What Is Actually Being Proposed?
Before anything else can be evaluated fairly, the proposal itself needs a clear introduction. Here is what Raeden is actually proposing to build in Colorado Springs — and how it got here.
Raeden is proposing to convert a former Intel semiconductor fabrication plant in northwest Colorado Springs into a data center known as Project Taurus. The proposal should be evaluated from the final engineering plans, utility studies, permits, and enforceable operating conditions—not from the label “data center” alone.
Raeden has described the project as supporting up to 50 megawatts of IT load and representing a projected $1 billion investment. Those are developer-stated figures unless and until confirmed in public filings. Household-equivalency comparisons can be useful for scale, but they vary with the assumed household load and whether 50 MW refers only to IT equipment or total facility demand. Raeden has also stated that it is seeking no local tax incentives and would fund directly assigned infrastructure upgrades; that does not by itself establish that there can be no broader system or rate effects.
The proposal would reuse an existing industrial building rather than develop an entirely new greenfield site. Claims about the building’s former electrical rating are relevant context, but a historical service capacity is not the same as a current utility determination that the grid can accommodate a new continuous load. The applicable utility study and interconnection requirements remain the better evidence.
A prior cryptocurrency-mining operation at the site generated documented noise and compliance concerns. That history is relevant because it shaped neighborhood trust, but it is not direct evidence of how a separately owned and differently engineered facility will perform. The prior record should inform permit conditions and monitoring—not substitute for evaluating the new design.
Community members near Chelsea Glenn and surrounding neighborhoods have raised concerns — primarily around noise, water use, power demand, heat impact, and economic benefit. This page reviews each of those concerns against the available evidence, and also flags where the evidence runs out and a real open question remains.
⚠ Disclosure
I have no contract, compensation, or business relationship with Raeden or Project Taurus. I work in the data center industry generally, which is a relevant bias readers should weigh when reading the rest of this page. Facts below are sourced where possible; opinion or interpretation is labeled as such. Corrections and counter-evidence are welcome via the contact section at the bottom.
Community Concerns, Addressed Directly
Six concerns have come up consistently in public meetings and community conversations. Each one is broken into three parts: what's documented, what additional context applies, and — where it exists — what's genuinely still unresolved. Select a topic below.
Data centers contain substantial mechanical and electrical equipment, and some interior equipment areas can require hearing protection. Community impact depends on what reaches nearby properties: equipment selection, enclosure design, setbacks, barriers, tonal and low-frequency content, nighttime operation, and maintenance practices. Colorado Springs has hosted several large facilities without the kind of widely reported controversy seen in some other markets, but this page has not completed an exhaustive audit of every complaint system.
Raeden has publicly stated a target below 50 dBA at the property line and has described acoustic enclosures and enclosed cooling equipment. A single sound level is not enough to evaluate impact without the measurement period, nighttime standard, tonal or low-frequency treatment, modeled receiver locations, and compliance method. The prior mining operation used a materially different equipment arrangement, so its noise record should not be transferred automatically to the proposed design.
The key unresolved questions are whether the acoustic target appears in an enforceable permit, which metric applies, where and when measurements occur, whether tonal penalties are used, who performs post-construction testing, and how residents can trigger an investigation. Those details are more informative than comparisons to rainfall or ordinary office buildings.
Water use depends primarily on the heat-rejection design. A closed internal coolant loop can be paired with dry air-cooled equipment, evaporative towers, or a hybrid system; “closed loop” alone does not determine site water consumption. Fully dry cooling can use little routine process water, while evaporative systems may consume substantial water but can reduce electrical demand under some conditions.
Raeden’s COO has stated that the project would use an air-cooled chilling system with closed piping. If the final design is fully dry-cooled, routine water use for heat rejection could be low. That statement should be checked against the final mechanical schedule, water-service estimate, and permit documents.
This page has not seen an engineering spec or water-utility filing confirming the cooling system design — the basis for this section is a public statement from Raeden's COO, not an independent technical document. If the final design differs from what's been stated, the water-use conclusion would need to change with it.
A 50 MW IT load is significant and may translate to a higher utility-meter load after cooling and electrical losses are included. Reuse of a previously industrial site and the presence of other local data centers are relevant context, but neither replaces a current utility study. Directly assigned upgrade costs, generation and transmission needs, reserve margins, and possible rate effects are separate questions and should be reported separately.
