AI Stocks in 2026: Are We in a Bubble or Just Getting Started?
NVIDIA's P/E is compressing while revenue keeps accelerating. Is the AI trade over, or are we entering the real growth phase? Breaking down the numbers.
Hynexly
The AI Trade Everyone Loves to Hate
Every week I see another headline asking if AI stocks are in a bubble. And honestly? I get it. When a single company (NVIDIA) adds more market cap in a year than the entire GDP of some countries, it does feel bubbly.
But here's what most of these hot-take articles miss: the valuation picture is way more nuanced than "expensive = bubble."
Wait, Valuations Are Getting Cheaper?
This is the part that trips people up. Let me show you:
Look at that chart carefully. NVIDIA's current forward P/E (35x) is actually below its 5-year average (45x). Same story with AMD. Google and Meta are trading near their historical averages.
How is that possible when these stocks are at all-time highs? Because earnings are growing faster than stock prices.
The Revenue Trajectory Is Insane
NVIDIA went from $26 billion quarterly revenue to an estimated $68 billion in two years. That's not normal. That's not even "strong growth." That's the kind of revenue acceleration you see once or twice per decade.
And it's not just NVIDIA:
- Microsoft's Azure AI revenue is growing 60%+ year-over-year
- Google's cloud division turned profitable primarily through AI workloads
- Meta is generating actual ROI from their AI investments in advertising optimization
The Bull Case (That I Mostly Believe)
Here's why I think we're closer to the beginning than the end:
1. Enterprise adoption is just starting. Most Fortune 500 companies are still in the "pilot phase" of AI implementation. The real spending wave — production deployments at scale — hasn't fully hit yet.
2. The infrastructure buildout creates its own demand. Every new AI model requires more compute, which requires more chips, which requires more data centers, which requires more power. This is a self-reinforcing cycle.
3. Sovereign AI is a new demand vector. Countries are now building their own AI infrastructure for national security reasons. This is government-level spending that didn't exist 18 months ago.
The Bear Case (That Keeps Me Honest)
I'm not a mindless AI bull. Here's what worries me:
- ROI pressure is real. At some point, CFOs will demand to see returns on their AI investments. If the productivity gains don't materialize, spending could slow dramatically.
- Concentration risk is extreme. NVIDIA controls 80%+ of the AI chip market. Any competitive threat (AMD, custom chips from Google/Amazon) could compress margins.
- Macro matters. If we get a recession, AI budgets get cut like everything else. "Transformative technology" doesn't make you immune to the business cycle.
My Take
I'm positioned bullish on AI but not blindly so. My approach:
- Core holding in NVIDIA — I think the P/E compression story is the most compelling bull case. Revenue growing 50%+ with a P/E in the 30s is actually reasonable for a company with this level of dominance.
- Diversified AI exposure through MSFT and GOOGL — These are "safer" ways to play the AI theme because they have diversified revenue streams.
- Hedging with options — I'm buying some puts as insurance against a sharp correction. When everyone is bullish, that's exactly when you need protection.
Bottom Line
The AI trade isn't a bubble in the traditional sense — valuations are actually supported by earnings growth. But it IS extremely concentrated, and any crack in the narrative could trigger a sharp selloff. Own these stocks, but size your positions wisely and have a plan for the downside.
Disclaimer: I hold positions in NVDA, MSFT, and GOOGL. This is not financial advice — always do your own research.
