The Long View: What Product Leaders Must Learn From Past Tech Cycles to Navigate the AI Hype

Why does every new wave of technology feel both inevitable and over-hyped? From the early Web to the two mobile revolutions, ecommerce booms and the dotcom bust, the pattern repeats: rapid promise, winner-takes-most dynamics, then messy consolidation — and often, slow decay. For product leaders today, AI is both the opportunity and the hazard. The long view helps you separate durable strategy from fashionable noise.

Read cycles, not headlines

Technological progress moves in waves. The industrial revolution, the birth of mass media, the first Web era, and the smartphone era each created three things: new users, new business models, and new winners — but also new failure modes. Familiar examples are Kodak and Blockbuster, whose business models were disrupted, and later, platforms such as some large marketplaces that showed signs of decline once extraction outpaced value.

Seeing cycles means asking different questions: is this change creating an ecosystem shift, or is it a feature-level improvement that will be absorbed by incumbents? AI today shows both. Foundation models offer new capabilities (ecosystem), while many startups trade on single features (feature-level). Your job as a leader is to map where your product lands on that spectrum.

Three durable patterns product leaders must internalise

• Commoditisation follows accessibility: When a capability becomes cheap and ubiquitous, differentiation moves from capability to experience and trust. If everyone can generate recommendations or text, your competitive edge becomes how reliably and responsibly you deliver those outcomes.
• Platform economics bend behaviour: Platforms that control value flows can start to prioritise short-term monetisation over long-term experience — a dynamic Cory Doctorow famously calls enshittification. Leaders must design incentives and governance to resist that drift.
• Institutional memory matters: The mistakes of past waves are often repeated because organisations lack a memory mechanism. Document the failures and the constraints you face — technical, regulatory, and human — and use them when assessing new bets.

Concrete moves: how to act now

Here are practical steps you can take this quarter to turn macro-arguments about AI into product advantage.

• Map value flows: Draw a simple map that shows who creates value, who captures value and who is dependent on whom in your product’s ecosystem. If you’re a marketplace, where does your margin come from? If you’re embedding AI, who bears the cost of model updates and data curation?
• Design anti-rot guardrails: Make product KPIs include “experience permanence” measures — retention of core user journeys, meaningful feedback loops, and a clear escalation path when platform incentives threaten user value. This is how you prevent slow decay that looks like short-term growth.
• Invest in composability not monoliths: Build modular systems so that when an external capability (say a new model family) improves, you can swap or combine it without expensive rewrites. The firms that survived multiple waves did so with architectures that let them adapt fast.
• Protect novel experiments: Run a two-speed operating model. Keep your mission-critical product teams focused on stability and measurable outcomes, while a protected innovation cell experiments with unproven business models — with explicit criteria for when an experiment is promoted or killed.
• Measure social licence: Track trust signals — transparency, explainability and accessibility — as first-class metrics. Consumers and regulators care about these signals; ignoring them invites both reputation risk and costly compliance later.

Case in point: markets that avoided and those that ignored the long view

Marketplace platforms offer instructive contrasts. Some big marketplaces expanded rapidly by optimising for supply-side growth and extraction, which worked until creators and sellers found it unrewarding. Coverage of platform decline in outlets such as The Guardian shows what happens when incentives shift away from users and partners.

By contrast, successful product organisations that survived multiple waves put users and partners first — they actively monitored ecosystem health and refused easy monetisation that eroded trust. Those are not romantic choices; they are pragmatic defences against the Rot Economy.

Three questions to ask at your next leadership meeting

• What part of our product’s value chain will become a utility within two years, and how will we differentiate when that happens?
• Which incentives in our platform could, if pushed, generate short-term revenue but long-term decay of user trust?
• Do we have a protected path for experiments to scale when they demonstrate real user value, and a clean kill switch when they don’t?

Where the long view leads you

Thinking in cycles makes you less reactive to headlines and more ruthless about what to build and protect. It encourages modular engineering, user-first economics, and governance that avoids the slow creep of extraction. If you treat AI as just another toolkit, you’ll be outcompeted by teams that treat it as a systems design problem — one that touches data, incentives, regulation and product experience.

Start by mapping your value flows, codify trust as a metric, and insulate real experiments from corporate gravity. Do that and you’ll find the AI wave offers genuine advantage instead of another bout of fashionable churn. If you want one practical outcome from this article: schedule a 90-minute leadership session next week to draw your value-flow map. It’s surprising what clarity that small investment delivers.