"Ratepayers pay nothing" describes the stated policy for who funds infrastructure upgrades, not a guarantee about every cost category. Nationally, data center growth has been linked to rising power costs in some markets even where individual facilities funded their own direct upgrades — the mechanism is usually grid-wide capacity costs, not a single project's line item.
Has Colorado Springs Utilities published a rate-impact analysis specific to this 50 MW addition, separate from Raeden's own infrastructure commitment? That analysis — if it exists — would be the strongest evidence either way, and this page has not located one.
Large high-availability data centers commonly use multiple standby generators, often diesel-powered, although batteries, gas engines, fuel cells, and other architectures also exist. Operating limits depend on federal rules and the facility’s final air permit. Colorado does not impose a universal 25-hour annual ceiling for every emergency generator.
Existing cumulative pollution burden is an important environmental-justice factor, but it does not make emissions irrelevant elsewhere. Generator count, engine rating, fuel, testing schedule, expected emergency use, dispersion modeling, nearby receptors, and enforceable permit limits should be evaluated for this site.
This page has not reviewed a final site-specific air permit identifying generator count, capacity, fuel, testing allowance, emissions controls, or reporting requirements. Those documents are necessary before annual operating hours or emissions can be stated confidently.
A March 2026 Cambridge study used NASA satellite data on 6,000+ data centers over 20 years and found land surface temperatures increased an average of 3.6°F nearby, with extreme cases up to 16°F up to six miles away. The study is not yet peer-reviewed and other climate scientists have raised the criticism that it may be measuring general suburbanization (roads, parking, commercial buildouts) rather than data-center heat specifically, and that it measures land surface temperature rather than air temperature people actually feel.
Results from very large or densely clustered campuses may not transfer directly to a single proposed facility, but size labels alone do not settle the question. Nearly all electricity consumed by IT and support equipment ultimately becomes heat that must be rejected. Economization can reduce compressor energy; it does not eliminate heat rejection. The relevant evidence is a site-specific thermal and mechanical design.
Raeden's own COO has acknowledged some heat rejection into surrounding air is inevitable — the company has not stated by how much, or whether it would accept third-party heat monitoring as a permit condition. Until the study is peer-reviewed, neither "this doesn't apply" nor "this is a serious risk" is fully settled.
Data centers can generate substantial construction spending and a smaller number of long-term technical, security, facilities, and support roles. Job and fiscal-impact estimates vary by operating model and should be tied to a project-specific economic study. Projected investment, payroll, tax revenue, incentives, and public infrastructure costs should be reported as separate figures rather than combined into a single benefit claim.
The 60–100 figure is smaller than many residents hoped for, and that's a legitimate disappointment, not a misunderstanding to be corrected. The broader case — construction-phase work, property tax revenue, infrastructure investment funded by Raeden rather than the city, and a signal to other investors — is real, but it's a different kind of benefit than direct local employment, and shouldn't be substituted for it in the conversation.
The Walmart $488M figure was a projection at the time it was announced, not an audited result — this page hasn't found a retrospective accounting of whether it was met. The same caveat would apply to any economic-impact figure cited for Raeden before the facility opens.
Colorado Springs Has Done This Before — Repeatedly
Colorado Springs has hosted major corporate data centers for nearly two decades. That history provides useful local context, but it is not a substitute for a complete complaint search, current operating-status verification, or project-specific engineering review.
North Colorado Springs
FedEx — Enterprise Data Center West
FedEx developed EDC-West as a major enterprise facility, then began moving western U.S. infrastructure to third-party capacity and announced a companywide goal of closing its remaining corporate data centers by 2024. No current primary source located for this review identifies EDC-West as an active FedEx production data center. The campus is therefore presented here as a historical and apparently decommissioned FedEx facility, while the building’s present ownership, occupancy, and reuse remain separate questions.
Historical / apparently decommissionedNorthwest Colorado Springs
HP / QTS — Colorado Springs Campus
HP proposed the purpose-built data center in 2009, with development occurring around 2010–2011. The broader corporate campus is older, but that should not be used as the operating age of the data-center building itself.
