An article in the July 19, 2025 issue of The Economist asked “Why is AI so slow to spread?” in its title. “Talk to executives and before long they will rhapsodise about all the wonderful ways in which their business is using artificial intelligence,” said the article. CEOs proudly bragged about the impact AI is already having in their companies, from “450 use cases” at a large bank to becoming “the new operating system” at their chain of restaurants. “In the first quarter of this year executives from 44% of S&P 500 companies discussed AI on earnings calls.”
But, despite the executives high expectations, “AI is changing business much more slowly than expected.” According to a recent Goldman Sachs article, while the “technical achievements in AI are racing ahead, … some important questions remain about the technology’s ultimate impact on the economy and corporations’ bottom line. Enterprise adoption of AI still must prove itself and overcome corporate inertia, and reasonable questions remain about the return on the extraordinary level of investment.”
“In recent months the firms’ share prices have underperformed the market,” added The Economist. “With its fantastic capabilities, AI represents hundred-dollar bills lying on the street. Why, then, are firms not picking them up? Economics may provide an answer.”
An article in the June 24, 2025 issue of the Harvard Business Review (HBR) by Wellmark CIO Paul Hlivko nicely explains why “The AI Revolution Won’t Happen Overnight.” “AI will transform industries,” wrote Hlivko. “However, this transformation will happen on enterprise time: longer, slower, and with far more friction than most expect, and much slower than Silicon Valley is selling.”
“Right now, companies are getting six fundamental things wrong about how AI will create value and how long it will take — and risk wasting resources, overpromising results, and eroding trust:
- AI’s real impact will take much longer than we think.
- We’re being wildly optimistic about enterprise AI adoption.
- The market is overestimating the value of AI companies.
- The real money isn’t in the models.
- We’re over indexing on startups.
- We’re obsessed with generative AI but it’s not the future.
Let me summarize each of these six fundamentals that, according to Hlivko, companies are getting wrong.
AI’s real impact will take much longer than we think
AI is a general purpose technology (GPT). And, as explained in “The Productivity J-Curve,” a 2018 article by Erik Brynjolfsson, Daniel Rock, and Chad Syverson, general purpose technologies (GPTs) “are the defining technologies of their times and can radically change the economic environment. They have great potential from the outset, but realizing that potential requires larger intangible and often unmeasured investments and a fundamental rethinking of the organization of production itself.” The authors called this phenomenon The Productivity J-Curve, because like the letter ‘J’, GPT productivity dips initially in its investment phase while later rising in the deployment phase.
We’ve seen many GPTs over the centuries, from the printing press and electricity to computers and the internet, and they all follow the same pattern. Even after reaching a tipping point of market acceptance, it takes considerable time, — often decades, — for these new technologies and business models to be widely deployed across industries and economies and for their full benefits to be realized.
For example productivity growth did not increase until 40 years after the introduction of electricity in the early 1880s, because It took until the 1920s for companies to figure out how to restructure their factories to take advantage of electric power with new manufacturing innovations like the assembly line. And, while the internet was already being used in research communities in the 1970s, it wasn’t until it became widely deployed across the economy in the 2000s that it transformed business models.
“There are compelling reasons to think that AI will follow the same slow but inevitable trajectory.” In an April, 2024 article, “The Simple Macroeconomics of AI.” MIT economist and Nobel laureate Daron Acemoglu wrote that only 5% of tasks will be profitably automated in the next decade, adding just 1% to the U.S. GDP — a far cry from the seismic shift many expect. The challenge, he argued, is that for most organizations, the costs of disruption, retraining, integration, and computing will outweigh the returns for most tasks.
In a subsequent Goldman Sachs Research report, senior strategist Allison Nathan interviewed Professor Acemoglu and asked him why he is less optimistic of AI’s potential economic impacts than other economists and financial analysts. Acemoglu replied that the forecast differences are primarily about the timing of AI’s economic impacts rather than about the ultimate promise of the technology. “Generative AI has the potential to fundamentally change the process of scientific discovery, research and development, innovation, new product and material testing, etc. as well as create new products and platforms.” But the economic impact of historically transformative technologies like AI will take time to play out.
