FACT CHECK: The shortage isn’t just a chip—it’s also 20% of the world’s LNG

Taiwan and South Korea

In “Electricity Becomes an Input—Taiwan and South Korea, Spring 2026,” section T2-22 notes that rising electricity costs have affected fabs in Taiwan and South Korea. The section focuses on Asian fabs, with one clear consequence: chip production is competing with residential heating. The sequence retains “Electricity Becomes an Input,” “Taiwan and South Korea,” and “Spring 2026.” A chip also depends on the grid.

A fab cannot treat energy as a minor detail when the cost of electricity threatens the continuity of production. The limitation is clear: the section does not break down cost increases separately by factory or by country. Fabs pay for electricity. The report links Asian fabs, section T2-22, and spring 2026.

The Consequence of “Electricity Becomes an Input”

For “Electricity Becomes an Input” — The Consequence of “Electricity Becomes an Input,” Spring 2026, Block T2-22 shows that Asian fabs are not an isolated constraint but a specific element of “Electricity Becomes an Input” within the material supply chain for chips and data centers. The focus is on the industrial supply chain of “Electricity Becomes an Input,” with a clear consequence: a delay or cost overrun at this stage ripples through to schedules, budgets, or expected volumes. The sequence includes “Electricity Becomes an Input,” “The Consequence of Electricity Becoming an Input,” and “Spring 2026.”

In “Electricity Becomes an Input,” technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is clear: the report does not quantify the exact share of this constraint in the final price of a chip or a data center. The supply chain remains intact. The report links the industrial supply chain of “Electricity Becomes an Input,” block T2-22, and Spring 2026.

2,100 gigawatts on hold

In “The U.S. Backlog—2,100 Gigawatts on Hold, 2026,” the T2-22 block estimates the U.S. interconnection backlog at more than 2,100 gigawatts, exceeding the grid’s total capacity. The report focuses on grid connections, with a clear consequence: data center developers’ timelines are being pushed back. The series continues with “The U.S. Pipeline,” “2,100 Gigawatts in the Pipeline,” and “2026.” The announcement does not power on any servers.

The figure describes projects awaiting a connection slot, not new capacity that is already available. The limit is specific: it does not indicate the volume of projects that will actually be built. The grid tracks projects; the report links grid connections to block T2-22 and the year 2026.

The Consequence of “The American Queue”

For the U.S. pipeline—the consequence of the U.S. pipeline, 2026—Block T2-22 shows that grid connections are not an isolated constraint but a factor specific to the U.S. pipeline within the physical supply chain of chips and data centers. The focus is on the industrial supply chain of “The American Line,” with a clear consequence: a delay or cost overrun at this stage ripples through to schedules, budgets, or expected volumes. The sequence includes “The American Line,” “The Consequence of ‘The American Line,’” and “2026.”

In the U.S. supply chain, technology production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is clear: the report does not quantify the exact proportion of this constraint in the final price of a chip or a data center. The chain retains its link; the report connects the U.S. industrial supply chain, block T2-22, and 2026.

Three to seven years

For timelines that extend beyond a cycle—three to seven years, 2026—block T2-22 estimates the duration of the connection processes to be between three and seven years. The point focuses on electrical connectivity, with a clear implication: the promise of computing capacity for 2026 is hampered by a hardware-related delay. The sequence retains “Timeframes Exceeding One Cycle,” “Three to Seven Years,” and “2026.” The manufacturing schedule is non-negotiable.

Transformers add their own multi-year delivery lead times, which lengthen a chain already dependent on the grid. The distinction is clear: these durations are orders of magnitude provided by the block, not the timeline for a specific project. Transformers dictate the timeline. The document links the electrical connection, block T2-22, and 2026.

The Consequence of Deadlines That Exceed a Cycle

For “Delays That Exceed a Cycle” — The Consequence of “Delays That Exceed a Cycle,” 2026, the T2-22 block illustrates that the electrical connection is not an isolated constraint but rather an element specific to “Delays That Exceed a Cycle” within the material supply chain of chips and data centers. The focus is on the industrial supply chain of Delays That Exceed a Cycle, with a clear consequence: a delay or cost overrun at this stage ripples through to schedules, budgets, or expected volumes. The sequence includes Delays That Exceed a Cycle, The Consequences of Delays That Exceed a Cycle, and 2026.

In “Delays Exceeding One Cycle,” technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is precise: the report does not quantify the exact proportion of this constraint in the final price of a chip or a data center. The chain retains its link: The report connects the industrial chain of “Delays Exceeding a Cycle,” block T2-22, and 2026.

