When Will AI Capex Payback Happen First?

Most of us would likely agree that artificial intelligence benefits are going to take a while to be seen almost anywhere except the financial results of infrastructure providers, who clearly will benefit. Nor would that ever be unusual when an important new technology--not to mention a possible new general-purpose technology--first emerges. 

Indeed, analysts at Goldman Sachs say “leading tech giants, other companies, and utilities to spend an estimated $1 trillion on capex in coming years, including significant investments in data centers, chips, other AI infrastructure, and the power grid.” 

Still, “this spending has little to show for it so far.” Nor would one realistically expect to see quantifiable results so early. The pattern with general-purpose technologies is that the platforms and infrastructure must be built first, before use cases and apps can be developed. 

Also, some functions are more susceptible to generative AI impact, for example, than others. 

Most of us would be willing to concede that customer service is one area where generative AI, for example, should produce results. Functions with many repeatable elements are commonly thought to be susceptible to AI automation. 

In a survey conducted for Bain, enterprise executives reported that better results were seen in sales; software development; marketing; customer service and customer onboarding, for example. Between October 2023 and February 2024, though, most other use cases seemed to produce less favorable outcomes than expected. 

source: Bain 

Generative AI thrives on well-defined patterns and processes, so jobs involving repetitive tasks with clear rules and minimal ambiguity are likely candidates for early change. 

But lots of functions and tasks are not routine or well structured; not simple but complex, so the range of use cases that can benefit near term is arguably limited. 

As the report notes, Daron Acemoglu, Institute Professor at MIT, estimates that only a quarter of AI-exposed tasks will be cost-effective to automate within the next 10 years, implying that AI will impact less than five percent of all tasks.

Most of us would be willing to concede that customer service is one area where generative AI, for example, should produce results. Functions with many repeatable elements are commonly thought to be susceptible to AI automation. Generative AI thrives on well-defined patterns and processes. Jobs involving repetitive tasks with clear rules and minimal ambiguity. 

All that noted, the first quantifiable results will be seen among suppliers of infrastructure, as apps cannot be built until the infrastructure is in place.   

GPT/Possible GPT	Infrastructure Provider	Early Revenue Gains
AI/Large Language Models	NVIDIA	171% year-over-year revenue increase in Q2 2023, driven by demand for AI chips
Internet	Cisco Systems	Revenue grew from $69 million in 1990 to $22.3 billion in 2001 as internet infrastructure expanded
Personal Computers	Intel	Revenue grew from $1.9 billion in 1985 to $33.7 billion in 2000 as PC adoption surged
Electricity	General Electric	Revenue increased from $19 million in 1892 to $1.5 billion in 1929 as electrical infrastructure spread
Railroads	Steel Companies (e.g. Carnegie Steel)	U.S. steel production grew from 68,000 tons in 1870 to 11.4 million tons in 1900

That noted, it also could be said that there has been overinvestment--at some point--in infrastructure for past general-purpose and new technologies. It also might be noted that application and device over-investment also occurs, early in the adoption of a new technology. 

Technology	Time Period	Description of Over-Investment
Railroads	1840s-1850s	Excessive railroad construction and speculation led to financial panics in 1857 and 1873 in the US and UK
Automobiles	1910s-1920s	Hundreds of car companies were founded, with most failing as the industry consolidated
Radio	1920s	Rapid proliferation of radio stations and manufacturers, followed by consolidation
Internet/Dot-com	Late 1990s	Massive speculation in internet-related companies led to the dot-com bubble and crash in 2000
Renewable Energy	2000s-2010s	Over-investment in solar panel manufacturing led to industry shakeout
Cryptocurrencies	2010s-2020s	Speculative frenzy around Bitcoin and other cryptocurrencies

But there is a difference between “over-investment” and the proliferation of would-be competitors in a new market. It always is normal to see more startups in any area of new information technology than there are surviving firms once the market is mature. 

The difference between over-investment and normal competition in a new market can be subtle. What might not be subtle is the lag time between capex investments and revenue realization, for firms not in the "picks and shovels" part of the ecosystem.

Infra suppliers already have profited.

Firm Executives Might Have Strategic Reasons for AI Investment, Even as Analysts Worry About Payback Periods

It is quite understandable that financial analysts covering public firms are concerned about the payback period for various forms of artificial intelligence. After all, huge capital investments in graphics processor units, for example, must be reflected in revenue upside at some point. The issue is whether expectations of near-term return are actually reasonable. 

As always, market forecasters, firm executives and others might lean towards the strategic implications, while financial analysts primarily look at the quarterly performance metrics. 

And, sometimes, investments are more “strategic” than “tactical.” In other words, a telco might have to invest heavily in fiber-to-home facilities simply to stay in business as competitors upgrade their home broadband infrastructures. 

The actual financial return on investment will matter, but might not be the driver. “You get to keep your business” or “you get to stay in business” might be the value, not simply increases in revenue after the investments are made. 

