Big Tech's AI bill is now bigger than its cash flow. What it means for your work

Skip the finance: Big Tech is betting more money than it makes on AI, so it will push AI into everything you use and look hard for ways to get paid back.

How much is Big Tech spending on AI?

More than ever, and faster than the money comes in.

Alphabet reported its third quarter on October 28, 2026. It spent US$51.4 billion on capital expenditures, or capex, the money companies put into long-lasting things like data centers, servers and chips. Its free cash flow, the cash left over after running the business and paying for all that equipment, came out at negative US$9.1 billion. In plain terms, Google's owner spent more than its businesses brought in as cash during the quarter, even with revenue up 25% to US$127.6 billion.

Meta, which reported a week earlier on October 21, showed the same pattern. Capex was US$34.9 billion and free cash flow was negative US$2.1 billion, while revenue grew 26% to US$64.6 billion.

Microsoft also reported on Wednesday. Revenue rose 19% to US$92.3 billion, and Microsoft Cloud grew 30% to US$63.9 billion. On the earnings call, Microsoft's CFO said capex should be about US$170 billion, according to a post by @TheTranscript_, which publishes quotes from earnings calls. Microsoft gives its outlook on the call, so that number doesn't appear in its written release.

These are some of the most profitable companies on Earth, and they are choosing to run cash-negative quarters to build AI capacity. They aren't short of money. They're spending it as fast as they can on one bet.

Why are tech companies spending more than they make?

Because they think the demand is real, and some of the numbers back them up. Google Cloud, which rents computing power and AI models to other businesses, grew 103% to US$30.7 billion last quarter. Google CEO Sundar Pichai said Google's own AI models now process nearly 32 billion tokens a minute (tokens are the small chunks of text AI is measured and billed by).

The catch is timing. A data center costs money the day the concrete is poured, and it pays off only once people and businesses use, and pay for, what runs inside it. That gap is why free cash flow went negative.

It's also why these companies need AI to start earning its keep in the products you already use. Google is now putting ads inside AI Mode, its chat-style search, and taking those AI Mode ads to international markets for the first time. Google's chief business officer said on the call that AI Mode currently earns less per search than classic Search, which helps explain the push to put ads there.

When a company spends more than it earns on something, it needs you to use that something.

Is AI spending costing jobs at big tech companies?

It depends on the company, and the numbers don't tell one clean story.

Meta reported 66,775 employees at the end of September, 15% fewer than a year earlier. Meta reported the number without tying it to AI, and the drop follows layoffs earlier this year. Spending and cuts happening at the same time doesn't prove one caused the other.

Alphabet went the other way. It had 204,757 employees, up from 190,167 a year earlier, so it is still hiring while it spends.

Outside the giants, Dropbox said Wednesday it is cutting about 200 roles, roughly 9% of its staff, while saying the business is profitable. Dropbox gave no AI reason for the cuts.

For the people affected, none of this is abstract. A cut role means a lost paycheck, health coverage and plans. What the numbers do show is that companies are making very deliberate choices about where money goes, and right now a growing share is going to buildings and chips. Fuller national data on October layoffs and hiring isn't out yet. It arrives in early November.

Will AI tools get more expensive?

Nobody has announced price hikes because of this spending, so treat this as an expectation. Still, companies with a bill this size have three levers: sell more AI to businesses, show more ads, and charge more for premium plans. You're already seeing the first two. Ads are moving into AI answers, and AI features keep landing in apps by default, often switched on before you ask.

For a solo creator or small business, that cuts both ways. Competition between Google, Microsoft, Meta and others keeps strong AI cheap or free today, which is great while it lasts. The free tier is also where these companies try to make you a habit. Know which tools you'd actually pay for if the free version shrank.

What to do this week

• List every AI tool you use and what you pay for each. Mark the ones you'd keep if prices went up.
• Running Google Ads? Check your campaign reports for new AI Mode placements, and read the news post on Google's AI Mode ads.
• Pick one AI skill tied to your actual job, like drafting proposals or cleaning up spreadsheets, and practice it for 20 minutes.
• Save a copy of anything important you keep only inside a free AI tool, such as prompts, chats or generated files.
• Work at a big tech company? Keep your résumé and portfolio current. That's good practice in any year.