Useful local design precedentNorth Colorado Springs
Progressive → Novva — One Campus
Progressive developed the facility and Novva later acquired the same campus, with Progressive initially leasing back space. They should not be counted as two separate local data centers. Current commissioned capacity should be distinguished from planned or ultimate buildout.
Deduplicated facility recordColorado Springs
Verizon — Historical Local Presence
Verizon has been associated with a converted industrial data-center facility in Colorado Springs. Current ownership, operating responsibility, capacity, and status should be confirmed through property, company, or permitting records before numerical claims are published.
Include only with current sourcingNorth Colorado Springs
Walmart — Corporate Facility
Walmart’s local facility is a relevant example of a large enterprise data center operating in a commercial corridor. Its existence provides context, but it does not prove that another site with different equipment, load, setbacks, and permit conditions will have the same impacts.
Context, not a guaranteeHow to use the local record responsibly. Colorado Springs has experience hosting large data-center and technology facilities, and the absence of widely reported controversy at several sites is relevant context. It is not proof of zero complaints, zero impact, or future performance. A reproducible complaint search would identify agencies, date ranges, facility aliases, search terms, and records-request results. Local precedent should inform the questions asked of Project Taurus—not replace project-specific engineering and enforcement.
A Prior Mining Operation Is Not the Same Proposal
A significant portion of community anxiety about this site comes from what happened there before. The facts of that prior tenancy and the facts of this proposal are listed side by side below — readers can draw their own conclusion about how much weight the history should carry.
Prior tenant: 3G Venture — Bitcoin Mining
- Improvised rigs using uninsulated shipping containers stacked with commodity ASIC hardware
- Industrial exhaust fans mounted externally with zero acoustic treatment
- Documented 24/7 noise described by neighbors as "outrageous" — covered by local TV news
- Repeatedly cited by Colorado Springs for noise ordinance violations
- Shipping containers toppled in a windstorm during attempted noise mitigation
- Faced foreclosure on a $6.5M loan and filed Chapter 11 bankruptcy
- No engineering design standards, no regulatory compliance pathway
- No connection to Raeden — they are simply purchasing a vacant building
Proposed: Raeden — Project Taurus
- Purpose-engineered facility in a former Intel plant — same conversion Verizon completed successfully here in 2006
- Fully enclosed, insulated, climate-controlled — all cooling is internal
- Developer-stated target below 50 dBA at the property line; final metric and enforcement not yet verified
- Developer-stated air-cooled closed-loop system; final water use depends on filed mechanical design
- $1 billion institutional investment with structured financing
- Full city planning and permitting process followed, with public input, before operations begin
- Developer states that no local tax incentives are requested
- Job and investment estimates are developer-stated; direct upgrade costs and broader grid effects should be separated
How Project Taurus Fits the Colorado Landscape
Colorado has hosted large data centers for decades. The table includes both current facilities and clearly labeled historical examples; former capacity should not be counted as part of today’s active statewide inventory. These comparisons provide scale and design context, but they cannot establish the impacts of this individual project.
| Facility | Location | Capacity | Proximity | Complaints |
|---|---|---|---|---|
| QTS Aurora Hyperscale | Aurora | 177 MW | Suburban corridor, adj. residential | Not established by this page |
| Novva Colorado Springs | COS — Sybilla | 40 MW | Near USAFA & residential | Not established by this page |
| JP Morgan Broomfield | Broomfield | 25 MW | Mixed commercial/residential | Not established by this page |
| Flexential Parker | Parker | 22.5 MW | Suburban commercial | Not established by this page |
| Walmart — COS | COS — Federal Dr | 16.3 MW | Business park; residential grew around it | Not established by this page |
| Verizon — COS | COS — Former semi plant | 10 MW | Similar conversion to Raeden | Not established by this page |
| FedEx EDC-West — Historical | Colorado Springs | Formerly reported at about 8 MW | Business-park setting | Apparently decommissioned as a FedEx corporate data center |
| Project Taurus (Proposed) | COS — Garden of the Gods | 50 MW | Chelsea Glenn nearby | Not yet built |
Scale in context: The QTS Aurora hyperscale campus runs at 177 megawatts — 3.5 times the size of Project Taurus — and public records show no documented complaints from its suburban Aurora surroundings. Relative to what Colorado already operates without an incident record, 50 MW is not an unusual figure for this state. That comparison addresses scale; it doesn't by itself resolve site-specific questions like proximity to Chelsea Glenn, which the concerns above address separately.
