How I used AI for writing about
Wisconsin Electric Usage and Data Centers

Brian K Verheyden

I needed another page for the TaxTimes. Several people have asked me about Data Centers and the power consumption.

I started with a few questions.

  • How much power has Wisconsin generated the past 10 years?
  • How is it created?
  • Have the sources changed?
  • How much power will be required for the 2 large data centers?
  • What will the Annual power consumption be?

I then jumped on the internet to research Wisconsin power consumption, learned I should look at power generation for specifics. The searches led me to US Energy and several links to Badger Institute, reading led me to more data sources.

I created the following table. Made several changes then consolidated the numbers to the categories needed.

I then created a chart of the data, to compare annual changes then went back and got more data, and updated the tables.

At this stage I jumped into MS Word and wrote the following:

Wisconsin’s Energy Picture and what happens with the data centers?

In order to get a handle on the massive future energy usage need for the two data centers being built in Wisconsin, we can to look at Wisconsin’s past electricity usage knowing that it will increase by 25%.

Over the past 10 years Wisconsin’s energy generation has remained relatively stable. In 2015 the total was about 66,389 Megawatts in 2025 that same total was 67,136 Megawatts. Three items of interest during that period. Natural Gas and Coal offset each other; natural gas increased from 20% to 35% and coal decreased from 56% to 35%. The other notable was in 2015 small-scale solar accounted for about 95% of solar to now utility-solar accounts for about 90% of solar. The increase in solar has largely decreased the usage of coal and natural gas.

The following was done with Google Gemini. Several changes needed be made and numbers from AI challenged in regards to annual usage and conversion of 67,136 thousand Megawatt hours and how it was referenced.

 

============ Brian prompt==========

Can you increase the following to 500 words?

Wisconsin’s Energy Picture and what happens with the data

centers?

In order to get a handle on the massive future energy usage need for the two data centers being built in Wisconsin, we can to look at Wisconsin’s past electricity usage knowing that it will increase by 25%.

Over the past 10 years Wisconsin’s energy generation has remained relatively stable. In 2015 the total was about 66,389 Megawatts in 2025 that same total was 67,136 Megawatts. Three items of interest during that period. Natural Gas and Coal offset each other; natural gas increased from 20% to 35% and coal decreased from 56% to 35%. The other notable was in 2015 small-scale solar accounted for about 95% of solar to now utility-solar accounts for about 90% of solar. The increase in solar has largely decreased the usage of coal and natural gas.

 

========================== Gemini Response===============

Wisconsin’s Shifting Energy Landscape: The Data Center Surge

Wisconsin stands at a pivotal crossroads in its utility history. For a decade, the state’s electrical grid has been a model of steady-state reliability, but the arrival of massive data center projects is set to disrupt this long-standing equilibrium. To understand the magnitude of this shift, we must first look at the remarkably stable foundation from which the state is departing.

A Decade of Stability (2015–2025)

Between 2015 and 2025, Wisconsin’s energy generation remained nearly flat. In 2015, the state generated approximately 66,389 Megawatts; by 2025, that figure had only crept up to 67,136 Megawatts. This marginal growth suggests a state that had mastered efficiency and met its industrial and residential needs without the need for aggressive expansion.

However, beneath this “flat” surface, a radical internal transformation occurred within the fuel mix. The most dramatic change was the “great swap” between coal and natural gas. In 2015, coal was the undisputed king of the North, powering 56% of the state. By 2025, coal’s share plummeted to 35%, with natural gas rising from 20% to 35% to fill the gap. This transition was driven by both economic factors and a push for lower-carbon fossil fuels.

The Solar Revolution

The other tectonic shift involved the architecture of renewable energy. A decade ago, solar power was a niche, decentralized effort; small-scale rooftop installations accounted for 95% of the state’s solar output. Today, the script has flipped. Utility-scale solar farms now account for 90% of Wisconsin’s solar generation. This scaling of solar infrastructure has finally begun to provide enough “bulk” power to actively chip away at the necessity of coal and gas, rather than just supplementing them.

The Data Center Challenge

This era of stability is now ending. With the construction of two massive data centers—driven by the global demand for Cloud computing and Artificial Intelligence—Wisconsin is facing a projected 25% increase in total energy demand. Unlike residential neighborhoods or standard manufacturing plants, data centers operate with a “flat” load profile, meaning they pull massive amounts of power 24 hours a day, 365 days a year.

