The map arrived in a cardboard tube, and it took the whole kitchen table to unroll. A pale red band ran across it at an angle that ignored every fence, contour and paddock name the family had used for four generations. The covering letter was courteous and precise. It explained that a corridor was under investigation, that the line shown was indicative only, that a survey team would like access in the coming months, and that compensation arrangements existed and would be explained in due course. Nobody at that table doubted the letter was written in good faith, and nobody at that table was against renewable energy. What they could not work out, reading it twice, was where in the process it described the word "no" was supposed to go.
Nine hundred kilometres away, a project engineer was refreshing a shipping tracker. The substation site was cleared, the concrete was cured, the crews were on day rates, and the transformer was on a vessel that had been rescheduled twice. Every week it did not arrive was a week of standing costs on a project whose approved budget had been set years earlier, in a different market for steel, copper and skilled labour. The towers and the conductor were the easy part. The equipment that turns a line into a working connection was the part nobody outside the industry had heard of, and it was the part running late.
Neither of these people is an obstacle. The farmer is protecting a productive asset and a family's working life against a permanent change imposed from outside. The engineer is trying to finish a piece of national infrastructure on a schedule set by a plan. Both are behaving exactly as a reasonable person in their position would. And between them sits the least discussed constraint on Australia's entire energy transition: not whether we can build wind and solar farms, but whether we can build, and energise, the wires that carry what they make.
What the plan actually says, as at the 2026 ISP
Start with the correction, because the popular version of this argument is now wrong and this site does not run arguments it knows to be wrong.
For several years the standard line in energy commentary — including in earlier drafts of this article — was that the Australian Energy Market Operator's Integrated System Plan required roughly 10,000 kilometres of new transmission, and that delivery was nowhere near it. That framing no longer matches the record. In its 2026 Integrated System Plan, AEMO puts the requirement at around 6,000 km of new transmission by 2050 under its central Step Change scenario — an extension of about 14 per cent, or roughly one-seventh, of the existing network of about 44,000 km (AEMO, 2026 Integrated System Plan, 2026). The Draft 2026 ISP published in December 2025 set out the same figure, and described it as down from about 7,600 km in Step Change in the 2024 ISP, and from close to 10,000 km measured across that plan overall (AEMO, Draft 2026 Integrated System Plan, Dec-2025).
The requirement went down. Anyone still saying "AEMO says we need ten thousand kilometres and we have built almost none of it" is quoting a superseded document. That matters, and it is stated first here, because the real argument is more interesting than the false one and survives it intact.
Who decides, who pays, and who actually builds
Australia does not have one transmission system with one owner. Understanding the delivery problem requires knowing who holds which lever, because the answers to "who caused this" and "who can fix it" are different for almost every step.
The plan is AEMO's. The Integrated System Plan is a biennial, whole-of-market roadmap for the National Electricity Market, produced under the National Electricity Rules. It identifies an optimal development path — the least-cost combination of generation, storage, transmission and, from 2026, distribution investment that meets forecast demand and legislated targets. AEMO says it tested more than a thousand potential paths across seven modelling iterations for the 2026 ISP (AEMO, 2026 ISP, 2026). Projects are classified as committed, anticipated, actionable, or future, and that classification is the closest thing Australia has to a national queue.
The money is regulated. Transmission network service providers — the state-based businesses that own and build the high-voltage network — recover their costs from customers through revenue determinations set by the Australian Energy Regulator, typically over five-year periods, with major projects assessed through a regulatory investment test for transmission or, increasingly, through jurisdictional pathways that sit outside it. The AER also reviews each draft ISP and reports on it publicly (Australian Energy Regulator, Review report — Draft 2026 Integrated System Plan, Jan-2026).
The Commonwealth's contribution is mostly financing, not construction. Rewiring the Nation, created in the October 2022 Budget, is a $20 billion concessional finance facility administered through the Clean Energy Finance Corporation, designed to lower the cost of capital on ISP transmission projects and therefore the eventual charge to consumers (Budget Paper No. 2, Oct-2022 Budget). Agreements with states followed, including agreed investment priorities of up to about $4.7 billion for New South Wales transmission (approx.; Commonwealth and CEFC announcements, 2022–2026).
