The Lighthouse Member Report

History, Industry, and the AI Adoption Curve

A member framework for understanding AI through history: industrial change, S-curves, bubbles, progress, disruption, and the market-selection pattern that repeats beneath each new technology wave.

Executive Summary Watchlist Framework
Member thesis: the technology can be real and the bubble can be real at the same time.

Transformative tools can improve life, raise productivity, and reorganize industries while still producing overinvestment, crowded narratives, and painful shakeouts along the way.

Overlapping technology adoption waves showing one tool rising, peaking, fading, and being followed by another wave.
Adoption waves are not clean handoffs. Older tools fade, linger, or get absorbed while the next wave becomes obvious.

Executive summary

AI should be read through history, not headlines alone.

Major technology shifts usually create a mix of real transformation and excessive investment enthusiasm. The useful framework is not "AI wins" or "AI is a bubble." The useful framework is adoption, overinvestment, shakeout, consolidation, and practical use.

History gives members a way to stay calm. The Industrial Revolution, the personal computer era, the dot-com period, and the mobile phone transition all show that new technologies can reshape the economy while early market assumptions still prove wrong.

Looking backward, the winners feel obvious. Looking forward, they rarely are. WealthVelocity members should use this report as a context map for separating real adoption from speculative overreach.

Progress and turbulence

Change is worth being grateful for, even when the transition is not gentle.

Productivity growth is one of the great positive forces in economic history. It helped lift incomes, improve living standards, pull labor out of subsistence work, and make cities, industries, and modern services possible.

That does not make every transition easy. A useful analogy is the caterpillar becoming a butterfly: the end result can be beautiful, but the process is not calm for the creature going through it.

AI should be treated with that same balance. It may become part of a long story of productivity improvement, but investors should still expect stress, displacement, overconfidence, and re-sorting before the mature market becomes clear.

Why history matters now

The landscape changes slowly, then suddenly.

Industrial revolutions reorganized labor, capital, geography, production, and ownership. The modern AI shift is different in form, but similar in effect: it changes where value is created, which skills matter, which companies gain leverage, and which business models become easier to replace.

01 Technology Appears

A new tool becomes possible, but early versions are incomplete, expensive, or awkward.

02 Capital Rushes In

Investors start pricing the future before customer behavior and margins are fully proven.

03 Weak Models Get Exposed

Useful technology does not protect every company from bad pricing, weak distribution, or poor execution.

04 Durable Use Remains

The market eventually sorts winners, useful niche tools, and also-rans.

05 Behavior Repeats

The content changes, but human preference around cost, performance, value, trust, and convenience repeats.

06 The Category Changes

What looked like the product can become only a stepping stone to the next platform.

S-curves and bubbles

Adoption and enthusiasm are related, but they are not the same curve.

Technology adoption often follows an S-curve. Early adoption is slow because the tool is incomplete, expensive, unfamiliar, or difficult to integrate. Then adoption accelerates as performance improves, costs fall, distribution expands, and users understand the value.

Investment enthusiasm can follow a different curve. Markets may rush ahead before adoption evidence is complete, especially when the potential market sounds enormous.

Prior waves

Looking backward, winners feel obvious. Looking forward, they rarely are.

Personal computers created durable platforms, but not every early hardware or software company survived. Wang Laboratories is a useful historical reminder: a company can be respected, useful, and associated with an important product category, yet still fail to become the long-term platform winner.

The internet was real. The dot-com bubble was also real. Many companies with weak models disappeared, while the infrastructure, consumer behavior, and business transformation continued.

Nokia looked durable before the smartphone transition changed what a phone was. BlackBerry became highly important for a period, then lost position as consumer expectations, developer ecosystems, touch interfaces, and app platforms shifted.

What this means for AI

The AI market will likely sort into leaders, laggards, also-rans, and losers.

AI tools are competing to become part of daily workflows. Some will become durable leaders. Some will lag but remain useful in narrower niches. Some will become also-rans with limited pricing power. Others will lose relevance as features are absorbed into larger platforms.

Leaders

Broadly adopted tools with distribution, trust, workflow integration, and pricing power.

Laggards

Specialized tools with vertical expertise, lower scope, or differentiated workflows.

Also-Rans

Products that may still provide value but lack durable distribution or pricing power.

Losers

Tools that lose relevance as better products, stronger platforms, or embedded features replace them.

Watchlist framework

What members should monitor.

These categories help separate real adoption from market enthusiasm.

01 Adoption Quality

Look for recurring use, not only trial accounts or headline user growth.

02 Pricing Power

Watch whether companies can charge for AI features, or whether AI becomes a cost of doing business.

03 Retention

Strong products should become habits. Weak products may see novelty-driven usage fade.

04 Workflow Integration

Embedded tools may have an advantage over tools that require users to change behavior too much.

05 Infrastructure Demand

Data centers, power, chips, cooling, networking, and cloud capacity can benefit even while app winners remain uncertain.

06 Margin Evidence

AI revenue is more valuable when it improves margins or creates durable recurring revenue.

Signal interpretation

How this should show up inside WealthVelocity.

Signal Bullish Confirmation Warning Sign
Adoption Users keep returning and usage becomes embedded in daily work. Usage is driven by novelty, free access, or promotional bundles.
Pricing Customers pay for AI features without high churn. Providers discount heavily or absorb rising compute costs.
Infrastructure Demand remains broad across chips, power, cooling, networking, and data centers. Spending narrows into a few crowded names.
Competition Leaders prove durable distribution and switching costs. Features become interchangeable and pricing power weakens.
Margins AI improves productivity, retention, or revenue quality. AI increases costs faster than monetization.

Portfolio posture

Use history as a filter, not a script.

The practical response is not to chase every AI name or reject the entire theme. Real technology shifts tend to create durable value, but not evenly. The first wave of market enthusiasm often funds too many competitors.

Later, the market sorts the field by adoption, pricing, distribution, product quality, and financial discipline. For members, this report should connect to AI infrastructure, software platforms, semiconductors, power demand, enterprise adoption, consumer behavior, market breadth, and valuation risk.

Reference packet

History, narrative, and source notes.

James Burke's Connections is the strongest conceptual reference for this report: complex technological history explained as a chain of related causes, inventions, dependencies, and adoption shifts. Fisher's Booms and Depressions is included as a historical reading reference because the market behavior still feels recognizable.

James Burke: Connections Connections episode list Irving Fisher: Booms and Depressions

Source note: Fisher is linked from the public WealthVelocity library. Additional reference PDFs can live in the same folder as future reports require them.