What is the realistic timeline and economics of the global energy transition — who's ahead, who's behind, and what's blocking progress

How Is the World's Big Switch to Clean Energy Actually Going?

April 12, 2026 | 147 nodes · 524 edges

↓ .md ↓ .db Take this into your AI — the full analysis + graph as markdown, ready to paste into ChatGPT, Claude, Gemini or any AI.

Based on analysis of a 147-node, 524-edge knowledge graph about the timeline, economics, and politics of the global energy transition.

The Short Version

The world is switching from burning fossil fuels to using solar, wind, and batteries. The good news: clean energy is getting dramatically cheaper, faster than almost anyone expected. The tricky news: cheap technology is no longer the main thing slowing the switch down. What’s slowing it down now is a different set of problems — and they’re harder to fix with engineering alone.

There’s One Engine Driving Almost Everything

Imagine a snowball rolling downhill. The bigger it gets, the faster it rolls, and the faster it rolls, the bigger it gets.

Solar power works like that. Every time humans build twice as many solar panels, the cost of making each panel drops by a predictable amount — roughly 20%. This pattern has held for decades and shows no sign of stopping. The knowledge graph calls this the “Solar Wright’s Law Deflation Engine,” and it is by far the most connected concept in the entire map — linked to 54 other things.

This one mechanism touches almost everything else in the story. Cheaper solar makes batteries more useful. Cheaper batteries make electric cars more practical. More electric cars reduce how much oil the world burns. Cheaper solar also makes “green hydrogen” (a fuel made by splitting water with electricity) potentially viable for industries that can’t easily run on batteries, like steel and cement.

So the falling cost of solar panels is not just a story about rooftops. It’s the central gear in a very large machine.

But the Gear Is Stuck

Here’s the thing about gears: they need room to turn. And in many places, that room doesn’t exist.

The two biggest physical blockages are:

The queue to connect to the grid. In the United States, if a solar farm or wind project wants to plug into the electricity network, it has to wait in line — sometimes for years. Right now, the line contains enough projects to power the entire US several times over, all sitting and waiting. This isn’t an engineering problem. It’s a paperwork and permitting problem.

The missing wires. Even when projects get approved, the high-voltage transmission lines needed to move electricity from windy plains or sunny deserts to cities often don’t exist yet. Building them is slow and expensive.

Both of these blockages share a surprising common cause: a US environmental review law (called NEPA) that was designed to protect communities from poorly planned projects but in practice delays clean energy infrastructure by years. The graph shows that NEPA is the single root cause feeding into both major bottlenecks at the same time. That structural coupling matters: fixing one problem without addressing NEPA would leave the other problem fully intact.

China Made Solar Cheap — Including for Itself to Lose Money

China now makes the vast majority of the world’s solar panels, batteries, and electric vehicle components. It got there by building factories at enormous scale, faster than anywhere else on Earth.

Here’s the strange part: China built so many solar panel factories that it now makes far more panels than the world currently installs. Chinese companies are selling panels for less than it costs them to make them. That sounds like a disaster for Chinese manufacturers — and it is. But it’s simultaneously making solar cheaper everywhere on Earth, faster than anyone expected.

The falling cost of Chinese solar panels is, paradoxically, being accelerated by the economic crisis in the Chinese solar industry. The graph shows no edge suggesting this self-destructive mechanism will naturally slow down anytime soon.

Meanwhile, Western countries — particularly the United States — have been pulling back on the policies that supported domestic clean energy manufacturing. Each pullback, the graph shows, strengthens China’s manufacturing advantage further. Five separate connections in the graph point from US policy retreats toward “China gets a stronger monopoly.” No edge runs the other direction.

The Chicken-and-Egg Problem That Has No Easy Fix

There is a circular trap at the heart of solar power that the graph describes as its tightest feedback loop.

