AZIO AI HOLDINGS, INC.

There were no material changes from the risk factors previously disclosed in our 2025 Annual Report, except as described below.

We have a limited operating history in our AI data center, GPU compute, and digital power business, and our strategy in this area is unproven at scale.

Following the Azio Acquisition, our business now encompasses the development, ownership, and operation of artificial intelligence data centers, enterprise GPU compute infrastructure, digital power solutions, and digital asset mining operations, serving enterprise, institutional, government, and technology customers across domestic and international markets. This business model requires us to execute across a complex and interdependent set of operational, technical, financing, and commercial challenges that our combined organization has not previously faced, including large-scale data center site selection and development, power procurement and interconnection, GPU procurement and systems integration, and the sale and distribution of enterprise GPU systems and server infrastructure. Because we have only recently adopted this expanded business strategy, we have a limited track record on which investors can evaluate our ability to execute it, and our historical financial results, to the extent they reflect our prior business, are not indicative of our future performance under this new strategy. We cannot assure you that our strategy will succeed, that we will achieve or sustain profitability in this business, or that we will be able to compete effectively against more established participants.

Our business strategy is evolving, and changes to that strategy, or our failure to execute it successfully, could adversely affect our results of operations and the market price of our common stock.

Our business strategy for our AI infrastructure, GPU compute, digital power, and digital asset mining operations continues to evolve as market conditions, customer demand, technology, and capital availability change. We may adjust our strategy, including the mix of AI hosting, GPU resale and distribution, compute-as-a-service, colocation, and digital asset mining activities we pursue, or the geographic markets and customer segments we target, in ways that are difficult to predict and that may not prove successful. Strategic shifts of this kind can result in unanticipated costs, the write-down or abandonment of assets or business lines, workforce reductions or realignments, and diversion of management attention, any of which could adversely affect our business, financial condition, and results of operations.

Our business depends on our ability to develop and operate large-scale, power-intensive AI data center campuses, and any delay, cost overrun, or failure in that development could materially harm our business.

We have announced plans to develop an AI infrastructure campus in Texas designed to support a substantial amount of power capacity, and our business strategy depends on our ability to complete such development, and any future campuses, on time and within budget. The development of large-scale AI data center campuses requires substantial upfront and ongoing capital investment, securing suitable sites and power interconnection capacity, engineering and construction expertise, and coordination with utilities, grid operators, municipalities, and equipment suppliers. Construction of this kind of project is subject to significant risk of delay or cost overrun as a result of, among other things, permitting and utility interconnection delays, supply chain constraints for critical equipment (including transformers, switchgear, cooling systems, and networking equipment), labor availability and cost, weather and other force majeure events, and unanticipated site, environmental, or engineering issues. We may also be exposed to the risk that a general contractor, subcontractor, or key supplier involved in our construction projects experiences financial difficulty, insolvency, or performance failure. Any failure to complete our planned infrastructure on the anticipated timeline or budget could impair our ability to meet customer commitments, expose us to contractual penalties or termination rights, damage our reputation, and have a material adverse effect on our business, financial condition, and results of operations. Comparable data center operators have identified similar construction, financing, and site-concentration risks in their public disclosures.

We may pursue development at a limited number of sites, and geographic concentration of our operations exposes us to regional regulatory, market, and physical risks.

To the extent our AI infrastructure operations become concentrated in a limited number of geographic markets, including Texas, we are and will be particularly exposed to the regulatory environment, wholesale power market structure and rules, grid operator requirements, weather patterns, and general economic and political conditions of those markets. Adverse regulatory, market, or physical developments affecting any single region in which we have concentrated operations could have an outsized and disproportionate effect on our overall business, financial condition, and results of operations compared to a more geographically diversified operator.

We depend on a concentrated customer and counterparty base for our GPU sales and infrastructure agreements, and the loss of, or failure to perform by, one or more of these counterparties could materially harm our results.

We have announced an agreement with a single counterparty covering the purchase of up to 128 NVIDIA HGX B300 GPU systems, which the Company has estimated could represent an aggregate hardware value of approximately $77 million if all contemplated systems are purchased, although actual revenues may differ materially and remain subject to future purchase orders and customary conditions. Our GPU sales, compute leasing, and infrastructure businesses may depend on a small number of customers and counterparties for a disproportionate share of revenue in any given period. Any failure by a significant customer or counterparty to submit anticipated purchase orders, to make timely payment, or to perform its other obligations to us, whether due to its own financial condition, changed business needs, breach, insolvency, or other factors, could have an outsized adverse effect on our revenue, cash flow, and results of operations. In addition, because several of our current and potential customers may be early-stage, private, or otherwise non-investment-grade counterparties, we are subject to heightened counterparty credit risk, and our ability to collect amounts owed to us, or to enforce contractual remedies, may be limited.

