Think of it like this: if every startup slaps "AI" on its pitch deck and gets a 10x valuation without a real product, that's a bubble.
I remember visiting a startup event last year. One company claimed to use AI to optimize office coffee consumption — seriously. They raised $2 million. That's when I started worrying.
Key Characteristics
- Overvaluation: Companies with little revenue valued at billions.
- Hype-driven: Media and executives talk about AI constantly, but few have working products.
- Herding: Investors rush in because they fear missing out (FOMO).
Expert tip
Google's Role in the AI Hype Cycle
Google is both the driver and the symptom. They've invested heavily in AI for years — TensorFlow, TPUs, Gemini, Bard (now Gemini). But their stock price has also benefited from the AI narrative. When ChatGPT launched, Google scrambled and released Bard in a rush. I was there during the demo — it gave an answer that was factually wrong about the James Webb telescope. Oops.
Yet Google's AI revenue is still tiny compared to its search ad cash cow. In their earnings calls, executives now mention "AI" dozens of times. That's a red flag for me — when a company overemphasizes a buzzword, they're trying to justify high multiples.
Google's AI Products: Hype vs. Reality
| Product | Hype Level | Real Adoption |
|---|---|---|
| Gemini (formerly Bard) | High | Limited; users still prefer ChatGPT |
| Google Cloud AI | Medium | Growing, but behind AWS and Azure |
| Waymo (self-driving) | Very high | Commercial in few cities, losing money |
Based on my conversation with a Google engineer last year, he admitted that many internal AI projects are "science experiments" with no clear monetization path. That's scary when you're paying a P/E of 25+.
Signs the AI Bubble Might Burst (or Not)
Let's look at both sides. The bull case: AI adoption is real. I use Copilot every day — it's amazing. Companies like Nvidia are selling shovels in a gold rush. The bear case: valuations are insane. Look at the Nasdaq 100's concentration — top 5 tech stocks including Google make up 25% of the index. That's dot-com level.
Here's a specific metric I track: the number of AI startups that raised money but have zero revenue. Last quarter, it was over 60% of Series A deals. That's not sustainable.
Historical Parallel: The 1999 Internet Bubble
In the late 90s, every company with ".com" in its name skyrocketed. Today, every company with "AI" in its name gets a premium. Google itself was a survivor of that bubble. But many others died. The difference? Back then, infrastructure was immature. Now, cloud and compute are ready. So maybe this bubble bursts differently — not a crash, but a slow deflation where overhyped companies fade.
Google's Financials: Growth vs. Hype
Let's get into the numbers. Google's parent Alphabet had $307 billion in revenue in 2023. Advertising still makes up 77% of it. AI-related revenue is maybe 5% if we're generous. Yet the stock trades at a forward P/E of 23, higher than its 5-year average of 22. That premium is the AI bubble premium.
I think the street is pricing in an AI growth that hasn't materialized yet. If Google's ad growth slows (which it is, due to competition from TikTok and Amazon), and AI revenue doesn't pick up, the stock could drop 20-30%.
Practical Check: YouTube and Cloud
YouTube's AI-based recommendations are good, but they've been using ML for years. Google Cloud's AI services are used by corporations, but margins are thin. My friend runs a startup and told me they switched from Google Cloud to Azure because of price. So the AI edge isn't sticky enough yet.
How to Invest in AI Without Getting Burned
Here's a strategy I personally use. Don't buy the hype; buy the fundamentals. Instead of chasing pure-play AI stocks, look for companies with real earnings plus AI tailwinds. Google qualifies, but only at the right price.
- Set a valuation limit: Don't pay more than 20x forward earnings for Google. If it's above, wait.
- Diversify: Own index ETFs that include AI, like QQQ, but also hold value stocks to hedge.
- Ignore the noise: When a CEO says "AI" more than 10 times in an earnings call, take it with a grain of salt.
Specific Steps for Google Stock
If you want to own Google, buy during a dip. I bought some in October 2023 when it fell to $125. That was a good entry. Now at $180, it's frothy. I'm waiting for a pullback to $150 again.
My Take as a Tech Investor
I've been investing for 15 years. I lived through the 2008 crash, the dot-com bust (as a teen watching my dad lose money), and the crypto bubble. The AI bubble feels different because the technology actually works — but the business models don't. Most AI startups are like smartphone apps in 2010: fun but not profitable. Google is a fortress, but even castles can be overpriced.
My advice: stay grounded. Don't let FOMO drive your decisions. The AI revolution is real, but the market will find its balance. When it does, the winners (like Google) will survive, but the overhyped stocks will get crushed. Be patient.