When too much solar power is generated at the same time — on a sunny afternoon, say — the price of electricity falls to nearly zero, or even goes negative. That sounds good, but it actually discourages investment in new solar, because investors can’t make money selling something at zero.

The fix is storage: giant batteries that absorb excess solar power during the day and release it at night or on cloudy days. But building enough storage is slow and expensive. And without enough storage, the price crashes keep happening, which keeps discouraging solar investment.

Each half of this problem makes the other half worse. There is no internal solution to this loop. It requires something from outside: a technology breakthrough in long-duration storage, a policy change, or a shift in how electricity markets set prices.

The “We Made It to the Top of the Hill” Moment That Isn’t Actually Safe

One of the graph’s most significant — and most carefully qualified — findings is this: around 2025, global greenhouse gas emissions may have peaked. That means total emissions might not grow anymore, and could begin to fall.

This sounds like a turning point, and structurally, it is. The graph treats it as “the single most important synthesis finding.” But the very same node is immediately surrounded by four separate connections that undermine its significance:

It is “insufficient for” stopping the carbon budget from running out
It is “undermined by” a gap in climate finance for developing countries
It is “threatened by” US policy reversals
It has “not yet” reached the industries most difficult to clean up (steel, cement, shipping, aviation)

Think of it like cresting a hill on a long drive. You’ve reached the top — that matters. But looking ahead, there are still many miles of road, some of it steep, and you’re not sure the car has enough gas.

When AI Is Both the Problem and Part of the Fix

Artificial intelligence uses enormous amounts of electricity. Training large AI models, running data centers, and powering the chips that make modern AI work requires power on a scale that is genuinely changing the electricity grid.

The graph shows AI connected to 43 other concepts — making it the second-most connected concept in the entire map. But unlike the solar engine (which is well-developed and well-understood), AI’s role is structurally unresolved. The graph contains roughly equal numbers of connections showing AI making things worse (more demand, more fossil fuel lock-in, more grid congestion) and connections showing AI making things better (flexible scheduling of computing workloads that could absorb excess solar, new materials discovered by AI that could improve batteries, efficiency gains from better chip design).

Whether AI ends up being a net drag or a net help depends on a race: can efficiency gains from smarter chips and software keep up with the sheer growth in demand for AI computing? The graph does not answer this question. It holds both possibilities open simultaneously.

The Non-Obvious Things the Graph Shows

A few findings are worth highlighting specifically because they run against intuition:

A policy meant to protect European industry may be helping Chinese exports. The EU created a carbon border tax (called CBAM) to discourage imports of goods made with dirty energy. One unintended effect: it preferentially favors solar panels made with clean manufacturing — which describes Chinese panels. The graph shows CBAM inadvertently opening export routes for the Chinese solar overcapacity that is crashing panel prices globally.

The failure of offshore wind is strengthening the case for nuclear. Offshore wind turbines have gotten dramatically more expensive to build in recent years, just as policy support has wavered. Several tech companies and data centers looking for reliable 24/7 clean power have responded by signing contracts with nuclear plants instead. One collapse enabled a different technology’s revival.

Your electric car is simultaneously causing and solving a grid problem. A large fleet of electric vehicles draws heavily on the electricity grid when charging — stressing the same interconnection queues that are already backed up. But those same cars, if managed properly, can feed power back to the grid during peak demand (called “vehicle-to-grid”). The same physical object is both a stressor and a solution, depending on the direction the electricity is flowing.

What the Graph Doesn’t Know Yet

Several of the most important concepts in the graph are listed as placeholders rather than fully worked-out mechanisms. “Carbon Budget Exhaustion,” “Energy Poverty vs. Decarbonization Dilemma,” and “Critical Minerals Becoming a Strategic Weapon” are among the most-connected concepts in the entire map — but they are described only as labels, not as explanations.

Think of them as boxes that say “BAD OUTCOME” with many arrows pointing in but few arrows pointing out. The graph is better at explaining how problems get worse than at modeling how terminal conditions are interrupted or resolved.

