In a recent LinkedIn post, John Barrows delves into the often-hyped trajectory of Artificial Intelligence adoption, questioning whether the industry’s perceived imminent “meltdown” is unfolding at the pace many headlines suggest. Barrows contrasts the rapid advancement of AI technology with its surprisingly slow real-world adoption rate, drawing parallels to historical technological shifts.
Barrows highlights data from the St. Louis Fed, noting that the number of individuals using AI daily for work remains largely stagnant. He observes that while experimentation is rife, most organizations have yet to fundamentally alter their operational paradigms. This observation leads him to consider the broader implications for the business landscape.
“People are experimenting, but most companies haven’t fundamentally changed how they operate yet.”
The Historical Adoption Curve of Technology
Drawing on insights from Citadel Securities, Barrows points out that major technological advancements, including personal computers, the internet, and electricity, have historically followed a predictable adoption curve. During these periods, predictions of mass unemployment were common, yet the reality was a transformation of work rather than its elimination.
As Barrows notes:
“Citadel’s point is that every major technology follows the same adoption curve. PCs, the internet, electricity. People predicted mass unemployment every single time, and what actually happened is the work changed. It didn’t disappear.”
However, Barrows offers a personal counterpoint, suggesting that the speed of AI development may distinguish it from previous technological waves, though he defers to economists on definitive conclusions.
Economic Factors Shaping AI’s Pace
An intriguing point raised by Barrows, echoing the Citadel Securities perspective, concerns the economic implications of widespread AI automation. He explains that a simultaneous push for automation by many entities could lead to an exponential increase in demand for computing resources—including chips, data centers, and energy. This surge, in turn, would escalate the cost of AI deployment.
According to Barrows:
“At some point, running an AI agent costs more than paying a person to do the work, and that creates a natural ceiling on how fast this all plays out.”
This economic reality suggests a potential natural brake on the speed of AI integration, challenging the notion of an overnight revolution.
Current Market Signals and Future Outlook
Further supporting his nuanced view, Barrows cites a surprising statistic: job postings for software engineers have increased by 11% year-over-year. He argues that this trend contradicts the narrative of AI actively replacing developers at present. Additionally, he observes that new business applications are at record highs, indicating a surge in entrepreneurship rather than widespread displacement.
In Barrows’ view:
“People aren’t sitting around waiting to be replaced. They’re starting companies.”
While Barrows maintains his conviction that the per-seat SaaS model faces significant challenges, evidenced by distressed software debt, he posits that a complete industry meltdown might be a more gradual process than widely anticipated. He concludes that companies possessing robust data, strong integrations, and demonstrable value will likely navigate this transition successfully, while those burdened by debt and selling undifferentiated products are more vulnerable.
Barrows concludes by posing a critical question to his audience: Is the current market dynamic a correction or an extinction event?
📝 About This Content
This article is based on insights shared by John Barrows on LinkedIn.
📅 Originally posted on April 30, 2026 | View original post on LinkedIn →