Integrating these facilities into the grid presents several hurdles:

  • Infrastructure Stress: A 25% jump in demand requires more than just new fuel; it requires high-voltage transmission lines and substations capable of handling the concentrated load.
  • Reliability vs. Renewables: While the state has successfully pivoted to utility solar, solar is intermittent. To keep data centers humming at midnight, Wisconsin will need to balance its green goals with firm “baseload” power, likely necessitating more battery storage or extended life for existing gas plants.
  • Economic Impact: While these centers bring investment, the sheer volume of energy they consume can influence local electricity rates and force utilities to accelerate multi-billion dollar capital expenditure plans.

Wisconsin’s energy picture is transitioning from a story of “maintenance” to a story of “massive expansion.” The next decade will not be defined by stability, but by how quickly and cleanly the state can build the capacity to power the digital engines of the future.

 

============ Brian prompt==========

can you rewrite with less positive on solar and less positive on data centers

========================== Gemini Response===============

Wisconsin’s Shifting Energy Landscape: The Data Center Strain

Wisconsin is entering a period of significant utility instability. For the past decade, the state’s electrical grid has operated under a predictable, steady-state model, but the introduction of massive data center projects is poised to break this equilibrium. To understand the risks ahead, we must examine how the state’s stagnant generation capacity will collide with a sudden, massive surge in demand.

A Decade of Stagnation (2015–2025)

Between 2015 and 2025, Wisconsin’s energy generation was characterized by inertia rather than growth. In 2015, the state’s total capacity sat at 66,389 Megawatts; by 2025, that figure had barely moved, reaching 67,136 Megawatts. While this lack of growth was manageable during a period of low demand, it has left the state with very little “breathing room” for new, energy-intensive industries.

During this period, the internal composition of the grid underwent a volatile shift. The state engaged in a forced trade-off between coal and natural gas. In 2015, coal was the primary reliable anchor, providing 56% of the state’s power. By 2025, coal’s share was slashed to 35%, replaced largely by natural gas, which climbed from 20% to 35%. This transition has made the state more dependent on gas pipelines and subject to the price fluctuations of a single fuel source.

The Solar Trade-off

The rise of solar power has further complicated the grid’s reliability. In 2015, solar was a minor, decentralized contributor, with small-scale rooftop installations making up 95% of the total. Today, the landscape is dominated by utility-scale solar farms, which now account for 90% of Wisconsin’s solar output.

While this expansion has been used to justify the decommissioning of coal and gas plants, it introduces the “intermittency problem.” Unlike the fossil fuel plants they replace, solar farms cannot provide power on demand during peak winter evenings or cloudy stretches. This shift has effectively reduced the state’s “firm” power—the energy that is guaranteed to be available regardless of weather conditions—at a time when demand is about to skyrocket.

The Data Center Burden

This fragile balance is now being tested by the construction of two massive data centers. These facilities are projected to increase the state’s total energy need by 25%, a staggering jump for a grid that has seen less than 2% growth in ten years.

The arrival of these data centers presents several critical concerns for Wisconsin residents:

  • Resource Competition: Data centers operate 24/7, pulling massive amounts of “baseload” power. This forces utilities to keep aging fossil fuel plants online longer than planned or build expensive new gas infrastructure to ensure the lights stay on for the data centers, even when the sun isn’t shining.
  • Infrastructure Costs: A 25% increase in load requires an aggressive overhaul of high-voltage transmission lines. The billions of dollars required for these upgrades are often passed down to residential ratepayers, potentially raising monthly bills for families to subsidize the operations of tech giants.
  • Environmental Conflicts: The sheer physical footprint of utility-scale solar required to offset even a fraction of a data center’s usage is enormous, often leading to the loss of valuable agricultural land.

Ultimately, Wisconsin’s energy picture is moving from a decade of stability into a period of forced, expensive expansion. The state must now figure out how to satisfy the immense appetite of these digital warehouses without compromising the reliability of the grid or the wallets of local taxpayers.

 

============ Brian prompt==========

The numbers 66,389 Megawatts is actually 66,389 thousand Megawatts, and 67,136 thousand Megawatts in annual production

========================== Gemini Response===============

That is a vital distinction. Shifting from Megawatts (MW) to Thousand Megawatt-hours (MWh)—or Gigawatt-hours—changes the scale from instantaneous capacity to the actual volume of “work” the state’s power plants performed over the entire year.

At over 67 million Megawatt-hours, the sheer scale of the energy being moved highlights how massive a 25% increase truly is. Adding nearly 17 million MWh of new demand to a grid that has been stagnant for a decade is a significant industrial lift.

Wisconsin’s Shifting Energy Landscape: The Data Center Strain

Wisconsin is entering a period of significant utility instability. For the past decade, the state’s electrical grid has operated under a predictable, steady-state model, but the introduction of massive data center projects is poised to break this equilibrium. To understand the risks ahead, we must examine how the state’s stagnant generation capacity will collide with a sudden, massive surge in demand.