Bottom Line

Here is what the structure of the knowledge graph shows, translated into plain terms:

The energy transition is real, and it is being driven primarily by the falling cost of solar power, which is falling faster partly because China is overproducing panels at a loss.

The main things slowing the transition down are no longer technological or economic — they are logistical and political. Permits, wires, and policy stability are now the binding constraints, not the cost of panels or batteries.

A circular trap (solar makes prices crash, crashes discourage investment, investment would fix the crash) has no self-correcting mechanism and needs intervention to break.

US policy retreat amplifies China’s manufacturing advantage, with no reverse mechanism modeled. Western pullback strengthens the very dependency it is ostensibly trying to reduce.

The 2025 global emissions peak is structurally significant but immediately qualified: it does not resolve the carbon budget problem, does not reach the hardest industries, and is threatened by ongoing policy reversals.

AI’s effect on the energy transition is genuinely uncertain and depends on a race between demand growth and efficiency gains that has not yet been decided.

The most powerful single policy lever visible in the graph is permitting reform — specifically, changing the rules that create both major physical deployment bottlenecks at once.

Structural Analysis: Global Energy Transition Knowledge Graph

Key Findings

1. The graph has two structural layers: a high-weight core and a set of stub attractors.

The 147 nodes split into two distinct populations. Roughly 130 nodes carry weights of 7–9.5, with detailed content. The remaining ~15 nodes — including Carbon Budget Exhaustion (w=1), AI Energy Demand Fossil Fuel Lock-In (w=1), Energy Poverty-Decarbonization Dilemma (w=1), Carbon Pricing Implementation Gap (w=1), and Critical Minerals Geopolitical Chokepoint (w=1) — are stubs: low-weight, content-thin nodes that nonetheless function as high-connectivity attractors (28, 43, 23, 25, and 24 edges respectively). In graph terms, these are terminal sinks — many mechanisms flow into them, few pathways lead out. They represent conceptual endpoints rather than explanatory nodes. The structural implication is that the graph’s explanatory density resides in its mechanisms, not in its outcomes.

2. A single mechanism (Solar Wright's Law Deflation Engine) serves as the primary connective tissue of the entire graph.

With 54 connections and weight 9, this node is the most connected AND the most developed node in the graph. It is simultaneously the driver of the transition (enabling cost crossovers, storage economics, EV adoption, green hydrogen viability) and a node under constraint from 10+ other mechanisms (grid queue, transmission deficit, trade barriers, capital costs, NIMBY, NEPA, VRE cannibalization). It does not appear as a terminal node — it is a throughput hub, taking in amplifiers and constraints and distributing effects outward.

3. The US policy reversal creates a unidirectional amplification cascade into China’s structural advantage.

Five distinct edges point from US policy events toward China Clean Energy Manufacturing Monopoly: US Clean Energy Policy Reversal 2025 --[amplifies]-->, US IRA Rollback Investment Shock --[amplifies]-->, US IRA Rollback Investment Collapse --[amplifies]-->, IRA Rollback Stranded Investment Shock --[triggers]-->, and EU Green Deal Political Retreat --[amplifies]-->. No edge in the graph runs in the reverse direction — there is no mechanism shown by which a strengthened Chinese monopoly feeds back into US domestic policy. This is a structural asymmetry: Western policy retreat is modeled as a one-way accelerant.

4. The graph models the energy transition as bottleneck-constrained, not cost-constrained.

In early sections, the cost deflation story (Wright’s Law, LFP batteries, BESS) is the dominant narrative. But the highest-weight bottleneck nodes — Grid Interconnection Queue Crisis (w=8.5, 22 edges), Transmission Infrastructure Deficit (w=8.5), Long-Duration Energy Storage Gap (w=8.5, 27 edges), Hard-to-Abate Sectors Decarbonization Gap (w=8.5) — collectively receive more constraint-type edges than the cost nodes receive enablement edges. The graph’s association structure implies that technology costs have been surpassed by physical deployment infrastructure as the binding constraint.