Thirty to fifty percent

Regarding the delay to 2028—30 to 50 percent, projection for 2026—the industry analysis cited in the section projects that 30 to 50 percent of the data center capacity planned for 2026 could be delayed until 2028. The point focuses on planned capacity, with a clear implication: investors must distinguish between a capacity projection and actual commissioning. The sequence retains “The Shift to 2028,” “Thirty to Fifty Percent,” and “Projection for 2026.” A projection does not fill a data center.

The delay stems from the combined constraints of interconnections, equipment, and energy demand. The caveat is clear: this range remains an analyst estimate and not a verified fact. Analysts are pushing back the capacity timeline. The report links planned capacity, the industry analysis cited in the section, and the projection for 2026.

The Consequence of the Delay to 2028

Regarding the postponement to 2028—the consequence of the postponement to 2028, projections for 2026, and the industry analysis cited in the section—it can be observed that planned capacity is not an isolated constraint but rather a factor specific to the postponement to 2028 within the supply chain for chips and data centers. The focus is on the industrial supply chain of “The Shift to 2028,” with a clear consequence: a delay or cost overrun at this stage carries over to schedules, budgets, or expected volumes. The sequence includes “The Shift to 2028,” “The Consequences of the Shift to 2028,” and “Projection for 2026.”

In “The Delay Until 2028,” technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is specific: the report does not quantify the exact share of this constraint in the final price of a chip or a data center. The Chain Remains Intact: This report connects the industrial chain from “The Delay Until 2028,” the industry analysis cited in that section, and the projection for 2026.

After the 2026 strikes

Helium Under Rationing — After the 2026 strikes: In 2026, Block T2-22 reports that the spot price of helium doubled following strikes that disrupted Qatari production. The report focuses on industrial helium, with a clear consequence: Taiwanese and South Korean factories are rationing a gas required for their processes. The sequence includes “Helium Under Rationing,” “After the 2026 Strikes,” and “In 2026.” This unassuming gas becomes a constraint.

Helium is used to cool wafers and detect leaks—two functions that link supply to yield. The timeframe is specific: the report covers March–April, and the block does not provide updated prices for August. Helium is slowing down fabs. The report links industrial helium, block T2-22, and the year 2026.

The Consequences of Helium Rationing

Regarding “Helium Under Rationing” — The Consequences of Helium Under Rationing, in 2026, the T2-22 block reveals that industrial helium is not an isolated constraint but a specific element of “Helium Under Rationing” within the supply chain for chips and data centers. The focus is on the industrial supply chain for “Helium Rationing,” with a clear consequence: a delay or additional cost at this stage affects schedules, budgets, or expected volumes. The sequence includes “Helium Rationing,” “The Consequences of Helium Rationing,” and “In 2026.”

In “Helium Rationing,” technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is specific: the report does not quantify the exact proportion of this constraint in the final price of a chip or a data center. The chain remains intact. The report connects the industrial chain of “Helium Under Rationing,” the T2-22 block, and “In 2026.”

In 2026

For Qatar’s One-Third Share — In 2026, Block T2-22 will account for approximately one-third of global helium production. The focus is on Qatar’s share, with a clear consequence: a local disruption takes on global proportions for industrial buyers. The sequence retains “Qatar’s one-third share,” “In 2026,” and “In 2026.” This concentration amplifies the impact.

This concentration explains why an incident at a hub can extend beyond the region’s borders and affect chip supply chains. The limit is precise: the share is given as approximately one-third, not as a percentage audited to the hundredth. Qatar accounts for one-third. The report links Qatar’s share, the T2-22 block, and “In 2026.”

The Consequence of the Qatari Third

For the Qatari third—the consequence of the Qatari third—in 2026, the T2-22 block reveals that the Qatari share is not an isolated constraint but a specific element of the Qatari third within the hardware supply chain for chips and data centers. The focus is on the industrial supply chain of the Qatari third, with a clear implication: a delay or additional cost at this stage affects schedules, budgets, or expected volumes. The sequence includes the Qatari third, the implications of the Qatari third, and in 2026.

In the Qatari third party, technology production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is specific: the report does not quantify the exact share of this constraint in the final price of a chip or a data center. The supply chain retains its link. The report connects the industrial supply chain of “The Qatari Third,” the T2-22 block, and “In 2026.”

In the spring of 2026

For The Price of Bromine — In the spring of 2026, in 2026, block T2-22 sets the price of bromine at $12,000 per metric ton. The focus is on bromine, with a clear consequence: circuit etching and flame retardants absorb additional cost pressure. The sequence includes “The Price of Bromine,” “In the Spring of 2026,” and “In 2026.” Each input factor contributes to the total.