The approvals and the land are the states'. New South Wales builds through the Electricity Infrastructure Investment Act 2020 and its Electricity Infrastructure Roadmap, Victoria coordinates through VicGrid, and Queensland runs its own priority transmission investment assessment. Each also operates a landholder access and compensation scheme, and AEMO's 2026 ISP counts those schemes among the policies it must model (AEMO, 2026 ISP, 2026).
So the attribution has to be drawn carefully, and this article draws it three ways. The Commonwealth inherited a 44,000 km network built over roughly seven decades by state monopolies, and it inherited the modern planning, environmental and native title framework those monopolies never had to work under. It caused the deadline pressure — the 82 per cent renewable electricity target for 2030, the Capacity Investment Scheme, and the financing architecture that assumes lines arrive on time. It presides over delivery, which is executed by state-owned and privately owned network businesses under state planning law, and which it can fund and coordinate but cannot directly build.
The strongest case for the policy
There is no precedent for this, the plan was always ambitious by design, and a planner that revises is doing its job
Take the government's and AEMO's case at its strongest, and it is strong enough that a fair critic has to concede most of it first. Nothing in Australian history looks like what is being attempted. The existing 44,000 km transmission network was assembled over roughly seventy years, overwhelmingly by state-owned electricity commissions with compulsory acquisition powers, minimal consultation obligations, no federal environmental assessment regime, no native title framework, and a political culture that treated a transmission tower as progress rather than as an imposition. Doing a meaningful fraction of that again in two decades, under modern law, with modern consultation expectations, is genuinely hard in a way the historical comparison flatters nobody.
The schedule was ambitious on purpose, and that was a defensible choice rather than a mistake. A plan built around the pace at which Australia has recently been able to build lines would have been a plan to miss every emissions and reliability objective the parliament has legislated. Setting a stretching target and falling somewhat short can leave a country further ahead than setting a comfortable one and meeting it.
Landholder consultation that slows a project is not a failure of the system. It is the system. A landholder who objects to a high-voltage easement across productive country has a genuine, legally recognised interest: a permanent restriction on land use, altered access for machinery and aircraft, changed valuation, and a structure they did not ask for on ground they work. A democratic process that takes time to weigh that is a legitimate cost of doing things properly, not friction to be engineered away. And the practical case points the same way as the principled one — riding roughshod over consultation invites litigation, and litigation delays projects far more effectively than consultation does. AEMO says as much in its own language, noting that projects with more complex community considerations typically have longer design and approval processes, that those costs are built into its modelling where they can be estimated, and that the effort is well spent because it usually produces better outcomes for the community and for consumers (AEMO, 2026 ISP, 2026).
The downward revision is, in significant part, good news and good planning. Storage, consumer energy resources, better-sited generation and demand-side response genuinely can displace the need for lines. AEMO's stated reasons for needing less transmission include exactly that: more generation and storage now sits in particular locations, and forecasts have moved. A planner that refuses to revise a plan when the facts change is not being disciplined; it is being negligent. Building a line that turns out not to be needed is a permanent cost to consumers and a permanent imposition on the land it crosses, and avoiding one is a genuine saving.
Finally, the alternative is worse. AEMO's own counterfactual modelling puts the case bluntly: on the Step Change scenario, a development path with no new ISP transmission would require about $17 billion more in capital for generation, storage and firming, and about $12 billion more in operating and fuel costs, with far more gas generation and carbon capture required to stand in for the lines (AEMO, 2026 ISP, 2026). Building generation with no network to move what it makes is not a cheaper option. It is a more expensive one that also works less well.
The evidence: a requirement that fell, and the composition of the fall
The honest argument about transmission has two strands. The first is about what the reduction in the plan is actually made of. The second is about the distance between a project being financed, being built, and being switched on. Neither strand requires the false 10,000 km claim, and both are visible in AEMO's own published numbers.
The headline numbers, as at the 2026 ISP
The 2026 ISP's optimal development path extends the network by about 6,000 km by 2050 under Step Change. Of that, about 3,500 km sits in eight committed and anticipated projects that are already underway and expected to be delivered over the next eight years. A further 1,660 km sits in twelve actionable projects to be delivered over the next twelve years. Beyond those, future ISP projects could add up to about 876 km depending on which scenario eventuates (AEMO, 2026 ISP, 2026). Around 40 per cent of the new length strengthens the connections between states; the other 60 per cent connects new capacity inside renewable energy zones or moves power to demand centres within a state.