Our customers may terminate agreements with us, decline to renew, or demand damages if we fail to meet contracted service levels, delivery timelines, or performance commitments.

Our AI hosting, colocation, and GPU supply arrangements may impose significant service-level, delivery, indemnity, and liability obligations on us. Our ability to meet these commitments depends on the timely completion of construction and infrastructure deployment projects, the continuous availability of power, cooling, and connectivity, the uninterrupted operation of our hardware systems, and our ability to source contracted hardware on the agreed timeline. Any failure to meet committed service levels, delivery dates, or performance standards, whether due to construction delays, equipment failures, supply chain disruptions, power curtailment, or other factors, could result in customer terminations, non-renewals, reputational harm, and significant financial liability, including damages, penalties, or the obligation to provide credits or refunds.

Our commercial partnerships with key suppliers and service providers are important to our business, and the loss of, or deterioration in, any of these relationships could adversely affect our operations.

We rely on relationships with key hardware suppliers, connectivity providers, and infrastructure partners to support our AI infrastructure business, including arrangements for the supply of enterprise GPU systems (including through reseller or distribution relationships) and for enterprise fiber connectivity to support low-latency AI training, inference, and cloud computing operations. The loss of, or a material adverse change in the terms of, any such relationship could disrupt our ability to source hardware, deliver contracted services, or maintain the network performance our customers require, and could require us to seek alternative arrangements on less favorable terms, at increased cost, or on a delayed basis, any of which could adversely affect our business, financial condition, and results of operations. We may have limited or no ability to control the business decisions of these third parties, including their pricing, allocation, or continued willingness to do business with us. These supplier and partner dependencies are in addition to the risks associated with our strategic relationships and joint ventures, including our former pilot partnership with Legacy Azio, described in our 2025 Annual Report.

Our engagement in enterprise GPU sales and distribution exposes us to inventory, pricing, warranty, and product-related risks that differ from our infrastructure hosting and compute services businesses.

In addition to developing and operating AI data centers, we source, integrate, and distribute enterprise-grade GPU hardware and server infrastructure. This activity exposes us to risks not present, or present to a lesser degree, in our hosting and compute services operations, including the risk of rapid price declines or volatility in GPU hardware pricing, the risk that hardware we acquire for resale becomes technologically obsolete or unsellable before it can be delivered or monetized, inventory and working capital risk, warranty and product liability exposure, exposure to defective or non-conforming products from our suppliers, and the risk that key suppliers limit, delay, condition, or discontinue our allocation of GPUs or other critical components. We may also be exposed to fluctuations in foreign currency exchange rates, import/export and customs requirements, and tariffs to the extent our GPU sourcing or distribution involves cross-border transactions. Any of these risks could result in inventory write-downs, margin compression, contract disputes, or an inability to fulfill customer orders, which could materially and adversely affect our results of operations. See also “Expansion of our business strategy into the AI infrastructure market could increase competitive, operational, legal and regulatory risks to our business in ways we cannot predict” in our most recent Annual Report on Form 10-K.

We operate in a rapidly evolving industry, and our business model, revenue streams, and cost structure may not develop as expected.

Our AI infrastructure business is intended to generate revenue from multiple potential sources, including AI data center hosting and colocation, GPU compute leasing (compute-as-a-service), enterprise GPU and server sales and distribution, power hosting, and digital asset mining, as well as through structured partnerships such as joint ventures, power purchase agreements, and equipment financing arrangements. There can be no assurance that any or all of these revenue streams will develop as we anticipate, that the relative mix of revenue among them will be favorable, or that our cost structure across these different activities will allow us to achieve or sustain profitability. Failure of one or more of these business lines to perform as expected could have a disproportionate effect on our overall results given the early stage of our operations in this sector.

We may seek to expand internationally, and our international operations, if any, would subject us to additional legal, regulatory, operational, and currency risks.

