The Operator’s View: Infrastructure Under Load
Planning Lag: Why Infrastructure Can’t Keep Up With Demand
Article 3
2 min read
Its not that infrastructure systems are failing, they are operating as designed. The issue is that the design assumed time and demand no longer does.
Delays are seen as execution problems: approvals too slow, delivery too complex, coordination lacking. The assumption is that with better management, the system will catch up. Now at small scale, this holds. When in increased at system scale, it breaks.
Infrastructure follows a sequential model: forecast, plan, approve, fund, deliver. Each step dependant on the last. While this works when demand is predictable and incremental. AI infrastructure and electrification are not. They create a step change demand, forming faster than planning cycles can adjust.
Planning responds to validated demand, not emerging demand. Validation introduces delay through modelling, approvals, and funding alignment. By the time demand is confirmed, conditions have already shifted. The result is structural: infrastructure is always being planned for yesterday’s demand.
Forecasts from the Australian Energy Market Operator show accelerating demand from data centres and electrification. At the same time, connection queues are growing and delivery timelines for transmission and substations extend over years. This is not backlog. It is timing mismatch between demand formation and infrastructure delivery.
What improves is planning accuracy, risk control, and capital discipline. What degrades is responsiveness and timing. It is not immediately visible because each project appears justified and controlled. Collectively, delays compound. Ownership is distributed across planners, regulators, and delivery organisations. Each operates correctly in isolation; the lag exists between them. It persists because reducing lag increases risk, and infrastructure systems are designed to minimise risk.
From my experience, the constraint is not just physical infrastructure. It is time embedded in the planning system. In this instance, AI and electrification compress timelines and infrastructure expands timelines. This cannot be solved by acceleration alone, as it requires earlier commitment to capacity and acceptance of uncertainty before demand fully materialises.
At a small scale, planning lag is manageable. At increased system scale, it becomes the constraint. Infrastructure will continue to be delivered, but not at the speed demand now requires. Its an important consideration during the planning process.
Continue the Series
Previous: Article 2 Energy Constraint When AI and Transport Compete for Power
Next: Article 4 Regional Constraint, Where Infrastructure Reality Becomes Visible
Please see below for full article and more details.
The Operator’s View: Infrastructure Under Load
Planning Lag: Why Infrastructure Can’t Keep Up With Demand
Article 3
Full Article (5–6 Minute Read)
Infrastructure systems are not failing.
They are operating exactly as designed.
The problem is the design assumes time.
And demand no longer does.
What people think
Infrastructure delays are often framed as execution issues.
Approvals take too long.
Projects move too slowly.
Delivery needs to accelerate.
The assumption is that with better coordination, the system will catch up.
At small scale, that can be true.
At system scale, it breaks.
What’s actually happening?
Infrastructure is built on sequential logic: Forecast = Plan = Approve = Fund = Deliver.
Each step depends on the one before it.
This works when demand is:
· stable
· predictable
· incremental
That condition no longer holds. AI infrastructure and electrification are driving nonlinear demand and not gradual increases.
Step changes, those step changes are occurring faster than planning cycles can adjust.
Where the system breaks
Planning does not respond to demand in real time.
It responds to validated demand.
That validation process introduces delay by design:
· modelling and forecasting
· stakeholder alignment
· regulatory approval
· capital allocation
By the time demand is confirmed, the conditions that created it have already shifted.
This creates a structural gap: Infrastructure is always being planned for yesterday’s demand.
Real-world signal
Recent projections from the Australian Energy Market Operator show accelerating demand from both electrification and data centres.
At the same time, connection queues for large loads are increasing, and delivery timelines for transmission and substation upgrades extend over multiple years.
This is not a backlog.
It is a timing mismatch between:
· how fast demand is forming
· how long infrastructure takes to deliver
Failure mode
What improves: planning accuracy, risk management, capital discipline.
What degrades: responsiveness, adaptability, delivery timing.
Why it is not immediately visible: Each individual project appears justified and well-sequenced.
The system looks controlled, however, collectively, delays are compounded.
By the time infrastructure is delivered, demand has already moved.
Ownership is distributed:
· planners validate demand
· regulators approve
· delivery organisations build
Each step is correct in isolation.
It’s the lag that exists between them.
Why it persists? Because reducing lag introduces risk.
Noting that: infrastructure systems are designed to minimise risk, not maximise speed.
What this means
The constraint is not just physical infrastructure.
It is time embedded in the planning system.
AI and electrification compress timelines.
Infrastructure expands them.
That mismatch cannot be resolved by acceleration alone.
It requires a shift in how planning is approached:
· earlier commitment to capacity
· acceptance of forecast uncertainty
· coordination across sectors before demand fully materialises
What happens next
At small scale, planning lag is manageable.
At system scale, then it becomes the constraint.
Not because infrastructure cannot be built.
It is because it cannot be built fast enough to match how demand is now forming.
The system will continue to deliver.
Just not in time.
Continue the Series
Previous: Article 2 Energy Constraint When AI and Transport Compete for Power
Next: Article 4 Regional Constraint, Where Infrastructure Reality Becomes Visible
Stay Connected
If you're interested in discussions on AI or how AI is actually delivered, across infrastructure, energy, networks, materials, and supply chains, please subscribe:
https://digitalbackbone-be8806.beehiiv.com/
Footnote
This article is part of a series exploring topics: AI is constrained by physical infrastructure, and increasingly shaped by economic behaviour at scale
Disclaimer
The views expressed in this article are my own and are intended for general information and discussion purposes only. They do not represent the views of any employer, organisation, or client.
© 2026 Rodney Terry – Digital Backbone. All rights reserved.