The network itself has grown. AEMO records that the NEM's transmission network stretched about 43,350 km in 2022 and spans about 44,000 km now — an addition of about 635 km, made up of the 365 km first stage of Project EnergyConnect between South Australia and its eastern neighbours, the 270 km Eyre Peninsula Link in South Australia, and a Far North Queensland connection into Woree (AEMO, 2026 ISP, 2026). That is a real, physical, energised addition to the national grid, and it should be counted as such.
What the reduction is made of
This is the analytical core, and it is a question of composition rather than headline.
The 2026 ISP states that the total build of new transmission is about 1,435 km less than projected in the 2024 ISP. It then decomposes that change in a footnote, and the decomposition is where the argument lives. About 165 km of projects progressed to operation and are therefore no longer counted in the forward plan. About 410 km was added by increases in project scope. And about 1,680 km of potential transmission is no longer needed, because there is more generation and storage in particular locations, or because of policy changes (AEMO, 2026 ISP, 2026).
The Draft 2026 ISP put the same idea in rounder terms three months earlier: the extent of new transmission had reduced compared with the previous ISP, with some projects downsized or no longer needed in response to policy or other changes (1,350 km), and others removed from the total because they had progressed to operation (365 km) (AEMO, Draft 2026 Integrated System Plan, Dec-2025).
Set those two components beside each other. On the Draft ISP's numbers, roughly four kilometres left the plan because they were downsized or dropped for every one kilometre that left because it was built and energised. On the final 2026 ISP's restated numbers, the gap is wider still — about 1,680 km against about 165 km, closer to ten to one.
A plan can be met by building it or by shrinking it. The ratio between those two tells you which one is happening.
The reasons AEMO gives, at full strength
A ratio is a headline, and headlines are where analysis goes wrong. So here are AEMO's reasons for needing less transmission, stated at their full weight, because several of them are entirely legitimate and one of them is close to unarguable.
First, substitution actually works. AEMO's stated reason for the largest component of the reduction is that there is more generation and storage in particular locations. If a battery, a pumped hydro scheme or a well-sited wind farm removes the need for a line, that is not a plan being quietly trimmed. That is the planning process doing precisely what a good planning process is for. Between July 2025 and January 2026 alone, an additional 5 GW of wind and solar and 11 GW/44 GWh of battery projects progressed to existing, committed or anticipated status (AEMO, 2026 ISP, 2026). Real assets, in real places, genuinely change what the network needs to do.
Second, the household sector moved. AEMO lifted its consumer energy resources forecast, in part reflecting stronger expected uptake under the Commonwealth's home battery program, with forecast investment in embedded storage by 2050 rising from 27 GW to 35 GW (AEMO, 2026 ISP, 2026). Power generated and stored on a roof does not need an interconnector to reach the kitchen below it. That is a genuine, physical reduction in network need, and it is a policy success rather than a retreat.
Third, one scenario got less ambitious. The 2024 ISP carried a Green Energy Export scenario built around very large hydrogen and green-commodity exports, implying a great deal of extra transmission. The 2026 ISP replaces it with a less expansive Accelerated Transition scenario, which AEMO says plainly reduces the potential need for transmission (AEMO, Draft 2026 ISP, Dec-2025). That is AEMO responding to a hydrogen export outlook that has weakened worldwide, not AEMO being timid. But it is worth naming precisely: a requirement that falls because the country quietly stopped planning for a hydrogen export industry is a different kind of good news from one that falls because batteries got cheap.
Fourth, a modelling assumption changed in transmission's favour. Previous ISPs applied a single 7 per cent weighted average cost of capital to all electricity infrastructure. At the recommendation of the ISP Consumer Panel and other stakeholders, the 2026 ISP applies 3 per cent to transmission — recognising that regulated revenues make these assets lower risk — and between 7 and 12 per cent to generation and storage (AEMO, 2026 ISP, 2026). The change is defensible and was made transparently. It also mechanically improves how transmission scores against alternatives. Both belong on the page.
Fifth, and cutting the other way, per-kilometre costs went up. AEMO notes that transmission costs in the plan fell despite higher capital costs per kilometre and more (smaller) projects, because the total length is shorter and some projects use less equipment (AEMO, 2026 ISP, 2026). Fewer, shorter lines can coexist with each remaining kilometre becoming harder and dearer to deliver. That is the bridge to the second strand.