To the extent we pursue international expansion of our AI data center, GPU compute, or digital power businesses, we would be subject to a variety of additional risks not present in our current U.S.-focused operations, including compliance with foreign laws and regulations (including data protection, export control, and anti-corruption laws), foreign currency exchange rate fluctuations, differing labor and employment practices, political and economic instability, difficulty enforcing contracts and collecting receivables, and the burden of complying with multiple, potentially conflicting, legal and regulatory regimes. Any of these risks could increase the cost and complexity of our international operations or limit our ability to expand internationally as planned.

Our operations depend on the availability of specialized GPU and related hardware from a limited number of suppliers, and shortages, allocation constraints, or price increases could impair our ability to grow our business.

The market for high-performance GPUs, including current-generation NVIDIA systems, and other specialized AI infrastructure equipment, is characterized by a limited number of suppliers, long lead times, significant global demand, and periodic allocation constraints. We compete for supply against much larger and better-capitalized companies, including hyperscale cloud providers. Any shortage, allocation restriction, price increase, or delay in the delivery of GPUs or related hardware could delay our infrastructure buildouts, impair our ability to fulfill customer or resale commitments, increase our costs, and place us at a competitive disadvantage relative to better-capitalized competitors with preferred supplier relationships.

Rapid technological change in AI hardware could render our GPU and data center infrastructure obsolete, requiring significant additional capital expenditures.

The AI infrastructure market is characterized by rapid and unpredictable technological change, including the frequent introduction of new generations of GPUs, networking infrastructure, and cooling technologies. If we are unable to upgrade, integrate, or resell our hardware in a timely and cost-effective manner, our existing infrastructure and inventory may become less competitive or economically unviable, we may be unable to attract or retain customers who demand state-of-the-art infrastructure, and we may be required to make substantial additional capital expenditures earlier than planned, any of which could materially and adversely affect our competitive position, financial condition, and results of operations. In addition, advances that reduce the amount of compute required to train or run AI models (for example, more efficient model architectures or open-source alternatives) could reduce demand for the type or quantity of infrastructure we provide.

Failures of critical systems or infrastructure at our data centers, or defects affecting our concentrated GPU hardware fleet, could result in significant operational disruptions, customer losses, and financial liability.

Our data center operations depend on the continuous functioning of critical systems, including power infrastructure, cooling systems, networking equipment, and compute hardware, many of which have long replacement lead times and may represent single points of failure. Because our GPU fleet may be concentrated in a limited number of hardware generations or suppliers, a serial defect or systemic issue could cause widespread underperformance or outages across multiple sites simultaneously. Any failure of these systems due to, for example, equipment malfunction, human error, natural disaster, cyberattack, or utility disruption, could result in material service interruptions, breach of customer service-level commitments, significant remediation costs, customer loss, and lasting reputational harm.

We may be unable to obtain sufficient insurance coverage for our operations, and losses that are uninsured or exceed our coverage could materially harm our financial condition.

Our AI data center, GPU compute, and digital power operations involve substantial capital assets and exposure to property damage, business interruption, equipment breakdown, cyber incidents, and third-party liability claims. We may not be able to obtain insurance coverage at commercially reasonable rates, or at all, for some or all of these risks, and our existing or future coverage may contain exclusions, sublimits, or deductibles that leave us exposed to material uninsured losses. A significant uninsured or underinsured loss could have a material adverse effect on our business, financial condition, and results of operations.

We operate in the digital asset mining business in addition to our AI infrastructure business, and our ability to allocate power and computing capacity between these uses, as well as the volatility of digital asset markets, subjects us to additional risk.

Our business includes digital asset mining operations alongside our AI data center and GPU compute businesses, and we may retain flexibility to deploy power and compute hardware between AI workloads and digital asset mining depending on power availability and relative market yields. Our results of operations may accordingly be affected by the substantial price volatility, regulatory uncertainty, and evolving legal treatment of bitcoin and other digital assets, independent of demand for our AI infrastructure services. Our operating results may fluctuate significantly and unpredictably as a result of movements in digital asset prices, changes in mining difficulty or network hash rate, and halving or other protocol-level events affecting mining economics.

Our ability to reallocate power or computing capacity between AI and digital-asset-mining uses may be constrained, reducing our operational flexibility.

Our ability to reallocate power or computing capacity between AI and digital-asset-mining uses may be constrained by contractual commitments to AI or hosting customers, by the differing technical requirements of each use case (including differences in hardware, cooling, and networking configurations), or by regulatory restrictions, and any such constraint could reduce our operational flexibility and adversely affect our revenue and profitability. A sustained decline in digital asset prices or an adverse change in the regulatory treatment of digital asset mining could have a material adverse effect on the portion of our business attributable to those operations, and could also affect our ability to monetize infrastructure originally built or intended for mining use.