The material is used in specific applications within the electronics industry; its price increase is therefore not solely attributable to energy costs. The limitation is clear: the block does not provide historical data or August prices for this metric ton. Bromine reaches $12,000. The report links bromine, the T2-22 block, and the year 2026.

The Consequence of the Price of Bromine

For “The Price of Bromine — The Consequence of the Price of Bromine, in 2026,” the T2-22 block demonstrates that bromine is not an isolated constraint but a factor specific to the price of bromine within the material supply chain for chips and data centers. The focus is on the industrial supply chain for “The Price of Bromine,” with a clear implication: a delay or additional cost at this stage affects timelines, budgets, or expected volumes. The sequence includes “The Price of Bromine,” “The Consequences of the Price of Bromine,” and “In 2026.”

In “The Price of Bromine,” technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is clear: the report does not quantify the exact share of this constraint in the final price of a chip or a data center. The chain maintains its link: The report connects the industrial chain from The Price of Bromine, the T2-22 block, and In 2026.

In 2026

Regarding ICL and market concentration — In 2026, the T2-22 panel estimates that the Israel-based ICL Group controls nearly 40% of the global supply of bromine. The analysis focuses on the ICL Group, with a clear implication: a concentrated market exposes manufacturers to greater dependence on a single player and a single production region. The sequence retains “ICL and market concentration,” “In 2026,” and “In 2026.” A single supplier can influence thousands of supply chains.

Here, bromine links the chemical properties required by the industry to a narrow supply structure. The threshold is precise: nearly 40% is an estimate by the bloc and not a figure certified by the company. ICL concentrates bromine. The report links ICL Group, the T2-22 bloc, and “In 2026.”

The Implications of ICL and the Concentration

For ICL and the concentration — The consequence of ICL and the concentration: In 2026, Block T2-22 reveals that the ICL Group is not an isolated constraint but a factor specific to ICL and the concentration within the material supply chain for chips and data centers. The focus is on the industrial supply chain of ICL and the concentration, with a clear consequence: a delay or additional cost at this stage affects schedules, budgets, or expected volumes. The sequence includes ICL and the concentration, The Consequences of ICL and the Concentration, and In 2026.

In “ICL and the concentration,” technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is specific: the report does not quantify the exact share of this constraint in the final price of a chip or a data center. The supply chain retains its link. The report connects ICL’s industrial supply chain and the merger, the T2-22 block, and “In 2026.”

April 20, 2026

For “Copper, Plain and Simple” — April 20, 2026: In 2026, Manufacturing Dive, citing Omdia’s analysis, describes a shortage of electricity, copper, and gas—all critical for semiconductors and AI. The focus is on copper, with a clear implication: infrastructure construction also depends on this metal, which is in high demand. The sequence retains “Copper, Plain and Simple,” “April 20, 2026,” and “In 2026.” Cables face their own shortage.

Copper connects networks, equipment, and facilities; a shortage of it prevents the crisis from being reduced to a single component. The limitation is clear: the cited article is a sector-specific analysis, not a comprehensive public assessment of all reserves. Omdia places copper at the center of the report, which links copper to Manufacturing Dive, citing the Omdia analysis and “In 2026.”

The Implications of “Copper Unveiled”

For “Copper Without Detours”—the consequence of “Copper Without Detours”—in 2026, Manufacturing Dive, citing Omdia’s analysis, shows that copper is not an isolated constraint but a factor specific to “Copper Without Detours” in the material supply chain for chips and data centers. The focus is on the industrial supply chain for “Copper Without Compromise,” with a clear implication: any delay or additional cost at this stage ripples through to timelines, budgets, or expected volumes. The sequence includes “Copper Without Compromise,” “The Consequences of Copper Without Compromise,” and “By 2026.”

In “Copper Without Detours,” technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is clear: the report does not quantify the exact share of this constraint in the final price of a chip or a data center. The supply chain retains its link. The report connects the industrial supply chain from Le cuivre sans détour—with Manufacturing Dive citing Omdia’s analysis—to En 2026.

July 16, 2026

Plus: Rare Earths — July 16, 2026. In 2026, Reuters reports that China’s controls on rare earths are putting additional pressure on Western technology supply chains. The focus is on rare earths, with a clear implication: the tensions already described regarding natural gas, helium, and bromine do not cancel each other out. The sequence retains “Rare Earths, Plus,” “July 16, 2026,” and “In 2026.” A supply chain can withstand multiple disruptions.

The constraints add up because they affect different inputs of the same technological production. The limitation is clear: the section refers to Chinese controls as an additional stressor without quantifying their direct effect on each factory here. Rare earths add to the pressure. The report links rare earths, Reuters, and “In 2026.”