The gap between financed, built and connected
The binding constraint on this program was never the headline kilometre count. It is the distance between three quite different states a project can be in, and the fact that public debate collapses them into one.
A project can be financed: approved by the regulator or a jurisdictional process, with a revenue allowance and a concessional loan attached. It can be built: easements acquired, towers standing, conductor strung. And it can be connected: energised, commissioned, carrying power at rated capacity into a market that can dispatch it. Money moves at the first stage, politics is announced at the first stage, and none of the electricity that matters exists until the third.
AEMO's own classification tracks this. Of the twelve actionable ISP transmission projects in the 2024 ISP, four had advanced to committed or anticipated status by the 2026 ISP, five remained actionable, and four were no longer actionable (AEMO, 2026 ISP, 2026). Advancing from actionable to committed is genuine progress — it means a proponent has taken an investment decision. It is not the same as electricity flowing.
The distinction is worth watching closely because status can change for reasons other than construction. Between the Draft and the final 2026 ISP, the New England REZ Network Infrastructure Project moved from actionable to anticipated. AEMO explains why: following advice from the New South Wales Government, it now considers the project sufficiently progressed through the state's framework to be classified as anticipated, and therefore did not retest its actionability in the 2026 ISP (AEMO, 2026 ISP, 2026, citing correspondence from the New South Wales Minister for Climate Change and Energy, 12 May 2026). That is a legitimate, transparent, properly footnoted planning judgement, and AEMO published the letter it relied on. It is also, plainly, an administrative reclassification rather than a construction milestone — and the project's listed in-service timing is January 2034.
Then there is the delivery rate. When AEMO modelled a Constrained Delivery sensitivity — a stress test of what happens if projects run late — it set its assumed build rates by reference to recent history, described as approximately 365 km per year during construction of Project EnergyConnect Stage 1 and the Eyre Peninsula Link (AEMO, 2026 ISP, 2026). That figure deserves handling fairly. It is not a national capacity ceiling; it is the observed pace on two specific projects, and AEMO's sensitivity assumed a 5 per cent year-on-year increase on it. Against roughly 6,000 km spread to 2050, a rate in that order is not arithmetically impossible.
The problem is not the total. It is the shape of the curve. AEMO says most of the new transmission is needed in the next decade, not spread evenly to 2050 — 3,500 km of committed and anticipated work inside eight years, and 1,660 km of actionable work inside twelve. That is a front-loaded profile against a demonstrated build history in the hundreds of kilometres a year, at a time when coal closures are scheduled and the 2030 renewable target is legislated.
AEMO's own stress test quantifies what slippage looks like. Under Constrained Delivery it assumed committed projects delayed six months, anticipated projects twelve months, and actionable and future projects an average of two years, with costs rising by an average of 30 per cent — attributing the cost rise to competition for skills and equipment as global demand rises, to the delays themselves, and to more costly conditions attached to planning requirements (AEMO, 2026 ISP, 2026). The model recovers the optimal trajectory by 2036. The intervening years are the ones households and businesses would actually live through.
What actually slows a line down
Five things, and they are not equally tractable.
Social licence and landholder consent. This is first because it is the largest and because it is most often described dishonestly. AEMO's own words are careful and worth taking seriously: new energy infrastructure and its construction can affect daily life, local environments and community identity, and there are concerns about the pace and scale of change, the visibility of infrastructure, whether all proposed investments are necessary, and how energy policy settings may evolve (AEMO, 2026 ISP, 2026). Those are not unreasonable concerns and they are not obstruction. A farming family asked to accept a permanent easement is being asked to absorb a specific, local, irreversible cost so that a diffuse, national, long-term benefit can be realised. That is a real asymmetry. It is exactly the sort of asymmetry that compensation schemes, benefit-sharing arrangements and proper consultation exist to address, and where they are done badly the delay that follows is a consequence of the failure, not of the objection. AEMO also identifies a structural gap that anyone who has sat at that kitchen table will recognise: communities need greater clarity on who is responsible for meeting their concerns, because primary responsibility rests with project proponents while approvals are granted by government bodies, to which communities naturally turn when concerns arise.
Planning approvals and easement acquisition. Transmission crosses jurisdictions, land tenures, native title determinations, environmental assessment regimes, and heritage protections. Each is a legitimate check, and each takes time. Several states have created bespoke pathways — the New South Wales infrastructure framework, Victoria's coordinated development arrangements, Queensland's priority assessment process — precisely to compress this, and AEMO notes those pathways provide an alternative to the standard regulatory investment test where they exist, but are not consistently available across the NEM.
