I have been deep diving into AI systems lately and the implication of Jevon’s paradox — how efficiency improvements can lead to reduced costs, increased consumption, and democratization of AI applications.
What about the inverse?
In my experience, lack of investment and continuous improvement in critical systems leads to stagnation. These systems become black boxes and minefields for tech leaders and users alike.
What Is Jevon’s Paradox?
Jevon’s Paradox suggests that as efficiency in using a resource increases, overall consumption of that resource often increases too. For example:
- More efficient lighting reduces the cost of use.
- Lower costs lead to increased usage.
- The result? Net resource use can actually go up due to a feedback loop.
But What Happens When the Opposite Occurs?
In tech systems, the inverse of Jevon’s Paradox emerges when:
- A system is deemed stable or good enough, so no further improvements are made.
- Year after year, it receives little to no funding.
- Over time, this results in an outdated system that is functional — but increasingly obscure, costly, and brittle.
For Example, Consider a 30-Year-Old Legacy Application
The scenario:
- A critical business system built in the early ’90s.
- It still functions, but hasn’t been updated in decades.
- Original creators are gone; documentation is missing.
- No budget is assigned for modernization — only critical fixes are encouraged.
The impacts:
- Usage declines: Only a few specialized developers and operators know how to use or maintain it.
- Change becomes expensive: Modifying a single feature requires hours of detective work, manual testing, and risk analysis.
- Technical debt compounds: Every change adds more instability to an already fragile system.
Underinvestment Shrinks the User Base
Unlike Jevon’s original paradox — where efficiency grows adoption — here, lack of investment drives usage down:
- Non-technical users can’t interface with the system.
- New employees avoid it due to steep learning curves.
- Only a small pool of experts can maintain or operate it, increasing dependence on tribal knowledge.
This creates a negative feedback loop:
- Fewer users → less justification for investment.
- Less investment → system becomes harder to use.
- Harder to use → even fewer users.
Exponential Cost of Change
Another hidden cost of neglect: the cost of change doesn’t stay linear — it balloons.
- Each attempted improvement introduces cascading risk.
- Without proper tooling or documentation, even a minor fix requires deep, manual intervention.
- What might take minutes in a modern platform takes weeks in a legacy system.
This is the interest on technical debt — and the longer you delay, the more it grows.
Why This Is the Inverse of Jevon’s Paradox
Instead of abundance through optimization, we get lower resource utilization (or stagnation) and poor user experience.
Breaking the Legacy Cycle
To escape this trap, organizations must:
- Recognize technical debt as risk: it’s not just a development issue — it’s a business liability.
- Encourage active usage: invest in interfaces and documentation that invite more users in.
- Modernize incrementally: start small, document often, and build momentum.
- Promote cross-training: prevent knowledge silos by democratizing access and understanding.
Final Thoughts: The Real Cost of Doing Nothing
Legacy systems don’t just grow old — they grow expensive, exclusive, and existentially risky.
Inverting Jevon’s Paradox teaches us this: when you stop investing in a technology, you don’t just halt progress — you shrink access, increase fragility, and inflate future costs.
So next time someone says, “It still works, why touch it?” — consider responding:
“Because the cost of doing nothing is already growing faster than we can measure.”