The Denver Moratorium — What It Actually Means
Denver's City Council approved a one-year moratorium on new data centers in May 2026. This decision is frequently cited as evidence that data centers are broadly problematic. It is not — and the distinction matters.
The controversy centers on CoreSite's DE3 campus being built in the Elyria-Swansea neighborhood — already ranked among Colorado's most polluted ZIP codes, with documented cumulative air quality burdens, high poverty rates, and insufficient green space. The community's concern is about environmental justice and cumulative pollution burden in an already-stressed neighborhood — not about data center operations being inherently loud or harmful.
The Colorado Springs context is categorically different. The city does not carry Elyria-Swansea's environmental burden. The proposed site is an established industrial corridor. And Colorado Springs already has 20 years of successful, complaint-free data center coexistence to point to.
The xAI Memphis Case, and How It Compares
The data center sector includes cases of documented regulatory non-compliance. This section reviews the most-cited recent case in full, then lists the specific, sourced differences between it and what's being proposed in Colorado Springs.
What Happened in Memphis
Elon Musk's xAI built its "Colossus" AI supercomputer campus in South Memphis, Tennessee beginning in 2024. To power it, the company installed 35 or more natural gas turbines on-site — without obtaining the required air permits from local or federal regulators. A second campus across the state line in Southaven, Mississippi added 59 more turbines, many operating under a "temporary-mobile" classification that placed them outside emissions monitoring entirely.
The impact is documented and serious. The Southaven turbines alone can emit more than 1,700 tons of smog-forming NOx per year — likely the largest industrial source of air pollution in the greater Memphis area. Memphis was already designated an "asthma capital of the nation" in 2024. Residents near the facility face cancer risks at four times the national average. The Southern Environmental Law Center filed suit under the Clean Air Act on behalf of the NAACP. xAI removed unpermitted turbines only after that legal action was threatened.
The differences between this case and what Raeden is proposing in Colorado Springs are listed below. They rest on what Raeden has stated publicly, not on an independent audit of a completed facility, since it hasn't been built yet.
- ⚡Power source. xAI used on-site natural gas turbines as primary power. Raeden has stated that it will draw from the local utility grid and use standby generation. The allowable testing and non-emergency operating hours must be taken from the final site-specific permit, not a universal statewide 25-hour limit.
- 📋Permitting sequence. xAI operated for months without permits before removing the turbines under legal pressure. Raeden is going through Colorado Springs' planning and permitting process before operations begin — though the process is not yet complete.
- 🌬️Air quality baseline. Memphis already bore a disproportionate pollution burden; Colorado Springs does not have a comparable cumulative burden documented at the proposed site.
- 📏Scale of emissions. xAI's turbines ran as continuous primary generation; Raeden's backup generators, at the stated maximum permitted frequency, would emit far less — assuming the facility operates as described.
Standards Used to Evaluate Any Operator
The xAI case and the 3G Venture case both involved regulatory citations and, in xAI's case, litigation. The standards below are applied consistently to evaluate any operator on this page, including Raeden, rather than written specifically in response to any one case.
Permits Before Operation
Facilities affecting air quality, water, or noise should hold required permits before operations begin, not after.
Public Information Before Construction
What's being built, how it will be powered, what it will emit, and what safeguards are in place should be public before ground is broken.
Ongoing, Named Monitoring
Noise, emissions, water, and grid-impact commitments are stronger when paired with ongoing third-party monitoring and a stated process for non-compliance.
Documented Local Impact
Impact on nearby residents — not just facility-level compliance — is the relevant unit of measurement.
Applying these standards to Raeden specifically: the 50 dB limit, the closed-loop cooling design, and the zero-tax-incentive position are public statements, not yet independently verified or written into an enforceable permit. Whether they become binding conditions with named monitoring is listed again in the Open Questions section near the end of this page.
Hyperscale vs. Enterprise: Not All Data Centers Are Created Equal
Some data center concerns are completely valid — and they center on a specific category of facility that is fundamentally different from what is proposed on the project area. Understanding the difference is essential to evaluating this project fairly.
The data center industry is not a monolith. It spans two very different categories of facility, and the legitimate complaints emerging across the country are concentrated almost entirely in one of them.