Cost escalation. Per-kilometre transmission costs have risen, in AEMO's telling, because of global competition for skills and equipment and because meeting planning requirements has become more costly. Every advanced economy attempting a grid build-out at once is bidding for the same steel, aluminium, cable and engineers.
Supply chains. Large power transformers, high-voltage cable, switchgear and HVDC converter equipment have long lead times and a small number of qualified global manufacturers. A line that is fully approved, funded and physically complete does not carry a single electron without them. This constraint is the least visible in political debate and the least amenable to domestic policy.
Workforce. Transmission construction needs specialised trades in numbers Australia has not required since the state commissions built the original network — in regional locations, simultaneously, in competition with mining, defence and civil construction.
The critique: a plan that is met on paper
Here is the argument, stated as narrowly as the evidence allows, because a narrow argument is harder to dismiss.
The transmission requirement in the 2026 ISP is smaller than the requirement in the 2024 ISP. That reduction is real and it is documented. But the reduction is overwhelmingly composed of lines leaving the plan, not lines entering service. About 1,680 km came off because it is no longer needed, against about 165 km that came off because it reached operation. Some of the "no longer needed" is unambiguously good — batteries and rooftop solar doing work that lines would otherwise do. Some of it is a scenario becoming less ambitious. And AEMO's own footnote names policy changes as one of the two reasons in the same breath as more generation and storage in particular locations, without disaggregating how much of the 1,680 km is attributable to each.
That non-disaggregation is the fair limit on this critique, and it is stated here rather than buried. Nobody outside AEMO can say with confidence how much of the reduction is substitution and how much is retreat, because the published breakdown does not separate them. This article does not claim to know. It claims that the question is material, that the ratio makes it material, and that it is not currently answerable from the public record.
The second limit is more important. Those ratios describe why the remaining plan changed. They are not a delivery scorecard. A project only leaves the plan by reaching operation at the very end of its life, and about 3,500 km of committed and anticipated work is physically underway right now, sitting inside the plan precisely because it has not energised yet. Treating that as failure would be as dishonest as treating the 10,000 km claim as current. The honest statement is narrower: on the composition of the change, the plan is being reduced faster than it is being delivered, and the public record does not let us say how much of that reduction is genuine substitution.
A target met by revising the plan looks identical, in the year it is met, to a target met by building the plan. The difference only appears later.
Where accountability goes
The structural problem is a mismatch of clocks. A transmission project takes the better part of a decade from identification to energisation. HumeLink, a 474 km 500 kV upgrade connecting Project EnergyConnect and the Snowy scheme to Bannaby, has an in-service date of December 2027. Project Marinus Stage 1 is listed at June 2030, the New England REZ network project at January 2034, and CopperString — an 840 km line to Queensland's North West Minerals Province — at 2032 (AEMO, 2026 ISP, 2026, project table). A federal parliamentary term is three years.
That gap means the political system reliably rewards the announcement and rarely encounters the outcome. The minister who announces a concessional loan, the premier who launches a renewable energy zone, and the government that legislates a target will mostly have moved on before the line is energised, before the cost reaches network charges, and before anyone can check whether the assumed substitution materialised. This is not a claim about motives. It is an observation about incentives that would apply to any government of any party facing the same asset lives.
It also means the ISP's biennial revision cycle does quiet work. Each edition is a fresh, defensible, independently reviewed document. Each revision that removes a line is individually justifiable. But a sequence of individually justifiable revisions can, over several editions, describe a smaller and smaller task without any single decision ever being made to attempt less. Nobody has to choose to lower the ambition. It can simply be re-optimised downward, transparently, one plan at a time. Whether that is what is happening is exactly the question the published breakdown does not answer.
The final piece is where the consequences land if the lines are late rather than unnecessary. Under-built transmission does not appear as a blackout on a single dateable evening. It appears as congestion, as renewable generation that is built but curtailed because the network cannot carry it, as coal closures deferred to manage reliability risk, and as higher wholesale prices in regions that cannot import cheaply. AEMO's Electricity Statement of Opportunities has repeatedly flagged reliability risk in the second half of this decade if generation, storage and transmission do not arrive close to schedule as coal exits (AEMO, Electricity Statement of Opportunities, 2024 and 2025 editions). Those costs are diffuse and almost impossible for a household to attribute to a decision — the same property this site's companion analysis of the transition's real cost documents in the network charge itself.