Digital asset mining hardware is subject to rapid technological obsolescence, physical degradation, and supply chain risk.

To the extent we own or operate bitcoin miners or other digital-asset-mining hardware, that hardware is subject to malfunction, technological obsolescence, and physical degradation, and may need to be replaced or upgraded frequently to remain competitive. We may also face disruptions or shortages in the supply chain for mining hardware and related components, which could impair our ability to maintain or expand our mining operations on favorable terms.

Regulatory developments affecting digital assets or digital asset mining could restrict our operations or increase our costs.

Government regulators may restrict the ability of electricity suppliers to provide electricity to digital asset mining operations such as ours, impose new taxes on electricity used for mining, or fully or partially ban mining operations in certain jurisdictions. Increasing regulatory and public scrutiny of the environmental impact of digital asset mining could result in new restrictions that adversely affect our ability to conduct mining operations or increase our compliance costs.

We are dependent on a reliable, cost-effective, and scalable supply of electricity, and any disruption, shortage, or increase in the cost of power could materially harm our business.

Our AI data center, GPU compute, and digital asset mining operations are highly energy-intensive, and electricity is a significant component of our operating costs. We have announced plans to develop infrastructure that includes substantial dedicated power capacity, and in some cases we may pursue behind-the-meter or on-site power generation (including natural gas generation) to reduce reliance on the traditional utility grid. Wholesale and retail power price volatility, transmission and interconnection constraints, curtailment risk, and our reliance on specific utilities, grid operators, or independent power arrangements each present risk to our ability to secure sufficient, reliable, and reasonably priced electricity for our planned expansion. Any electricity shortage, outage, curtailment, or material cost increase could impair our ability to meet uptime and performance commitments to customers, trigger contractual termination rights, delay our development plans, and have a material adverse effect on our business, financial condition, and results of operations.

Regulatory actions or grid operator requirements relating to large electricity loads could increase our costs or restrict our operations.

As our data center operations scale, we may become subject to heightened scrutiny from regulators, utilities, and grid operators with respect to our electricity consumption and the impact of our operations on grid reliability, including with respect to interconnection processes for large loads, grid stability, voltage and frequency ride-through requirements, and curtailment obligations. New rules, tariffs, or operational restrictions could require us to incur significant additional costs, delay expansion plans, reduce the operational flexibility of our facilities, or limit our ability to participate in ancillary services or demand-response markets, any of which could have a material adverse effect on our results.

Our reliance on behind-the-meter or on-site power generation, if pursued, subjects us to additional operational, regulatory, and commodity price risk.

To the extent we rely on on-site power generation, including natural gas generation, to supply our data center or mining operations, we are subject to risks associated with the availability and price of natural gas or other fuel, the operation and maintenance of generation equipment, compliance with air quality and other environmental permitting requirements, and the risk that behind-the-meter arrangements are challenged or restricted by regulators or utilities. Any disruption to our on-site generation arrangements could require us to rely on grid power on less favorable terms, or could disrupt our operations entirely.

Severe weather, natural disasters, and other physical events could disrupt our operations and cause material losses.

Our data center and power infrastructure facilities are susceptible to damage from severe weather events, including floods, storms, extreme heat, winter weather events, and other climate-related phenomena, as well as earthquakes, fires, and other natural or man-made disasters. Any such event could result in extended outages, significant repair and replacement costs, loss of customer contracts, and potential liability, and our insurance coverage may prove insufficient to cover the full extent of any such losses.

We face significant cybersecurity risks, and a breach of our systems or those of our customers or third-party providers could result in substantial harm to our business and reputation.

Our data centers and compute services may house sensitive or proprietary customer workloads, including AI model training data, and our information technology and operational technology systems may be vulnerable to cyberattacks, including from sophisticated and well-resourced threat actors. Any breach or unauthorized access affecting our systems or those of our third-party partners could expose us to significant legal liability, regulatory investigation, litigation, remediation costs, loss of customer confidence, and reputational harm. The integration of the acquired AI business with our legacy systems, and the consolidation of new personnel, vendors, and technology platforms, may introduce additional vulnerabilities during the transition period beyond those addressed by our existing cybersecurity risk management program described under Item 1C of our most recent Annual Report on Form 10-K.