We’re being wildly optimistic about enterprise AI adoption
“When ChatGPT launched, AI felt like magic — an overnight revolution,” noted Hlivko. “Earnings calls were flooded with AI mentions. Venture capital shifted into overdrive. Headlines promised AI’s transformation would be instant and all-encompassing. We’ve seen this kind of overheated hypecycle before — with the early personal computers, dot-com bubble, the blockchain boom, and even the very early days of cloud computing — and we’ll likely make this mistake again.”
He explained that we tend to misjudge the impact of major technological changes due to three cognitive biases:
The planning bias makes us underestimate how long transformations take. “This phenomenon sometimes occurs regardless of the individual’s knowledge that past tasks of a similar nature have taken longer to complete than generally planned. The bias affects predictions only about one’s own tasks. On the other hand, when outside observers predict task completion times, they tend to exhibit a pessimistic bias, overestimating the time needed.”
The optimism bias convinces us that the adoption of the new technology will be smooth and easy. This cognitive bias “causes someone to believe that they themselves are less likely to experience a negative event. It is also known as unrealistic optimism or comparative optimism. It is common and transcends gender, ethnicity, nationality, and age.”
The recency bias leads us to believe AI’s viral consumer adoption will translate seamlessly into the enterprise. This third cognitive bias “favors recent events over historic ones. A type of memory bias, recency bias gives greater importance to the most recent event such as the final lawyer’s closing argument a jury hears before being dismissed to deliberate.”
The market is overestimating the value of AI companies
“Investors are making a critical error around AI: They’re treating AI companies like high-growth, asset-light software firms, when in reality they’re capital-intensive, high-cost, and infrastructure heavy. AI-heavy tech stocks have traded at a 20–40% premium, assuming future profits that haven’t materialized.”
This is not only a market misread, but also an execution trap. “Inflated valuations set unrealistic expectations that trickle down into the enterprise: pressure to move fast, to pilot something flashy, to be seen doing AI. The result? Rushed rollouts, misaligned priorities, and investments in the magic rather than margin performance. In a market priced for miracles, the real advantage lies in restraint — leaders who prioritize integration over spectacle and long-term value over short-term visibility.”
The real money isn’t in the models
“Even if AI model companies turn a profit, they won’t be able to defend their advantage. AI’s biggest breakthroughs—like neural networks and attention mechanisms — are just math, and math can’t be patented. … The real test of AI isn’t whether we can build something new. It’s whether we can embed it deeply enough into business systems to generate durable, measurable value.”
We’re over indexing on startups
“The market hype is fixated on AI startups, but big incumbents have the real advantage in the enterprise.” AI is clearly a disruptive technology, but it’s real economic value will be how quickly it can be deployed across industries and economies. While startups will play a major role in pushing AI innovation forward, the incumbents, — such as Microsoft, Google, and Salesforce, — have the ability and budgets to embed AI into their existing enterprise stacks, as well as to provide the necessary enterprise support. “That’s how AI adoption happens — whoever owns the enterprise and consumer workflow wins.”
We’re obsessed with generative AI but it’s not the future
“We’re fixated on generative AI, but the future lies beyond chat-based models. Today’s AI excels at summarizing reports and drafting emails but struggles with real-world complexity. It lacks situational awareness, complex reasoning, and the ability to synthesize multiple types of changing information in real-time. That’s why AI adoption lags in fields like medicine and logistics — where decisions require more than historical text. A chatbot can draft a contract, but it can’t diagnose every patient or optimize a failing supply chain.”
“The future of AI isn’t about building a better chatbot. It’s about designing systems that see, hear, analyze, and act in concert — at scale, and in sync with the complexity of the real world.”
“Can we think smartly about machines?,” asks Hlivko in the article’s concluding section. For the last 75 years we’ve been evaluating AI based on its human-like intelligence and cognitive abilities. “Companies are treating AI as if it’s a silver bullet, throwing billions at models while neglecting the harder work of integration, infrastructure, and real business value. But one thing is certain: AI’s ubiquity will erode its exclusivity. Its impact won’t be in who owns it, but in how we use it.”