The Consequence of “Rare Earths Plus”

For “Rare Earths Plus”—the consequence of “Rare Earths Plus”—in 2026, Reuters highlights that rare earths do not constitute an isolated constraint but rather a factor specific to “Rare Earths Plus” within the material supply chain for chips and data centers. The focus is on the industrial supply chain for “More on Rare Earths,” with a clear implication: a delay or additional cost at this stage affects timelines, budgets, or expected volumes. The sequence includes “More on Rare Earths,” “The Consequences of More on Rare Earths,” and “In 2026.”

In “Rare Earths Plus,” technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is specific: the report does not quantify the exact proportion of this constraint in the final price of a chip or a data center. The supply chain retains its link; the report connects the industrial supply chain of “Rare Earths Plus,” Reuters, and “In 2026.”

July 9, 2026

More Expensive Transformers — July 9, 2026: In 2026, Reuters reports that analysts anticipate a 4% to 10% year-over-year increase in transformer costs. The focus is on transformers, with one clear consequence: connecting new data centers may cost more even before a single watt is delivered. The sequence retains “More Expensive Transformers,” “July 9, 2026,” and “In 2026.” The equipment is taking the brunt of the delay.

A transformer is critical equipment: its price and delivery time affect the same commissioning phase. The caveat is clear: this is an analyst forecast, not an increase that has already materialized for every buyer. Reuters quantifies the transformer; the report links transformers, Reuters, and “In 2026.”

The Consequence of More Expensive Transformers

Regarding “More Expensive Transformers” — The consequence of “More Expensive Transformers” in 2026: Reuters highlights that transformers are not an isolated constraint but a specific factor within the chip and data center supply chain. The focus is on the industrial supply chain for “More Expensive Transformers,” with a clear consequence: a delay or additional cost at this stage affects timelines, budgets, or expected volumes. The sequence includes “More Expensive Transformers,” “The Consequence of More Expensive Transformers,” and “In 2026.”

In “More Expensive Transformers,” technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is specific: the report does not quantify the exact share of this constraint in the final price of a chip or a data center. The report traces the industrial chain from “More Expensive Transformers,” Reuters, and “In 2026.”

Horizon 2030

From 24 to 110 gigawatts — Horizon 2030: In 2026, Wood Mackenzie, cited in the report, forecasts U.S. data center capacity to reach 110 gigawatts, up from approximately 24 gigawatts currently. The focus is on U.S. capacity, with a clear implication: the projected demand makes the electricity issue more significant than software growth alone. The sequence includes “From 24 to 110 gigawatts,” “Horizon 2030,” and “In 2026.” AI also requires power plants.

The same analysis estimates that this trajectory would consume eight times more electricity than electric vehicles over the period. The caveat is clear: this 2030 scenario is a forecast and does not guarantee either investments or grid connections. Wood Mackenzie targets 110 gigawatts; the report links U.S. capacity, citing Wood Mackenzie in the section, and the year 2026.

The Implications of 24 to 110 gigawatts

For 24 to 110 gigawatts — The consequence of 24 to 110 gigawatts: In 2026, Wood Mackenzie, cited in the section, shows that U.S. capacity is not an isolated constraint but a factor specific to 24 to 110 gigawatts within the supply chain for chips and data centers. The focus is on the industrial supply chain of “From 24 to 110 Gigawatts,” with a clear implication: a delay or cost overrun at this stage ripples through to schedules, budgets, or expected volumes. The sequence includes “From 24 to 110 Gigawatts,” “The Consequences of ‘From 24 to 110 Gigawatts,’” and “By 2026.”

In the “24 to 110 gigawatts” section, technological production depends on simultaneous stages: energy, equipment, materials, and connectivity are not interchangeable. The limitation is specific: the report does not quantify the exact proportion of this constraint in the final price of a chip or a data center. The supply chain remains intact. The report links the industrial supply chain from From 24 to 110 Gigawatts, Wood Mackenzie (cited in the report), and In 2026.

The shortage is therefore not simply a lack of components. 20% of LNG, more than 2,100 gigawatts on hold, rationed helium, and bromine priced at $12,000 all point to distinct constraints that converge on the same production process. Projected capacity alone does not produce any chips. Only an actual connection does.

Manufacturing Dive summarizes the issue as a strategic reconfiguration in which scarcity becomes the profitable product. This phrase is attributed to its Omdia analysis; it does not allow us to claim that the scenarios for 2028 or 2030 are a foregone conclusion. Historical figures are more reliable than a promise of capacity.

By Maxime Marquette, columnist