Most new information technologies take some time before we tend to see measurable benefits. That has been true for many technologies. So the issue is whether various forms of AI are more like social media or smartphones or PCs, the internet and automated teller machines. 

Technology	Approximate Lag Time to Measurable Outcomes (Years)
PCs	10-15
Internet	10-15
ATMs	8-10
HDTV	5-8
Cloud Computing	5-7
Wireless/Wi-Fi	5-7
Smartphones	3-5
Social Media	3-5

Applying various forms of AI to various use cases across industries might reasonably produce varied payback periods, from rapid to lengthy, suggesting that investment tied to particular use cases is a reasonable approach.

Most of us likely can imagine clear performance benefits in areas ranging from e-commerce, search and social media recommendations fairly quickly. As AI already is used to support such personalization features. 

Other use cases, including manufacturing or healthcare, might take longer, in part because many parts of the value chain have to be altered at the same time to take advantage of AI. 

Industry/Use Case	Estimated Payback Period
E-commerce (general)	1.2 - 1.6 years
E-commerce (search optimization)	6 months - 1 year
Social Media & Content	1 - 2 years
Manufacturing	2 - 3 years
Business Services (Law, Accounting, Consulting)	1.5 - 2.5 years
Smartphones (AI features)	1 - 2 years
PCs (AI-enhanced software)	2 - 3 years
Cloud Computing (AI services)	1 - 2 years
Healthcare (AI diagnostics)	2 - 4 years
Financial Services (Fraud Detection)	1 - 1.5 years

Obviously there are many variables. Larger-scale implementations may see faster payback due to economies of scale, so long as they are targeting major functions that can affect financial return. 

Some AI applications, such as fraud detection in financial services, may see quicker returns compared to more complex implementations in healthcare or manufacturing, and also be easier to measure. 

Existing information technology infrastructure and past success integrating information technologies, probably also will matter. Companies that have more-developed IT might see faster payback periods compared to firms whose existing infra is less well developed. 

Fast-moving industries such as  e-commerce and social media might realize benefits quicker than more traditional sectors, simply because they face fewer regulatory issues that must first be addressed. 

Regulatory environment: Industries with strict regulations (e.g., healthcare, finance) may have longer payback periods due to compliance requirements.

As always, the particular use cases will have different payback periods, when implemented at scale. 

Industry	AI Use Case	Payback	Source
E-commerce	Product Recommendation Engines	6-12 months	McKinsey, AI in Retail
Social Media & Content	Personalized Content & Ad Targeting	12-18 months	Forrester, AI in Marketing
Manufacturing	Predictive Maintenance	18-24 months	PwC, AI in Manufacturing
Business Services	Legal Document Review & Due Diligence	24-36 months	Accenture, AI in Professional Services
Smartphones	Voice Assistants & Virtual Companions	3-5 years (Long-term brand value)	CB Insights, AI in Mobile
PCs & Cloud Computing	Resource Optimization & Server Management	12-18 months	Bain, AI in Cloud Computing

When Will $1 Trillion in AI Capex Pay Off?

About $1 trillion in expected spending on generative AI capital investment in data centers, chips and servers, the power grid and connectivity might not produce the anticipated benefits in the short term, say Goldman Sachs equity analysts Ashley Rhodes, Jenny Grimberg and Allison Nathan.

But much of the disparity in views about AI is in the timing of benefits, not ultimate value.

The key phrase might be “in the short term.” Among the “pessimists” cited is Daron Acemoglu, MIT Institute professor.

Acemoglu forecasts about a 0.5 percent increase in productivity and about a one percent increase in gross domestic product in the next 10 years, compared with Goldman Sachs estimates of a nine percent increase in productivity and 6.1 percent increase in GDP.

“The forecast differences seem to revolve more around the timing of AI’s economic impacts than the ultimate promise of the technology,” he argues.

And much could hinge on “how” generative AI develops and what it replaces. For example, if GenAI winds up replacing low-wage jobs with costly technology, without producing other value, then the investments might be wasted.

One might argue that the opposite has been the case for some successful technology transitions of the past, including the internet, where relatively low-cost technology replaced costly incumbent solutions.

Seen in that light, a potential problem with generative AI is that it is a costly investment that almost has to displace complex problems to provide value. And that might take time to develop.

Impact might also vary across the ecosystem. Suppliers of “picks and shovels” might profit in the short term even if “gold seekers” do not uniformly benefit.

Also, even if value does not appear in GDP statistics, it still is possible that revenue and profits earned by at least some companies in the AI value chain will show positive changes. Think Nvidia and other graphics processing unit suppliers, or possibly “AI as a service” revenues earned by cloud computing as a service providers such as Amazon Web Services.

As providers of infrastructure, such firms might profit even if others who purchase products and services from infra suppliers do not show revenue or profit gains in the short term.

In other words, there might be infrastructure supplier winners in the short term, even if many other entities make big investments in generative AI that do now show revenue or profit impact in the near term.