5. The 2025 Global Emissions Peak Inflection node carries a structurally paradoxical status.

It is labeled “THE single most important synthesis finding” (w=9) but is simultaneously --[insufficient_for]--> Carbon Budget Exhaustion, --[undermined_by]--> Global Clean Energy Finance Gap, --[threatened by]--> US IRA Rollback Policy Reversal Risk, and --[not_yet_affecting]--> Hard-to-Abate Sectors Decarbonization Gap. The peak is real in the graph’s model, but four separate associations immediately qualify or undermine its significance. This node functions as a structural inflection that does not resolve the system’s terminal conditions.

Feedback Loops

Loop A: Solar Price Cannibalization ↔ Long-Duration Storage Gap (tight bidirectional reinforcement)

Solar Price Cannibalization Problem --[triggers, w=9]--> Long-Duration Energy Storage Gap
Long-Duration Energy Storage Gap --[worsens, w=9]--> Solar Price Cannibalization Problem

This is the tightest circular structure in the graph. Solar oversupply drives daytime prices toward zero, creating demand for storage; but until that storage exists at scale, the price signal cannibalizes solar revenue, reducing investment incentives. The loop has no internal resolution mechanism — it requires an exogenous intervention (technology breakthrough, policy, or demand-side flexibility).

Loop B: China Manufacturing → Solar Overcapacity → Deflation → China Manufacturing (reinforcing)

China Clean Energy Manufacturing Monopoly --[amplifies, w=8]--> Solar Wright's Law Deflation Engine
Solar Wright's Law Deflation Engine --[co_activated / LFP Battery Chemistry Cost Revolution --[deepens]]--> China Clean Energy Manufacturing Monopoly
China Solar Overcapacity Deflationary Export --[results_from, w=9]--> China Clean Energy Manufacturing Monopoly
China Solar Overcapacity Deflationary Export --[amplifies, w=9]--> Solar Wright's Law Deflation Engine

China’s manufacturing scale drives down global solar costs, which expands deployment, which drives further Chinese manufacturing scale. The overcapacity condition — economically destructive for Chinese producers — simultaneously reinforces the deflation mechanism. This loop has no negative-feedback component in the graph; it is modeled as uniformly self-reinforcing.

Loop C: AI Demand → Grid Congestion → AI Lock-In → AI Demand (reinforcing)

AI Energy Demand Fossil Fuel Lock-In --[amplifies, w=8]--> Grid Interconnection Queue Crisis
Grid Interconnection Queue Crisis --[amplifies, w=8]--> AI Energy Demand Fossil Fuel Lock-In

Directly stated as a bidirectional amplification. AI demand strains the interconnection queue; grid congestion makes new renewable connections harder to obtain, pushing AI operators toward gas-backed power, increasing the fossil lock-in. A secondary path: Grid Interconnection Queue Crisis --[triggers]--> Hyperscaler Energy Vertical Integration --[responds_to]--> AI Energy Demand Fossil Fuel Lock-In, showing a partial bypass (hyperscalers build private grid solutions) rather than a resolution.

Loop D: Carbon Budget → Fossil Stranded Assets → Political Resistance → Carbon Budget (reinforcing)

Carbon Budget Exhaustion --[amplifies, w=9]--> Fossil Fuel Stranded Asset Threat
Fossil Fuel Stranded Asset Threat --[causes, w=8.8]--> COP30 Belém Fossil Fuel Roadmap Failure
COP30 Belém Fossil Fuel Roadmap Failure --[deepens, w=8.5]--> Carbon Budget Exhaustion

As the carbon budget tightens, the implied writedown of fossil asset values increases, increasing industry resistance to binding agreements; that resistance produces weaker outcomes at multilateral fora, which deepens the budget problem. Fossil Fuel Stranded Asset Carbon Bubble --[accelerated_by]--> 2025 Global Emissions Peak Inflection adds a secondary path where the peak emissions narrative itself accelerates stranded asset risk, which feeds back through financial markets.

