Many dot-com bubble predictions for frictionless commerce failed because the technology wasn't capable. Today's powerful AI agents represent the maturation of that tech, finally enabling the seamless disintermediation that was envisioned decades ago.

Arif Hilali of Bain Capital Ventures warns investors against mistaking Silicon Valley hype for mainstream adoption. He uses cloud computing's slow, multi-decade rollout as a parallel for AI, suggesting that even when a trend seems obvious inside the tech bubble, its true market penetration takes much longer than anticipated.

The AI era's high velocity of change, where market leaders can be displaced in 1-2 years, resembles the volatile dot-com bubble, not the last decade's predictable SaaS growth. This means founders must consider that even massive scale doesn't guarantee durability, making exit timing a critical strategic question.

Unlike the dot-com bubble driven by fleeting startups, the AI boom is a sustainable "megatrend." It's led by established giants like Microsoft and Google, developing on a compressed 5-7 year timeline (vs. 15 years for the internet), and operating at a scale 1000x larger, suggesting longevity over a sudden collapse.

Cresta's CEO argues that while the internet's evolution from 1995-2001 was somewhat foreseeable, the advancements in AI since 2019 would have been unimaginable even to the experts who wrote the foundational papers. This highlights the unprecedented nature of the current technological shift.

Drawing parallels to the Industrial Revolution, Demis Hassabis warns that AI's societal transformation will be significantly more compressed and impactful. He predicts it will be '10 times bigger' and happen '10 times faster,' unfolding over a single decade rather than a century, demanding rapid adaptation from global institutions.

To grasp AI's potential impact, imagine compressing 100 years of progress (1925-2025)—from atomic bombs to the internet and major social movements—into ten years. Human institutions, which don't speed up, would face enormous challenges, making high-stakes decisions on compressed, crisis-level timelines.

A leading AI expert, Paul Roetzer, reflects that in 2016 he wrongly predicted rapid, widespread AI adoption by 2020. He was wrong about the timeline but found he had actually underestimated AI's eventual transformative effect on business, society, and the economy.

The risk of an AI bubble bursting is a long-term, multi-year concern, not an imminent threat. The current phase is about massive infrastructure buildout by cash-rich giants, similar to the early 1990s fiber optic boom. The “moment of truth” regarding profitability and a potential bust is likely years away.