And even some who are skeptical about the magnitude of positive impact in the short term might well concede that the long term impact is going to be evident.

By way of perspective, about $5 trillion in information technology investments are made every year, according to researchers at Gartner.


source: Goldman Sachs Global Investment Research

“Generative AI has the potential to fundamentally change the process of scientific discovery, research and development, innovation, new product and material testing an so forth, as well as create new products and

Platforms,” he notes. “But given the focus and architecture of generative AI technology today, these truly transformative changes won’t happen quickly and few—if any—will likely occur within the next 10 years.”

Again, the key phrase might be “today.” Generative AI is expected by some to achieve human-level performance in most technical capabilities by the end of this decade, and compete with the top 25 percent of human performance in all tasks before 2040, according to McKinsey.

If so, both optimists and pessimists have a valid point. In the short term, gains might be muted; in the long term just the opposite could occur.

One study suggests “that around 80 percent of the U.S. workforce could have at least 10 percent of their work tasks affected by the introduction of LLMs (large language models), while approximately 19 percent of workers may see at least 50 percent of their tasks impacted,” the authors estimate.

Significantly, though, they do not speculate about the amount of time those changes will take, and when they will be realized. Again, there is a cost-benefit issue. To provide lots of value, generative AI has to prove it can address complex problems that displace high-priced labor or create other sources of value that drive growth, new products or markets.

McKinsey suggests a longer time frame as well. For specific capabilities, the timeline for achieving human-level performance has been pulled forward, compared to earlier forecasts. They suggest human level performance happening perhaps two decades earlier than previously seen:

• Creativity: from around 2048 to 2023

• Logical reasoning and problem solving: from around 2043 to 2023

• Natural language understanding: from around 2055 to 2025

• Social and emotional reasoning: from around 2050 to 2033

Still, all those developments are far outside the financial return window for capital investments to be made over the next several years, which might be expected to produce breakeven results on investment in three to five years, with gains thereafter.

The point is that operating profits from large capex programs typically are not seen in a matter of a few quarters. Granted, software firms might often expect capital investment “breakeven” points to be reached in two years or less. More capital-intensive “utility-type” firms might expect capex breakeven in two to five years.

Measurable generative AI returns should not take five years, as cost savings should be quantifiable, for some use cases, within a year or so. Measurable returns for other use cases might not be so easy, or so swift.

The ultimate results may well turn on how fast generative AI is able to prove useful for complex tasks. As always, much hinges on the assumptions we make. How much benefit will accrue from automation, and how much from faster rates of innovation?

For example, Acemoglu assumes that generative AI will automate only 4.6 percent of total work tasks, while Goldman Sachs economists estimate that generative AI will automate 25 percent of all work tasks following the technology’s full adoption.



source: Goldman Sachs Global Investment Research

“Acemoglu’s framework assumes that the primary driver of cost savings will be workers completing existing tasks more efficiently and ignores productivity gains from labor reallocation or the creation of new tasks,” say Goldman Sachs economists. “In contrast, our productivity estimates incorporate both worker reallocation—via displacement and subsequent reemployment in new occupations made possible by AI-related technological advancement—and new task creation that expands nondisplaced workers’ production potential.”

“Differences in these assumptions explain over 80 percent of the discrepancy between our 9.2 percent and Acemoglu’s 0.53 percent estimates of increases in total factor productivity over the next decade,” the Goldman Sachs authors say.

As always with forecasts, the assumptions are key. How much value, and when that value is obtained, all vary based on the assumptions.

Have Home Broadband Prices Gone Up, Down or Sideways since 2000?

It is hard to say, for certain, whether home broadband prices in the U.S. market have increased, decreased or stayed the same since about 2000, for several reasons. One can measure without adjusting for inflation. One can measure the “price most people pay,” which does not account for hedonic changes (better performance for the same price). Or one can measure any specific speed tier and its price over time. 

Without adjusting for inflation; consumer product choices or hedonic product changes, one might argue that prices have risen. Then there is the matter of whether survey respondents report the full price or only the advertised price before taxes and fees.  

Based on 2000 and 2023 typical prices, not adjusted for inflation, some might argue prices have stayed consistent, ranging between $30 and $75 a month for most of two decades. 

But there has been hedonic change, as downstream speeds have grown about two orders of magnitude, even if typical prices have remained roughly in the same ranges over time.  

 

Year	Average Download Speed (Mbps)	Typical Pricing Range	Sources
2000	0.5 - 3	$30 - $100+	AllConnect, NCTA
2005	3 - 6	$20 - $80+	AllConnect, NCTA
2010	10 - 25	$40 - $70+	AllConnect, NCTA
2015	25 - 50	$50 - $80+	AllConnect, NCTA
2020	50 - 100	$40 - $70+	AllConnect, NCTA
2023	100 - 200+	$30 - $100+	AllConnect, NCTA