Loop E: Just Transition Failure → Policy Reversal → Worse Transition Conditions → Political Failure (reinforcing)

Just Transition Political Economy Trap --[triggered, w=8]--> US IRA Rollback Policy Reversal Risk
Just Transition Political Economy Failure --[enables, w=8]--> US Clean Energy Policy Reversal 2025
US Clean Energy Policy Reversal 2025 --[amplifies, w=8]--> LNG Infrastructure Lock-In Trap [and multiple other nodes]
LNG Infrastructure Lock-In Trap / other nodes --[amplifies]--> Carbon Budget Exhaustion
Carbon Budget Exhaustion / disruption --[amplifies]--> Just Transition Political Economy Failure [indirectly via energy cost pressures]

This loop is less direct than A–D but structurally consequential: transitions that displace workers generate political conditions that reverse the transitions, which extend fossil infrastructure, which create more stranded asset risk and energy volatility, which create more political instability. Carbon Pricing Implementation Gap --[compounds, w=7.5]--> Just Transition Political Economy Failure closes the circuit.

Non-Obvious Connections

1. China solar overcapacity accelerates the transition it financially undermines. China Solar Overcapacity Deflationary Export --[amplifies, w=9]--> Solar Wright's Law Deflation Engine. The economic distress condition in China’s solar sector — firms selling below cost, industry consolidation — is modeled as an accelerant of global cost deflation. The mechanism that is destroying Chinese solar firm margins is simultaneously the mechanism driving the global energy transition forward. No edge captures any negative consequence of this for deployment speed.

2. The EU’s CBAM inadvertently enables Chinese solar overcapacity exports. EU CBAM Carbon Arbitrage Mechanism --[inadvertently_enables, w=6]--> China Solar Overcapacity Deflationary Export. The carbon border tax creates price incentives for importers of carbon-intensive goods to source from low-carbon manufacturers — which is China, for solar panels. The policy intended to reduce carbon arbitrage enables a different form of arbitrage: Chinese overcapacity finds export markets through CBAM-preference effects. This is the single most structurally counterintuitive edge in the graph.

3. AI flexible compute is modeled as a grid stabilizer, not only a grid stressor. AI Compute Demand Flexibility Paradox --[enables, w=7]--> Solar Wright's Law Deflation Engine and --[alleviates, w=7]--> Grid Transmission Infrastructure Bottleneck. AI data centers, which can shift workloads temporally, are shown as able to absorb excess solar generation and reduce transmission congestion. This runs directly counter to the dominant framing of AI as purely additive demand. The same node --[contradicts, w=9]--> AI Energy Demand Fossil Fuel Lock-In, making AI’s net grid effect structurally ambiguous in the graph.

4. Offshore wind’s collapse strengthens the case for nuclear. Offshore Wind Cost-Policy Double Collapse --[enables, w=7]--> Nuclear-AI Hyperscaler PPA Wave and Offshore Wind Policy-Cost Compound Collapse --[amplifies, w=6.5]--> Nuclear-AI Hyperscaler PPA Nexus. The failure of one 24/7 clean power source (offshore wind, which cannot be dispatched at will but is reliable) increases the demand signal for another (nuclear). This is a substitution effect: the collapse of one high-CAPEX, low-LCOE technology redirects capital toward another.

5. NEPA creates two independent bottlenecks simultaneously. NEPA Permitting Paralysis --[causes, w=9.5]--> Grid Interconnection Queue Crisis and --[deepens, w=8.5]--> Transmission Infrastructure Deficit. A single regulatory framework generates both major physical deployment constraints in the US system. This structural coupling means the two bottlenecks are not independent — fixing one without reforming NEPA would leave the other intact.

