Big Tech AI Spending Hits $650 Billion. Amazon’s $200 Billion Bet Just Spooked Wall Street.

Big tech AI spending just spooked Wall Street like nothing since the DeepSeek panic. Amazon, Alphabet, Microsoft, and Meta revealed plans to spend a combined $650 billion on AI infrastructure this year, and the market responded by wiping over $1 trillion in combined value. Amazon’s stock dropped roughly 7% on February 6 after the company announced $200 billion in planned 2026 capital expenditures, more than $50 billion above what analysts expected. Investors are now asking whether the AI boom justifies these enormous bets or whether Big Tech is throwing money at a bubble.

Thursday evening, Amazon dropped their Q4 earnings report. Revenue was up 14% to $213.4 billion. AWS backlog grew 40% year over year to $244 billion. By any normal standard, those are great numbers. But then came the capex figure. $200 billion. For 2026 alone. And the market didn’t care about the revenue anymore.

Why Amazon Got Hit Hardest by Big Tech AI Spending Fears

Amazon wasn’t the only company announcing massive AI spending. Alphabet said it plans to spend $175 billion to $185 billion in 2026. Meta bumped its guidance to between $115 billion and $135 billion, almost double what it spent in 2025. Microsoft’s run rate puts it on pace for about $145 billion.

So why did Amazon take the biggest punch?

A few reasons. First, the surprise factor. Analysts expected Amazon’s capex to land around $146.6 billion according to FactSet. Amazon showed up with $200 billion. That $53 billion gap between expectation and reality is what killed the stock, not the spending itself.

Second, Amazon was already the weakest Magnificent Seven performer heading into 2026. The stock was down about 4% for the year through Thursday’s close. Investors had less patience for another spending surprise from a stock that wasn’t delivering returns.

Third, Amazon also narrowly missed earnings estimates. The company posted $1.95 per share versus the $1.97 consensus. A two cent miss sounds tiny, but when you’re simultaneously telling investors you plan to burn $200 billion, every penny counts.

D.A. Davidson actually downgraded Amazon’s stock to neutral from buy on Friday, citing concerns about spending plans, risks to AWS dominance, and the chance that AI could actually erode Amazon’s retail business. That last point is interesting. The very technology Amazon is spending billions on could end up helping competitors eat into its e-commerce market share.

$650 Billion in Big Tech AI Spending. The Number That Changed Everything.

Step back and look at the full picture. Amazon, Alphabet, Microsoft, and Meta are collectively planning to spend close to $650 billion to $700 billion on capex in 2026. That’s a 67% to 74% increase from the $381 billion these same companies spent in 2025. About 75% of that goes straight to AI infrastructure. Chips, servers, data centers.

Michael Field, chief equity strategist at Morningstar, put it bluntly to CNBC. “The bet is becoming binary. Either a big payoff if these investments come good, or a huge waste of shareholder’s cash if it goes wrong.”

That quote sticks with me because it captures exactly what’s going through every retail investor’s mind right now. These companies are making bets so large that there’s barely a middle ground anymore.

We’ve Seen This Movie Before (Sort Of)

If you were around for the DeepSeek panic in January 2025, this week probably felt familiar. Back then, the Chinese AI startup released a model that supposedly cost just $5.6 million to train compared to OpenAI’s $100 million plus for GPT 4. Nvidia lost $589 billion in a single day, the largest one day market cap loss in U.S. stock market history.

The market recovered from DeepSeek. Nvidia bounced back. The AI spending narrative continued. (We covered how the AI boom echoes past market bubbles when OpenAI’s own numbers raised eyebrows.)

This time feels different. DeepSeek was an external shock, a surprise from outside the system. This February 2026 sell off is coming from inside the house. The companies themselves are telling you they need to spend more money than anyone predicted, and the market is responding by saying “prove it was worth it.”

Is This an AI Bubble or Just Growing Pains?

Depends who you ask.

Sam Altman, CEO of OpenAI, said in 2025 that he believes an AI bubble is happening. Jamie Dimon at JPMorgan said he thinks “AI is real” but some money invested now will be wasted, and there’s a higher chance of a “meaningful drop” in stocks than the market was pricing in.

On the other side, Morgan Stanley analysts argued the bubble fear is “misplaced” or “premature.” Their reasoning? The biggest tech companies today have roughly three times the cash flow and capital reserves compared to companies during the dot com bubble. They can afford to spend this kind of money. Whether they should is a separate question.

I think the honest answer is probably somewhere in between. AI is real. The technology works. Companies are using it, but the current level of spending assumes AI will generate returns at a scale and speed that nothing in tech history supports. We’re in a period where the spending is certain and the payoff is a guess. (I wrote a deeper breakdown of whether the AI bubble is real back when this debate was just getting started.)

What This Means If You Own Big Tech Stocks

Here’s where I want to be practical. If you have money in an S&P 500 index fund, you’re exposed to this. In late 2025, the five largest companies made up 30% of the S&P 500 and 20% of the global MSCI World index. That’s the greatest concentration in half a century.

One oddly reassuring detail is the equal weight S&P 500 actually hit a record high during this same week. The rest of the economy is doing fine. This sell off is a tech story, not an economy story.

If your portfolio is diversified, you’re probably okay. If you’re heavily concentrated in Magnificent Seven stocks, this is a good week to think about whether that concentration still makes sense.

Nobody Has a Crystal Ball (Including Me)

I’m not going to tell you whether to buy this dip or run for the exits. I don’t know. What I do know is that $650 billion in spending needs to produce results, and 2026 is the year the market will start demanding receipts.

The AI trade has shifted from “buy anything with AI in the name” to “show me the revenue.” That’s probably healthy, even if it hurts in the short term. If you want to see what experts predicted for AI this year before this sell off happened, we covered 15 AI predictions for 2026 backed by real research data.

Frequently Asked Questions

Why did Big Tech stocks lose $1 trillion in February 2026?

Amazon, Alphabet, Microsoft, and Meta all announced massive increases in AI infrastructure spending for 2026, with combined capex guidance approaching $650 billion to $700 billion. Amazon’s $200 billion projection was $53 billion above analyst expectations, and its stock dropped about 7% in a single day. Investors worried that the spending may not generate enough returns to justify the enormous costs.

Is the AI stock bubble going to burst?

Opinions are split. OpenAI CEO Sam Altman has said he believes an AI bubble exists, while Morgan Stanley analysts argue the fear is premature because today’s tech giants have triple the cash reserves of dot com era companies. The answer depends on whether AI spending produces measurable revenue growth in 2026 and 2027.

Should I sell my tech stocks during this sell off?

That depends on your timeline and concentration. The broader economy remains strong (the equal weight S&P 500 hit a record high during this same week), so this correction is tech specific. If you’re diversified and investing long term, sharp sell offs in strong companies have historically been buying opportunities, but if your portfolio is heavily concentrated in a few tech names, consider whether that level of risk still fits your situation.

How does Amazon’s $200 billion AI spending compare to other tech companies?

Amazon’s $200 billion capex plan is the largest among Big Tech. Alphabet guided $175 billion to $185 billion, Microsoft is running at about $145 billion annually, and Meta expects $115 billion to $135 billion. Combined, these four companies plan to spend roughly $650 billion on infrastructure in 2026, with about 75% going to AI.