Conditions That Increase Community Impact
- Very high or rapidly expanding electrical load
- Dense clustering of multiple facilities near homes
- Routine on-site fossil generation or extensive generator testing
- Evaporative cooling in a water-constrained location
- Short setbacks, inadequate acoustic treatment, or strong tonal noise
- Unclear permits, weak monitoring, or expansion before public review
- Large public subsidies without transparent community benefit
- High existing pollution burden or vulnerable nearby receptors
Questions That Matter More Than Labels
- What is the current and ultimate total facility load?
- How is heat rejected, and what is annual water demand?
- What are the modeled nighttime noise levels at actual receptors?
- How many generators are proposed, and what does the air permit allow?
- Who funds direct upgrades, and what broader utility impacts were studied?
- Which commitments are enforceable permit conditions?
- Who monitors compliance, and are results public?
- What remedies exist if modeled performance is not achieved?
Real Cases Where Concerns Were Completely Justified
These are documented cases — not speculation — where data center operators caused genuine harm to communities. They deserve acknowledgment, not dismissal. And they all share a common thread that distinguishes them from Project Taurus.
xAI — Colossus Supercomputer Campus
Elon Musk's xAI installed 35+ natural gas turbines on-site without air permits — then added 59 more across the state line under a "temporary-mobile" exemption that bypassed emissions monitoring entirely. The Southaven turbines alone can emit 1,700+ tons of NOx annually. Residents face cancer risks four times the national average. The NAACP filed suit under the Clean Air Act. xAI removed unpermitted turbines only after threatened with federal litigation.
Hyperscale Data Inc. — Alliance Cloud Services
A digital asset mining facility that converted to AI and high-performance computing operations expanded massively without adequate noise mitigation or community consultation. Residents describe a constant industrial hum "like a vacuum cleaner running all the time" that never stops — day or night. A federal class-action lawsuit filed in 2026 represents 1,300 residential properties within one mile. Crucially, the company removed a row of trees that buffered noise between the facility and nearby homes — without warning or permission. The mayor publicly demanded transparency the company refused to provide. The facility had not applied for required permits at the time of expansion.
Unnamed Hyperscale Facility — Brittany Heights Neighborhood
Beginning in late 2014, residents of the Brittany Heights neighborhood began experiencing constant noise from a neighboring data center's cooling systems — a hum that never stopped, even at night. Residents used noise-cancelling headphones and earplugs to try to cope inside their own homes. Complaints to local authorities went largely unaddressed for years. The situation was significant enough that Chandler adopted zoning code amendments in 2022 making it harder to site data centers, and the city council unanimously rejected a new proposed data center in 2025 — nearly a decade after the original complaints began.
Northern Virginia Hyperscale Corridor
Home to nearly 300 operating data centers — roughly 14% of all data centers worldwide — Northern Virginia represents the most concentrated hyperscale buildout in history. A 2024 Virginia Joint Legislative Audit Review Commission report found that nearly one-third of the state's data centers sit within 200 feet of residentially zoned properties. Residents in Prince William County report noise levels routinely exceeding 60 decibels. Amazon has begun retrofitting facilities with acoustic shrouds in response. The scale and density of development in this corridor is without precedent — and is driving legislative action across the country.
A Pattern Worth Naming
Documented harm is often associated with some combination of scale, proximity, cumulative development, inadequate acoustic design, substantial water use, routine combustion, weak permitting, or poor enforcement. None of those factors is exclusive to a single industry label.
Project Taurus should not be presumed harmful because other data centers have caused harm, and it should not be presumed safe because other local facilities have operated quietly. Its likely impact depends on the final design, operating limits, utility studies, monitoring, and enforceable remedies.
The educational purpose of this comparison is to replace a blanket “all data centers are bad” narrative with a more demanding standard: identify the actual risk pathways, require evidence for each one, and hold the operator to measurable conditions.
Notable Industry Cases
The data center industry includes operators with very different track records. This section documents specific, named cases — what happened, when, and what came of it — without ranking them on a good/bad scale. Readers can draw their own conclusions from the record.
Operators With Public Community Agreements
These are large corporations with real environmental footprints, not perfect actors. What distinguishes this group is that their commitments to host communities are written, specific, and have been carried out over time.