A Decade of Stagnation (2015–2025)

Between 2015 and 2025, Wisconsin’s annual energy production was characterized by inertia rather than growth. In 2015, the state’s total output sat at 66,389 thousand Megawatt-hours; by 2025, that figure had barely moved, reaching 67,136 thousand Megawatt-hours. While this lack of growth was manageable during a period of low demand, it has left the state with very little “breathing room” for new, energy-intensive industries.

During this period, the internal composition of the grid underwent a volatile shift. The state engaged in a forced trade-off between coal and natural gas. In 2015, coal was the primary reliable anchor, providing 56% of the state’s power. By 2025, coal’s share was slashed to 35%, replaced largely by natural gas, which climbed from 20% to 35%. This transition has made the state more dependent on gas pipelines and subject to the price fluctuations of a single fuel source.

The Solar Trade-off

The rise of solar power has further complicated the grid’s reliability. In 2015, solar was a minor, decentralized contributor, with small-scale rooftop installations making up 95% of the total solar output. Today, the landscape is dominated by utility-scale solar farms, which now account for 90% of Wisconsin’s solar generation.

While this expansion has been used to justify the decommissioning of coal and gas plants, it introduces the “intermittency problem.” Unlike the fossil fuel plants they replace, solar farms cannot provide power on demand during peak winter evenings or cloudy stretches. This shift has effectively reduced the state’s “firm” power—the energy that is guaranteed to be available regardless of weather conditions—at a time when demand is about to skyrocket.

The Data Center Burden

This fragile balance is now being tested by the construction of two massive data centers. These facilities are projected to increase the state’s total energy need by 25%, a staggering jump for a grid that has seen less than 2% growth in ten years.

The arrival of these data centers presents several critical concerns for Wisconsin residents:

  • Resource Competition: Data centers operate 24/7, pulling massive amounts of “baseload” power. This forces utilities to keep aging fossil fuel plants online longer than planned or build expensive new gas infrastructure to ensure the lights stay on for the data centers when the sun isn’t shining.
  • Infrastructure Costs: A 25% increase in load requires an aggressive overhaul of high-voltage transmission lines. The billions of dollars required for these upgrades are often passed down to residential ratepayers, potentially raising monthly bills to subsidize the operations of tech giants.
  • Environmental Conflicts: The sheer physical footprint of utility-scale solar required to offset even a fraction of a data center’s usage is enormous, often leading to the loss of valuable agricultural land.

Ultimately, Wisconsin’s energy picture is moving from a decade of stability into a period of forced, expensive expansion. The state must now figure out how to satisfy the immense appetite of these digital warehouses without compromising the reliability of the grid or the wallets of local taxpayers.

 

============ Brian prompt==========

please keep what you have and adjust initial increase is 20% with future increase to 50%

========================== Gemini Response===============

Wisconsin’s Shifting Energy Landscape: The Data Center Strain

Wisconsin is entering a period of significant utility instability. For the past decade, the state’s electrical grid has operated under a predictable, steady-state model, but the introduction of massive data center projects is poised to break this equilibrium. To understand the risks ahead, we must examine how the state’s stagnant generation capacity will collide with a sudden, overwhelming surge in demand.

A Decade of Stagnation (2015–2025)

Between 2015 and 2025, Wisconsin’s annual energy production was characterized by inertia rather than growth. In 2015, the state’s total output sat at 66,389 thousand Megawatt-hours; by 2025, that figure had barely moved, reaching 67,136 thousand Megawatt-hours. While this lack of growth was manageable during a period of low demand, it has left the state with very little “breathing room” for new, energy-intensive industries.

During this period, the internal composition of the grid underwent a volatile shift. The state engaged in a forced trade-off between coal and natural gas. In 2015, coal was the primary reliable anchor, providing 56% of the state’s power. By 2025, coal’s share was slashed to 35%, replaced largely by natural gas, which climbed from 20% to 35%. This transition has made the state more dependent on gas pipelines and subject to the price fluctuations of a single fuel source.

The Solar Trade-off

The rise of solar power has further complicated the grid’s reliability. In 2015, solar was a minor, decentralized contributor, with small-scale rooftop installations making up 95% of the total solar output. Today, the landscape is dominated by utility-scale solar farms, which now account for 90% of Wisconsin’s solar generation.

While this expansion has been used to justify the decommissioning of coal and gas plants, it introduces the “intermittency problem.” Unlike the fossil fuel plants they replace, solar farms cannot provide power on demand during peak winter evenings or cloudy stretches. This shift has effectively reduced the state’s “firm” power—the energy that is guaranteed to be available regardless of weather conditions—at a time when demand is entering an unprecedented vertical climb.