Our reliance on third-party service providers, software, and infrastructure creates additional operational and security risk.

We rely on third-party vendors, service providers, and software platforms to support our operations, including for networking, connectivity, monitoring, and administrative functions. A failure, security incident, or service disruption affecting any of these third parties could impair our ability to operate our infrastructure or serve our customers, and we may have limited visibility into, or control over, the security and reliability practices of these providers.

We compete against larger, better-capitalized, and more established participants in the AI infrastructure, GPU compute, and digital power markets.

We compete for customers, sites, power, hardware, and personnel against specialized AI infrastructure and cloud providers, large-scale bitcoin-mining-to-HPC conversion companies, and general-purpose hyperscale cloud providers, many of which have substantially greater financial, technical, marketing, and operational resources, greater name recognition, longer operating histories, and stronger customer and supplier relationships than we do. We expect competition in this sector to intensify as additional entrants, including well-capitalized technology companies and financial sponsors, pursue AI infrastructure opportunities.

Demand for AI infrastructure may not develop as we anticipate, may develop more slowly than expected, or the industry may develop excess capacity relative to demand.

The AI infrastructure sector has attracted substantial capital investment industry-wide, and there is a risk that the industry could develop excess compute or data center capacity relative to customer demand. A slowdown or reprioritization of AI-related spending, weaker-than-expected enterprise or institutional adoption of AI workloads, or technological advances that reduce the amount of compute required for AI training or inference, could reduce demand for our infrastructure, depress pricing, and adversely affect our revenue growth and results of operations. Conversely, if industry-wide demand outpaces available power and site capacity, we may be unable to secure the resources necessary to grow as planned, or may face increased costs in doing so.

Our AI infrastructure and digital power operations are capital-intensive, and we may require significant additional financing that may not be available on acceptable terms, if at all.

The development, construction, and operation of AI data center infrastructure, GPU compute capacity, and associated power infrastructure requires substantial upfront and ongoing capital investment. Our ability to execute our growth strategy will depend on our ability to access debt and equity capital markets, project financing, equipment financing, or other structured arrangements on acceptable terms. Any inability to raise sufficient capital when needed could result in delays to or curtailment of our expansion plans, impairment of our ability to meet customer or supplier commitments, and a material adverse effect on our business. Additional equity or convertible debt financing, if obtained, could result in significant dilution to existing stockholders, and additional debt financing, if obtained, could include restrictive covenants that limit our operational and financial flexibility. This capital-intensive expansion is undertaken against the backdrop of our existing history of losses, negative working capital, and management’s conclusion that there is substantial doubt about our ability to continue as a going concern as discussed in Item 2 “Management’s Discussion and Analysis of Financial Condition and Results of Operations,” Part I of this Quarterly Report, which increases the risk that financing for our AI infrastructure buildout may not be available to us on acceptable terms, or at all.

Our capital expenditure plans and expected returns are based on assumptions that may not prove accurate, and actual returns on our infrastructure investments may be lower than expected.

Our decisions to invest in AI data center capacity, GPU hardware, and power infrastructure are based on assumptions regarding future customer demand, pricing, utilization rates, and the useful life of our assets. If actual demand, pricing, or utilization differs materially from our assumptions, the returns on our capital expenditures may be lower than expected, or we may be required to impair the value of our assets, which could have a material adverse effect on our financial condition and results of operations.

We are subject to environmental, health, safety, and land use regulation, and increasing scrutiny of the energy intensity of AI and digital-asset-mining infrastructure could increase our costs or restrict our operations.

Our data center and digital asset mining operations are subject to a variety of federal, state, and local environmental, health and safety, zoning, building code, and land use regulations. Increasing regulatory and stakeholder scrutiny of the energy intensity and environmental impact of AI infrastructure and digital asset mining, including potential new restrictions on large electricity loads, water usage for cooling, noise, or the availability of power for such uses, could require us to incur significant additional capital or operating expenditures, restrict or delay our development activities, or otherwise adversely affect our results of operations and reputation.

Climate-related regulatory developments could increase our costs, restrict our access to energy, or otherwise adversely affect our operations.

Legislative and regulatory responses to climate change, including potential restrictions on greenhouse gas emissions, carbon pricing mechanisms, or requirements affecting the availability of power for energy-intensive uses, could increase our operating costs, restrict our access to energy on favorable terms, or otherwise adversely affect our ability to develop and operate our infrastructure as planned