6. The V2G fleet constrains its own enabling condition. V2G Virtual Battery Fleet --[constrains, w=6]--> Grid Interconnection Queue Crisis and --[constrains, w=5.5]--> Long-Duration Energy Storage Gap. EVs are both a demand stressor (EV-Grid Demand and V2G Feedback Loop amplifies Grid Queue) and a storage solution that partially reduces the storage gap and queue pressure. The same fleet of vehicles simultaneously causes and mitigates the grid congestion problem, depending on the direction of energy flow.

7. JETP and CBAM contradict each other. JETP Climate Finance Credibility Gap --[contradicts, w=7.5]--> EU CBAM Carbon Arbitrage Mechanism. JETPs pay developing nations to accelerate transitions; CBAM taxes their carbon-intensive exports, creating negative incentives for export competitiveness in the interim. These are the EU’s two primary developing-world climate instruments, and they point in opposing directions in the graph.

Central Mechanisms

Solar Wright’s Law Deflation Engine (54 connections, w=9)

Structurally, this node functions as the primary transmission belt of the graph. It receives amplification from China Solar Overcapacity Deflationary Export, LFP Battery Chemistry Cost Revolution, China Peak Emissions Structural Shift, EU RePowerEU Geopolitical Energy Acceleration, and China Clean Energy Manufacturing Monopoly. It distributes effects outward to Fossil Fuel Stranded Asset Threat, Grid-Scale BESS Deployment Wave, Green Hydrogen economics, EV Oil Demand Displacement Curve, Africa Pay-As-You-Go Solar, and 2025 Global Emissions Peak Inflection. Ten separate constraint mechanisms point back at it (NEPA, Grid Queue, Transmission Deficit, Trade Barriers, Capital Costs, NIMBY, IRA Rollback, OBBBA, VRE Cannibalization, Global South Cost). Its high connectivity reflects that it is both cause and effect throughout the graph — every enabling mechanism eventually routes through it, and every bottleneck eventually constrains it.

AI Energy Demand Fossil Fuel Lock-In (43 connections, w=1)

The highest-connectivity node with the lowest weight. This structural anomaly indicates a recently added concept that has been rapidly associated with many pre-existing nodes but not yet fully developed in the graph. It receives amplification from 15+ sources (Grid Queue, LNG Lock-in, OBBBA, NVIDIA Architecture, etc.) and is countered by 10+ sources (Custom Silicon, LoRA/PEFT, Nuclear PPAs, V2G, BESS, AI Compute Flexibility, etc.). It functions as a bifurcation hub: its net effect on the transition depends on the relative magnitudes of its positive (demand-creating) and negative (efficiency/offsetting) associations, none of which are resolved by the graph.

China Clean Energy Manufacturing Monopoly (34 connections, w=9)

Unlike Solar Wright's Law, this is a geopolitically concentrated mechanism — it both enables the transition (cheap panels, batteries, EVs) and creates strategic dependencies. Its amplifiers are almost entirely exogenous to China (US policy reversals, G7 deployment gaps, EU retreat). Its constraints are primarily trade mechanisms (CBAM, US Export Controls) and supply-chain vulnerabilities (Critical Minerals Chokepoint). The graph models it as self-reinforcing: as Western deployment increases demand, Chinese scale deepens; as Western policy retreats, the monopoly deepens further. No internal correction mechanism is modeled.

Carbon Budget Exhaustion (28 connections, w=1)

Pure sink node. Receives flows from 20+ distinct mechanisms; produces outflows to Fossil Fuel Stranded Asset Threat and Climate Uninsurability Property Cascade. Its low weight means it is underdeveloped as a concept — it functions as a label for an outcome, not as a mechanism. The fact that 28 edges route to it suggests the graph treats it as the terminal condition rather than an explanatory variable.

