AI’s Big Investment Push Has A Smaller Payoff Beyond Tech

🔍 Read the full analysis: AI’s Big Investment Push Has A Smaller Payoff Beyond Tech on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get the latest gadgets 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 recent corporate and productivity data show no clear financial payoff from AI outside technology, even as companies commit hundreds of billions of dollars to computing infrastructure. His October 24 note compares performance at seven large technology companies with the other 493 members of the S&P 500, while the scale of investment raises questions about when businesses and their customers will see returns.

Apollo’s charts show revenue per worker and profit margins rising among the “Magnificent 7”—Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla—while the figures for the other 493 S&P 500 companies are broadly flat. The note also cites San Francisco Federal Reserve productivity measures that, according to Apollo, show no clear AI-related boost to productivity growth. These are economy-wide and company-group trends, not a direct measurement of AI’s effect on each business.

On the spending side, Meta’s third-quarter results reported $34.91 billion in capital expenditures. Its operating cash flow was $32.79 billion, leaving free cash flow at negative $2.12 billion for the quarter, compared with positive $10.6 billion a year earlier. Meta narrowed its 2026 capital-spending forecast to $135 billion to $145 billion.

Meta’s October 22 filing also listed about $355.73 billion in leases that have not started, $454.52 billion in non-cancelable contractual commitments, and as much as $14.72 billion in contingent cloud-capacity purchases. Together, those items total about $825 billion in future commitments, mostly tied to data centers, servers and cloud capacity. The figure is not all AI spending; the filing also includes items such as Reality Labs hardware.

The Gap Between AI Costs and Returns

The contrast matters because AI infrastructure requires substantial spending before many companies can show corresponding revenue or productivity gains. Meta’s commitments are a liability to fund over time, not an immediate cash payment of $825 billion, but they indicate how much capacity the company has contracted for. The investment could support future services and growth; it also increases exposure if demand or returns fall short of expectations.

Financing is becoming part of the debate. The source material reports that SpaceX borrowed $40 billion to buy AI chips and that the cost of insuring debt issued by Meta and cloud-computing provider CoreWeave rose. Those developments do not establish that the companies cannot meet their obligations, but they point to investor attention shifting from access to chips and power toward how rapidly expanding infrastructure will be financed.

For businesses adopting AI, the aggregate results are not a verdict on every use case. They do suggest that purchasing models or computing capacity alone is not proof of a return. Companies need to track whether specific deployments improve output, cut costs or increase sales—and account for the costs of running them.

What Apollo’s Comparison Captures

Apollo’s October 24 “Daily Spark” note, titled “No Signs of AI Payoff Outside the Tech Sector,” draws on several indicators rather than a controlled test of AI adoption. Its company comparison shows the seven largest technology firms pulling ahead of the rest of the S&P 500 in revenue per employee and profit margins. The figures can show where financial performance is concentrated, but they cannot by themselves establish that AI caused the difference.

The note also uses Ramp data on corporate purchases of AI services. In that dataset, the top 10% of firms account for 99.5% of spending on model serving and inference, 99.0% of neocloud spending and 95.6% of spending on OpenAI and Anthropic. These figures describe Ramp’s business customers, not the S&P 500 as a whole. Apollo also reports that 80.4% of the top 1% of AI spenders are venture-backed, compared with 13.1% of the median spender.

Evidence from specific tasks is more mixed than the broad headline suggests. The supplied material cites customer-support research showing 14% more issues resolved per hour, rising to 34% for novice workers, and retail experiments with sales effects ranging from zero to 16.3%, depending on the workflow. It also cites a survey in which 83% of nearly 9,000 executives reported no productivity impact so far, with average productivity growth attributed to AI at 0.2% a year. These results have different methods and scopes and should not be treated as one unified estimate.

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

— Torsten Slok, Apollo chief economist, in the October 24 Daily Spark note

What the Aggregate Data Cannot Show

Apollo’s “zero” framing is stronger than what its charts alone can establish. Flat averages across 493 companies can conceal both gains and losses, and the measures do not isolate AI’s effects from changes in labor costs, interest rates, tariffs or other business conditions. The Ramp spending data cover Ramp customers and should not be read as a full census of corporate AI use.

The source material does not provide a company-by-company accounting of AI costs and returns, nor a consistent measure of gains across the customer-support and retail examples. Meta’s commitments also span years and include spending that is not exclusively for AI. The reported borrowing and debt-insurance costs do not, by themselves, establish that financing will be unavailable or that planned projects will be canceled.

Returns and Funding in Focus

The next test will be whether companies can connect AI deployments to durable improvements in productivity, margins or sales while keeping computing and financing costs under control. Investors and executives will have more evidence as companies report results and update their capital-spending plans. Meta’s 2026 spending outlook and its future infrastructure obligations will remain key measures to watch.

For now, the available evidence supports a narrower conclusion than a blanket claim that AI produces no value: aggregate gains outside tech are not yet clear, while some task-level studies report measurable improvements. The timing and scale of broader returns—and whether those returns will justify the infrastructure commitments—remain unresolved.

Key Questions

What did Apollo mean by “zero financial payoff”?

Apollo chief economist Torsten Slok used the phrase to describe aggregate indicators in which revenue per worker and margins rose for seven large technology firms but were broadly flat for the other 493 S&P 500 companies. It is an interpretation of those trends, not proof that every non-tech company has received no benefit from AI.

What makes up Meta’s roughly $825 billion in commitments?

Meta’s filing lists leases that have not started, contractual commitments and contingent cloud-capacity purchases. The total is about $825 billion, mostly related to data centers, servers and cloud capacity; it is a multiyear commitment figure, not a bill due immediately, and not all of it is AI-related.

Are there examples of AI producing measurable business gains?

Yes. The supplied material cites a customer-support study reporting 14% more issues resolved per hour, with a 34% gain for novice workers, and retail experiments with sales effects from zero to 16.3%, depending on the workflow. Those findings apply to specific tasks and do not establish economy-wide returns.

Does the report show that AI investment is failing?

No. Apollo’s data point to no clear broad payoff outside tech so far, but they do not isolate AI as the cause of company performance and cannot rule out gains at individual firms. The timing, size and distribution of future returns remain uncertain.

Source: ThorstenMeyerAI.com