How Much Is AI Paying Off Outside The Technology Industry?

🔍 Read the full analysis: How Much Is AI Paying Off Outside The Technology Industry? on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get everyday essentials delivered free with Prime

• Fast, free delivery on millions of items
• Prime Video, Amazon Music and more included
• Member-only deals all year

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

Apollo chief economist Torsten Slok says AI has yet to produce a visible financial payoff across corporate America outside technology, citing flat revenue per worker and profit margins among 493 S&P 500 companies compared with gains at seven major tech firms. His October 24 analysis lands as companies commit more money to data centers, chips and cloud services, while research on individual business processes shows that some AI uses can improve performance.

Apollo’s analysis compares the so-called Magnificent Seven — Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla — with the other 493 members of the S&P 500. Its charts put the seven at about $780,000 in inflation-adjusted revenue per employee in the third quarter of 2026, while the other companies were around $525,000 and had remained in a similar range since 2021. The seven’s profit margins were about 30%, compared with roughly 10% for the rest.

Apollo also cited San Francisco Federal Reserve measures of productivity, which it said showed no sign of AI boosting productivity growth. Data from Ramp, a business spending platform with more than 85,000 U.S. customers, show AI-related spending concentrated among a small group: the top 10% of firms accounted for 99.5% of spending on model serving and inference. Those figures describe Ramp customers, not all U.S. businesses.

Other research points to gains in particular tasks. The source report cites a customer-support study in which workers resolved 14% more issues per hour, with a 34% gain for novice workers. Retail experiments ranged from no sales increase to a 16.3% increase, depending on the workflow. By contrast, a survey of nearly 9,000 executives found 83% reported no productivity impact from AI so far; the reported average productivity effect was 0.2% a year.

The Gap Between AI Bills and Returns

The debate matters because businesses are making expensive, long-lived commitments before broad returns show up in standard measures. Meta’s third-quarter results, released October 21, reported $34.91 billion in capital expenditures and negative free cash flow of $2.12 billion, compared with positive free cash flow of $10.6 billion a year earlier. The company forecast capital spending of $135 billion to $145 billion for 2026.

Meta’s October 22 filing listed roughly $825 billion in future commitments, including leases for infrastructure and contracts for cloud capacity, servers and network equipment. The figure is not all AI spending: it also includes other items, such as Reality Labs hardware, and spans commitments over multiple years. It is not equivalent to cash already spent.

For executives and investors, the central issue is whether specific deployments improve revenue, reduce costs or raise output enough to justify infrastructure and service costs. For workers and customers, gains may depend on the task and how a company changes its processes. Neither the aggregate figures nor selected successful trials settle the broader question on their own.

What Apollo’s Figures Compare

Apollo’s October 24 note, titled “No Signs of AI Payoff Outside the Tech Sector,” uses company-level financial comparisons and productivity indicators. Slok’s “zero financial payoff” wording is an interpretation of aggregate trends, not a direct measurement isolating AI’s effect on every company. Revenue per worker and profit margins can be affected by many forces, including labor costs, interest rates and changes in business mix.

The comparison also groups seven technology companies against a much larger and varied set of firms. A flat average among the other 493 could conceal companies that benefit from AI and others that do not. Ramp’s spending data offer a separate view of who is buying AI services, but represent Ramp’s customer base rather than the S&P 500 as a whole.

Spending commitments are rising alongside the debate over returns. The source report says SpaceX borrowed $40 billion to buy AI chips and cites forecasts of nearly $780 billion in 2026 capital spending by Amazon, Alphabet, Meta, Microsoft and Oracle combined. Those figures are not all realized expenditures: the five-company total is a forecast, while Meta’s filing lists contractual commitments and leases with different payment schedules.

“AI has delivered zero financial payoff to corporate America outside of tech.”

— Torsten Slok, Apollo chief economist, as quoted in Apollo’s October 24 note

What the Aggregate Data Cannot Show

The figures do not isolate AI’s contribution from other changes in the economy or within companies. Apollo’s comparison cannot show whether AI caused the gap between the seven technology companies and the other S&P 500 firms, and a group average can mask differences between businesses. The Ramp figures also cannot be treated as a representative measure of all corporate AI spending.

It remains unclear how quickly gains from individual AI deployments will appear in company-wide productivity, revenue or margins, and whether those gains will exceed the cost of computing, software and implementation. The source material provides examples of task-level improvements, but not enough detail to establish whether they persist across companies or translate into lasting profits.

Meta’s reported $825 billion is a total of several categories of future obligations, not an AI-only bill or a single-year cash outlay. The available figures do not establish how much of those commitments will ultimately be used, what returns they will generate, or how financing costs will change. Forecasts for spending by major technology companies are projections, not final expenditure totals.

Tracking Returns Against Spending

The next test will be whether companies report sustained gains in output, revenue or margins as AI tools move from trials into routine use. Investors and corporate managers will also watch spending plans, cash flow and financing conditions as major technology firms build data-center and cloud capacity.

Further company filings and productivity data may help clarify whether current task-level improvements spread across non-tech sectors. For now, the evidence supports a narrower conclusion than Apollo’s headline: no broad payoff is visible in the cited aggregate measures yet, while some specific uses show measurable gains. How widespread and durable those gains become remains unsettled.

Key Questions

What did Apollo mean by saying AI has delivered no payoff outside tech?

Apollo’s note compared revenue per worker and profit margins at seven large technology companies with the other 493 S&P 500 firms. The latter group showed flat readings in the measures cited. The phrase is Slok’s interpretation of those aggregate trends, not proof that no non-tech company has benefited from AI.

Do the figures show that AI has no productivity benefit?

No. Apollo cited productivity measures with no visible AI boost so far, while the source report also describes task-level studies showing gains in customer support and some retail experiments. These findings measure different things and do not establish how widely or how durably the gains apply.

What does Meta’s roughly $825 billion in commitments include?

Meta’s filing includes leases that have not started and non-cancelable contracts, largely for cloud capacity, servers and network equipment, plus contingent cloud purchases. The total covers obligations over time and includes items beyond AI, so it is not an AI-only spending figure or an amount already paid.

Why are companies still spending heavily if returns are unclear?

Companies are investing in computing capacity and AI services in anticipation of future demand and possible business gains. The source material shows large commitments and some process-level benefits, but does not establish whether returns will exceed costs across the wider corporate sector.

Source: ThorstenMeyerAI.com