Google has matched 100% of its electricity consumption with renewable energy purchases every year since 2017 — and has committed to running on carbon-free energy 24/7 by 2030. More relevantly to communities: Google publicly pledges to cover 100% of its own power costs and all infrastructure upgrades required by its growth, explicitly stating it will not pass those costs to residential ratepayers or small businesses. In Cedar Rapids, Iowa, Google and QTS negotiated a written community benefit agreement that restricted water use to 20,000 gallons per day, mandated clean energy reliance, set strict noise limits, and committed $400,000 per year to the city for 15 years. That is the model — specific, written, enforceable, and verifiable.
Microsoft has committed to being carbon negative by 2030 — not just neutral — and has contracted 40 gigawatts of new renewable energy since 2020. Over 60 of its data centers globally hold LEED Gold certification. In West Des Moines, Iowa, Microsoft negotiated a community agreement promising to run entirely on renewable energy and investing in local workforce development. Perhaps most notably, Microsoft has begun switching backup generators at some facilities to run on low-carbon fuel blends rather than straight diesel — a specific, measurable operational commitment that goes beyond pledges. These are the kinds of concrete, auditable commitments that should be the baseline expectation for any large facility.
Most data centers receive little public attention, often because impacts are limited, managed, or simply not widely reported. Colorado Springs has several relevant local examples, but ownership, operating status, complaint history, and facility capacity should be verified individually. The absence of prominent headlines is useful context—not proof of perfect compliance or office-like impact. Project Taurus should be evaluated on its own engineering and enforceable conditions.
Cases With Public Pledges but No Verification Mechanism
These involve companies that have made public commitments — sometimes substantial ones — that are not backed by independent enforcement, audit, or penalty structures. Documenting the gap between pledge and mechanism, not assigning intent.
In March 2026, Amazon, Google, Meta, Microsoft, OpenAI, Oracle, and xAI signed the White House "Ratepayer Protection Pledge," committing to build or buy their own power and cover grid upgrade costs rather than passing those costs to residential ratepayers. The pledge is voluntary, with no disclosed oversight mechanism, enforcement authority, or penalty structure for non-compliance. A Consumer Reports investigation found power price nodes near data center concentrations increased by as much as 267% between 2020 and 2025. A CNBC analysis found existing PJM grid market structures continue to route data center infrastructure costs to residential ratepayers under current rate rules, independent of the pledge.
Amazon has signed the ratepayer pledge, has agreed to fund grid upgrades for specific projects including its Falls Township, Pennsylvania facility, and has begun retrofitting some facilities with acoustic shrouds following noise complaints. Northern Virginia — home to roughly 300 data centers, many Amazon-owned — has the highest concentration of data centers in the country. Residents in Prince William County have reported noise levels exceeding 60 decibels. A 2024 Virginia Joint Legislative Audit and Review Commission report found nearly one-third of the state's data centers sit within 200 feet of residentially zoned property. The retrofits followed years of resident complaints.
Georgia estimated $2.5 billion in lost revenue from data center tax breaks in a single year; Virginia estimated $1.6 billion; Texas estimated $1 billion. These figures represent public funds that would otherwise be available for other state and local priorities. A documented pattern in the industry involves companies leveraging their capital scale to negotiate large tax incentive packages from state and local governments, with the option to build elsewhere if terms aren't met. Raeden is requesting zero tax incentives from Colorado Springs, which differs from this industry-wide pattern.
Cases With Documented Permit Violations or Community Harm
These are cases with court filings, regulatory citations, or municipal records documenting harm to surrounding communities. Described here using the same documentary standard as the rest of this page — what happened, sourced, without added characterization.
xAI installed 35+ natural gas turbines at its Memphis Colossus campus without obtaining air permits, then added 59 more across the Mississippi state line under a "temporary-mobile" classification that placed them outside standard emissions monitoring. South Memphis already had one of the highest documented pollution burdens in the country; residents there face cancer risk estimates four times the national average. The Southaven turbines are estimated capable of emitting 1,700+ tons of NOx annually. The Southern Environmental Law Center filed suit under the Clean Air Act on behalf of the NAACP in 2025. xAI removed the unpermitted turbines after the suit was filed.