Long-Duration Energy Storage Gap (27 connections, w=8.5)

The most structurally active bottleneck node. It is simultaneously constrained by Grid-Scale Battery LCOE Collapse (which “undermines” it as batteries improve), partially addressed by Quantum-AI Battery Materials Acceleration (could break it), partially bypassed by V2G and Virtual Power Plants, and directly worsened by the Solar Cannibalization loop. It also receives --[enables]--> edges pointing outward: --[enables]--> LNG Infrastructure Lock-In Trap (extended gas demand during storage gap) and --[reduces_need_for]--> Nuclear-AI Hyperscaler PPA Wave. This node is a keystone: resolving it would remove a large cluster of downstream associations.

Tensions & Open Questions

1. Does AI’s net effect on the transition accelerate or retard it? AI Energy Demand Fossil Fuel Lock-In has 43 connections spanning both amplifying and counteracting directions. On the demand side: it amplifies Grid Queue, LNG Lock-in, Carbon Budget. On the mitigation side: it generates Nuclear PPA demand, Custom Silicon efficiency, V2G integration, flexible compute grid services. The graph contains both directions but provides no weighting mechanism to determine which set of effects dominates at system scale.

2. The VRE Price Cannibalization Spiral is simultaneously being exacerbated and mitigated by the same forces. China Fixed-Tariff Abolition 2025 --[exposes]--> VRE Price Cannibalization Spiral. As market-based pricing replaces administered tariffs, cannibalization becomes visible. Industrial Demand Response as Virtual Storage --[mitigates]--> and Virtual Power Plant Grid Flexibility --[partially_solves]--> offer demand-side responses. But NIMBY Local Opposition Siting Crisis --[worsens]--> and Long-Duration Energy Storage Gap --[worsens]--> continue deepening it. The graph does not resolve whether mitigation or deepening prevails.

3. SMR economics: the nuclear revival path is constrained by the same financial structure it needs to displace. SMR Economics Valley of Death --[depends_on]--> Hyperscaler Nuclear PPA Demand Signal for viability, but Grid-Scale Battery LCOE Collapse --[competes_with]--> SMR Economics Valley of Death. If batteries cheapen fast enough, the hyperscaler demand signal may evaporate before SMRs reach commercial scale. The graph holds both trajectories open without resolution.

4. The CBAM creates opposing effects at the same time. EU CBAM Carbon Arbitrage Mechanism --[inadvertently_enables]--> China Solar Overcapacity Deflationary Export while simultaneously --[threatens]--> China Clean Energy Manufacturing Monopoly. These are not sequentially ordered — the enabling and threatening effects run in parallel on different dimensions (export market access vs. long-term competitiveness). The net effect on Chinese manufacturing dominance is structurally indeterminate.

5. The Africa solar leapfrog is modeled as both bypassing and reproducing the poverty trap. Africa Pay-As-You-Go Solar Leapfrog --[bypasses]--> Developing World Cost of Capital Trap and --[bypasses]--> JETP Concessional Finance Structural Failure. But Africa Solar Leapfrog-Poverty Premium Paradox --[created_by]--> Developing World Cost of Capital Trap and --[exemplifies]--> Energy Poverty-Decarbonization Dilemma. Distributed solar access bypasses the finance trap for grid access, but the same cost-of-capital structure reappears as a premium on the distributed system itself.

6. The weight-1 stub nodes have no resolution pathways modeled. Carbon Budget Exhaustion, Energy Poverty-Decarbonization Dilemma, Carbon Pricing Implementation Gap, and Critical Minerals Geopolitical Chokepoint are all high-connectivity sinks with weight 1. They receive many flows but have limited or no outward resolution edges. The graph is structurally more complete in describing how problems deepen than in describing how terminal conditions are resolved.