What they got right, and the honest caveats
Several things belong on the credit side of this ledger, and they are not small.
The Integrated System Plan is a genuinely sophisticated planning instrument and an internationally unusual one. Few countries produce a whole-of-market, publicly consulted, independently reviewed, scenario-tested development path to 2050, revise it every two years, publish the modelling assumptions, invite challenge to them, and footnote the ministerial correspondence they relied on. Every number in this article comes from AEMO's own documents. We can have this argument, with real figures, because AEMO publishes them and the Australian Energy Regulator reviews them in public. That is a public asset, built across governments of both persuasions, and it deserves saying plainly.
Roughly 2,800 km was underway at the Draft 2026 ISP, and about 3,500 km is committed or anticipated and underway in the final plan, scheduled for delivery over the next eight years. That is substantial, and more than commentary generally credits. Project EnergyConnect Stage 1 is energised. The Eyre Peninsula Link is built. Four projects that were merely actionable in 2024 have taken investment decisions. The pipeline has moved.
Re-optimisation that avoids unnecessary infrastructure is a saving, not a failure. Every kilometre correctly identified as unnecessary is capital consumers never repay through network charges, and land never encumbered. If storage and rooftop solar genuinely displace 1,680 km of line, that is one of the better outcomes available in this policy area, and treating it as a scandal would be exactly the reflex this site exists to avoid.
The financing architecture is also doing real work: Rewiring the Nation lowers the cost of capital on regulated assets whose costs are ultimately recovered from customers, which is one of the few levers a federal government holds over a system it does not own or build.
And AEMO stress-tested itself. The Constrained Delivery sensitivity exists precisely because AEMO expects delivery risk to be real. Its conclusion is worth quoting in substance: if generation and storage delivery is slower and more costly, the value of the transmission in the optimal development path increases, even though the transmission itself is also slower and more costly to deliver. On AEMO's analysis, delay strengthens rather than weakens the case for building these lines.
The caveats on the credit are equally honest. The 3,500 km underway is scheduled, not delivered, and schedules in this sector have historically moved. The reduction in the plan is documented but not disaggregated between substitution and retreat. Reclassifying a project from actionable to anticipated is an administrative change, transparently made, that nonetheless removes it from the actionability test in that edition. And the demonstrated build rate remains, on AEMO's own reference point, in the hundreds of kilometres a year against a requirement concentrated in the next decade.
The risk to the future
A target met on paper, with the bill arriving after the term that set it
The forward risk is not a blackout and it is not a scandal. It is subtler and, for that reason, more likely. It is that a national transmission plan gets progressively smaller, edition by edition, through a sequence of individually defensible re-optimisations, and that a meaningful share of those reductions turns out to have been assumptions about substitution — batteries, consumer energy resources, better-sited generation, demand-side response — that materialise later or smaller than modelled. If that happens, nothing dramatic occurs on any particular day. Congestion rises. Renewable output is curtailed because the network cannot carry it. Coal closures slip to manage reliability, at a cost. Regional wholesale prices separate. And the network charge on every bill absorbs the difference, quarter by quarter, indefinitely.
Reason about probability rather than certainty here, because certainty is not available. It is entirely possible that AEMO's substitution assumptions prove conservative, that storage and rooftop solar do even more than modelled, and that 6,000 km turns out to have been generous. AEMO has been right about the direction of consumer energy resources more often than its critics have. It is also possible that delivery runs closer to the Constrained Delivery sensitivity — six to twenty-four months of slippage and 30 per cent higher costs — and that the network arrives late rather than never. The plausible range of outcomes is wide and much of it is manageable.
What is not symmetric is who carries each outcome. The upside of an optimistic plan is a modest saving spread thinly across all consumers. The downside is concentrated: on households in regions that cannot import cheap power, on the communities hosting the assets that do get built, and on landholders whose easement negotiations are compressed if the schedule tightens.
The narrow, defensible test to apply from here is not whether AEMO's kilometre number rises or falls in the 2028 ISP. It is whether the next editions publish a disaggregation of the reductions — how much came from substitution that actually happened, how much from policy change, how much from scenario revision — and whether the committed 3,500 km energises inside its stated eight years. Both are checkable. Both will be checked here.