The Complexity Generation Gap
What happens when producing complexity becomes almost free
There is a ratio inside every functioning institution that almost nobody measures.
On one side is the number of things arriving that require a decision:
applications, claims, filings, disputes, permits, appeals, registrations and inspections.
On the other is the institution's capacity to resolve them.
As long as resolution capacity keeps pace with generation capacity, the system works.
When it doesn't, a backlog begins.
And something fundamental changes.
The signal
Global patent applications reached 3.7 million in 2024, according to WIPO — a record and nearly double the level of 2010.
Now consider a very different institution.
Major high-risk provisions of the EU AI Act were scheduled to become applicable in August 2026. Their implementation was deferred because national authorities, technical standards and compliance infrastructure were not sufficiently ready.
These appear to be unrelated developments.
They point toward the same structural problem.
The world is becoming capable of producing matters requiring institutional judgment faster than institutions can expand their capacity to judge them.
Artificial intelligence accelerates this dramatically.
But AI did not create the trend.
Regulatory accumulation creates more assessments.
The energy transition creates permits and environmental reviews.
Trade fragmentation creates sanctions checks, origin verification and export-control decisions.
Ageing populations increase healthcare, pension and benefit determinations while simultaneously reducing the working-age population from which assessors are recruited.
Several independent forces are feeding the same bottleneck.
What happens next?
Probably not institutional collapse.
Institutions have another option:
reduce the complexity they are required to process.
We expect four responses to appear increasingly often:
Deferral — move deadlines.
Simplification — replace individual assessments with categories and thresholds.
Higher barriers to entry — increase fees or tighten requirements.
Reduced discretion — stop making some individual judgments altogether.
Our base case assigns a 55% probability to institutional simplification becoming the dominant response.
Another 25% probability goes to two-speed institutions: machines process standard cases while scarce human judgment is concentrated on exceptions.
Together, these scenarios describe a world where the ordinary case may actually become faster.
The price is paid by the exception.
What this means for you
Individuals
File earlier than you think necessary.
Keep documentation complete and reproducible.
Where the cost is small, prefer arrangements that fit established categories rather than depending on discretionary assessment.
Business
Add institutional throughput to your risk register.
A regulator may not reject your project. A permit may simply arrive eighteen months later than expected.
That difference can be just as consequential.
Structure ordinary processes so they can increasingly be assessed through standardised, machine-readable procedures.
Capital
Separate refusal risk from delay risk.
They are not the same.
A project receiving approval two years late can destroy returns even if the approval itself was never seriously in doubt.
Over time, administrative capacity may need to be evaluated alongside taxation, political stability and rule of law when comparing jurisdictions.
Our forecast
By mid-2027: generation volumes continue rising faster than qualified resolution capacity.
Probability: 75%
By 2029: deferral and simplification become increasingly normal responses to institutional overload.
Probability: 70%
By 2031: automated adjudication begins handling more high-volume standard cases, with humans increasingly concentrated on exceptions and appeals.
Probability: 55%
By 2036: institutional processing capacity may become a meaningful factor in where economic activity locates.
Probability of a pronounced version: 45%
The critical distinction is this:
The future may not contain fewer institutions or fewer rules.
It may contain institutions capable of recognising fewer distinctions.
The standard case becomes easier to process.
The unusual case becomes harder.
And that changes the economics of being an exception.
Chaos is the rising cost of the next choice.
When institutions simplify, some choices eventually stop being available at any price.
THRIVE IN CHAOS
Decision Intelligence for an Uncertain World
Analysis → Forecast → Recommendations
Signal → Meaning → Action → Stability
Signal Over Noise