A facility that began as a Bitcoin mining operation converted to AI and high-performance computing, then expanded its operations and power capacity without the permits required for that expansion. Residents describe a continuous industrial hum from the facility. The company removed a row of trees that had buffered noise between the facility and neighboring homes without prior notice. A federal class-action lawsuit filed in 2026 represents 1,300 residential properties within one mile. The facility had not applied for the permits required for its expansion at the time the lawsuit was filed.
The Brittany Heights neighborhood in Chandler began reporting constant noise from a neighboring data center's cooling systems in late 2014. Residents filed complaints with local authorities; the complaints went largely unaddressed for several years. Chandler adopted zoning code amendments in 2022 making it harder to site new data centers, roughly eight years after the original complaints began, and the city council rejected a new proposed data center in 2025.
Where Project Taurus Stands
Applying the same standard used above to this page's actual subject: Raeden hasn't operated this facility yet, so it doesn't yet have the multi-year audited record that Google and Microsoft do. It also has no documented permit violation or harm case attached to it. It sits in an unrated category — a proposal with stated commitments that haven't yet been tested or made enforceable.
What Raeden has said publicly — grid power, <50 dB at the property line, closed-loop air-cooled chilling, zero tax incentives — would place it alongside the operators with written, enforceable commitments if it operates as described and those commitments become permit conditions with real monitoring. What's missing, compared to Google's Cedar Rapids agreement or Microsoft's West Des Moines agreement, is a published written community benefit agreement specific to this project. That's the single most useful document for residents and the city to ask for before approval.
Summary of This Section
The cases above span a range, from facilities governed by detailed public agreements to facilities associated with documented permit violations or community harm. Many facilities fall between those endpoints, with limited public information about day-to-day performance.
Project Taurus remains unrated. Its proposal differs from some of the industry’s worst cases, but category comparisons cannot establish future performance. The decisive evidence will be the final engineering design, utility and environmental review, enforceable permit conditions, commissioning results, public monitoring, and response if actual performance differs from predictions.
What This Page Doesn't Know Yet
A page that resolves every question in one direction isn't being thorough — it's being one-sided. These are the specific things this page could not verify, listed plainly so residents, reporters, and the city can go ask for them directly.
1. Is there a written, enforceable community benefit agreement? Google's Cedar Rapids and Microsoft's West Des Moines deals are public, written, and specific. No equivalent public document for Raeden's Colorado Springs proposal was found while researching this page. If one exists, it should be linked here; if one doesn't yet, that's worth asking for before approval, not after.
2. Who independently verifies the acoustic limit, water-use design, generator operating conditions, and emissions once the facility is operating? "Committed to" and "permitted to require" are not the same as "monitored by a named third party with public reporting." This page hasn't found a monitoring plan specific to this site.
3. Has Colorado Springs Utilities published a rate-impact study for this specific 50 MW addition? The "ratepayers pay nothing" claim describes who funds direct infrastructure upgrades, not a guarantee about grid-wide rate effects. A site-specific study, if one exists, would settle this more convincingly than precedent from other facilities.
4. What happens if Raeden's stated commitments aren't met? Named consequences, an enforcement body, and a public complaint process would turn these from promises into commitments. This page hasn't seen those mechanisms specified anywhere yet.
5. What does the cooling system actually look like in the final engineering plan? The water-use conclusion on this page rests on a public statement from Raeden's COO, not a filed engineering spec. If the final design changes, that conclusion should be revisited.
Why this section exists: Every other chapter on this page leans on documented local precedent to suggest this project is likely to behave like Colorado Springs' other data centers. That's a reasonable inference from a real track record — but it's an inference, not a guarantee, and the gaps above are where that inference is currently resting on stated intentions rather than verified, enforceable commitments. Closing those gaps is what would actually settle the question, for skeptics and supporters alike.
Have Questions — or Corrections? Let's Talk.
This page is built on the belief that informed communities make better decisions than uninformed ones, regardless of which way that information cuts. Whether you support the project or oppose it, if you have corrections, additional sourcing, a counter-example, or a copy of a document this page hasn't found yet — the Open Questions section above is a direct invitation to send it.
I work in the data center industry, which is the disclosed bias at the top of this page. I'm not a neutral party in the abstract sense, and I've tried to flag where that shows up. If something here reads as one-sided despite that effort, say so — that's useful feedback, not an inconvenience.
info@coloradodc.com