The Data Center Burden

This fragile balance is now being tested by the construction of two massive data centers. These facilities are projected to trigger an initial 20% increase in the state’s total energy need. Even more concerning is the long-term forecast: as these facilities reach full capacity and secondary infrastructure follows, that demand is expected to balloon to a 50% future increase. This is a staggering requirement for a grid that has seen less than 2% growth in ten years.

The arrival of these data centers presents several critical concerns for Wisconsin residents:

  • Resource Competition: Data centers operate 24/7, pulling massive amounts of “baseload” power. This forces utilities to keep aging fossil fuel plants online longer than planned or build expensive new gas infrastructure to ensure the lights stay on for the data centers when the sun isn’t shining.
  • Infrastructure Costs: A 50% jump in load requires an aggressive overhaul of high-voltage transmission lines. The billions of dollars required for these upgrades are often passed down to residential ratepayers, potentially doubling down on costs to subsidize the operations of tech giants.
  • Environmental Conflicts: The sheer physical footprint of utility-scale solar required to offset even a fraction of a data center’s usage is enormous, often leading to the loss of valuable agricultural land.

Ultimately, Wisconsin’s energy picture is moving from a decade of stability into a period of forced, expensive expansion. The state must now figure out how to satisfy the immense appetite of these digital warehouses without compromising the reliability of the grid or the wallets of local taxpayers.

 

============ Brian prompt==========

the increases are annual

========================== Gemini Response===============

Wisconsin’s Shifting Energy Landscape: The Data Center Strain

Wisconsin is entering a period of significant utility instability. For the past decade, the state’s electrical grid has operated under a predictable, steady-state model, but the introduction of massive data center projects is poised to break this equilibrium. To understand the risks ahead, we must examine how the state’s stagnant generation capacity will collide with a sudden, overwhelming surge in annual demand.

A Decade of Stagnation (2015–2025)

Between 2015 and 2025, Wisconsin’s annual energy production was characterized by inertia rather than growth. In 2015, the state’s total output sat at 66,389 thousand Megawatt-hours; by 2025, that figure had barely moved, reaching 67,136 thousand Megawatt-hours. While this lack of growth was manageable during a period of low demand, it has left the state with very little “breathing room” for new, energy-intensive industries.

During this period, the internal composition of the grid underwent a volatile shift. The state engaged in a forced trade-off between coal and natural gas. In 2015, coal was the primary reliable anchor, providing 56% of the state’s power. By 2025, coal’s share was slashed to 35%, replaced largely by natural gas, which climbed from 20% to 35%. This transition has made the state more dependent on gas pipelines and subject to the price fluctuations of a single fuel source.

The Solar Trade-off

The rise of solar power has further complicated the grid’s reliability. In 2015, solar was a minor, decentralized contributor, with small-scale rooftop installations making up 95% of the total solar output. Today, the landscape is dominated by utility-scale solar farms, which now account for 90% of Wisconsin’s solar generation.

While this expansion has been used to justify the decommissioning of coal and gas plants, it introduces the “intermittency problem.” Unlike the fossil fuel plants they replace, solar farms cannot provide power on demand during peak winter evenings or cloudy stretches. This shift has effectively reduced the state’s “firm” power—the energy that is guaranteed to be available regardless of weather conditions—at a time when annual demand is entering an unprecedented vertical climb.

The Data Center Burden

This fragile balance is now being tested by the construction of two massive data centers. These facilities are projected to trigger an initial 20% annual increase in the state’s total energy need. Even more concerning is the long-term forecast: as these facilities reach full capacity and secondary infrastructure follows, that annual demand is expected to balloon to a 50% increase over current levels. This is a staggering requirement for a grid that has seen less than 2% growth over the last ten years.

The arrival of these data centers presents several critical concerns for Wisconsin residents:

  • Resource Competition: Data centers operate 24/7, pulling massive amounts of “baseload” power. This forces utilities to keep aging fossil fuel plants online longer than planned or build expensive new gas infrastructure to ensure the lights stay on for the data centers when the sun isn’t shining.
  • Infrastructure Costs: Supporting a 50% jump in annual load requires an aggressive overhaul of high-voltage transmission lines. The billions of dollars required for these upgrades are often passed down to residential ratepayers, potentially raising monthly bills to subsidize the operations of tech giants.
  • Environmental Conflicts: The sheer physical footprint of utility-scale solar required to offset even a fraction of a data center’s massive annual consumption is enormous, often leading to the loss of valuable agricultural land.

Ultimately, Wisconsin’s energy picture is moving from a decade of stability into a period of forced, expensive expansion. The state must now figure out how to satisfy the immense annual appetite of these digital warehouses without compromising the reliability of the grid or the wallets of local taxpayers.

 

 

 

8 hour project 😉