Hypotheses

H1. NEPA reform would have multiplicative effects exceeding any single technology breakthrough. The graph shows NEPA Permitting Paralysis as the shared root cause of both Grid Interconnection Queue Crisis (w=9.5) and Transmission Infrastructure Deficit (w=8.5). Since both bottlenecks are among the top constraints on Solar Wright's Law Deflation Engine and on 2025 Global Emissions Peak Inflection, a single policy change removing NEPA paralysis would simultaneously relieve the two most connected physical deployment constraints. Testable prediction: permitting reform produces a larger near-term deployment acceleration than any individual storage, generation, or efficiency technology breakthrough.

H2. Resolving the Long-Duration Energy Storage Gap would break at least three independent feedback loops. Loop A (Solar Cannibalization ↔ Storage Gap) would break directly. Long-Duration Energy Storage Gap --[enables]--> LNG Infrastructure Lock-In Trap would weaken, reducing gas lock-in. Long-Duration Energy Storage Gap --[reduces_need_for]--> Nuclear-AI Hyperscaler PPA Wave would reduce the nuclear demand signal. Testable prediction: any LDES technology achieving cost parity would produce observable declines in new LNG infrastructure commitments and in nuclear PPA deal flow within 3–5 years.

H3. US policy retreat increases, not decreases, US import dependence on Chinese clean energy manufacturing. Five graph edges amplify China Clean Energy Manufacturing Monopoly as direct results of US IRA rollback. No edges model a reverse effect (domestic US manufacturing gaining share from policy retreat). Testable prediction: post-OBBBA US solar, battery, and EV component import volumes from China increase year-over-year, contrary to the stated rationale of reducing dependence.

H4. The Just Transition → Policy Reversal loop predicts that energy transitions executed without worker displacement programs will be reversed at higher rates than transitions with explicit just transition mechanisms. Just Transition Political Economy Failure --[enables]--> US Clean Energy Policy Reversal 2025 is modeled as a causal edge. Cross-national comparison: transitions in countries with strong displacement compensation (Germany Kohleausstieg payments, US Appalachian transition programs where implemented) should show lower policy reversal rates than transitions without such programs, controlling for initial fossil fuel dependency.

H5. AI’s net effect on grid decarbonization is determined by the race between inference efficiency gains and training/deployment demand growth. Training-to-Inference Economic Shift --[undermines]--> AI Compute Demand Flexibility Paradox shows that the shift toward inference (which can be temporally flexible) improves grid compatibility. Custom AI Silicon Energy Efficiency Dividend --[counteracts]--> AI Energy Demand Fossil Fuel Lock-In shows hardware efficiency reducing demand. If inference efficiency (Custom Silicon ASIC Economics, LoRA/PEFT) scales faster than total AI compute demand, the net grid effect becomes stabilizing rather than destabilizing. Testable via datacenter grid contracts: if 24/7 firm power contracts (nuclear PPAs, gas PPAs) grow faster than interruptible renewable PPAs, demand-growth dominates; if the reverse, efficiency effects dominate.

H6. The CBAM → China Solar Overcapacity inadvertent enablement will produce a measurable global solar cost deflation faster than CBAM proponents modeled. If CBAM preferencing routes Chinese overcapacity panels to EU-adjacent markets (and indirectly to emerging markets via price arbitrage), the deflationary export mechanism is strengthened by the same policy intended to protect European manufacturers. Testable prediction: global average solar panel spot prices in non-EU, non-US markets fall faster in post-CBAM years than IEA cost trajectory models projected, attributable to Chinese export market redirections.

H7. The stub-node cluster (w=1, high connectivity) represents the graph’s highest-priority development frontier. The five nodes — Carbon Budget Exhaustion, AI Energy Demand Fossil Fuel Lock-In, Energy Poverty-Decarbonization Dilemma, Carbon Pricing Implementation Gap, Critical Minerals Geopolitical Chokepoint — function as outcome containers rather than explanatory mechanisms. Expanding these into full mechanism nodes (with internal sub-structure) would likely reveal additional feedback loops and resolution pathways not currently visible. Their current structure conceals the question of how terminal conditions are reached and what would interrupt them.