{"url_path":"/sec/agpu/8-k/2026-06-09/body","section_key":"body","section_title":"Body","topic":"sec","document":{"doc_type":"8-K","doc_date":"2026-06-09","source_url":"https://www.sec.gov/Archives/edgar/data/1446159/0001171843-26-004008-index.html","accession_number":"0001171843-26-004008","cik":"0001446159","ticker":"AGPU","issuer_name":"Axe Compute Inc.","edgar_url":"https://www.sec.gov/Archives/edgar/data/1446159/0001171843-26-004008-index.html","primary_entity_key":"0001446159","primary_entity_name":"Axe Compute Inc."},"word_count":3400,"has_tables":false,"body_markdown":"EX-99.1\n2\nexh_991.htm\nEXHIBIT 99.1\n\n**EXHIBIT 99.1**\n\n** **\n\n**RISK FACTORS**\n\n** **\n\nThe following risk factors supplement and update the risk factors disclosed in Part I, Item\n1A of the Annual Report on Form 10-K of Axe Compute Inc. (the \"Company,\" \"we,\" \"us,\" or \"our\")\nfor the fiscal year ended December 31, 2025 (the \"2025 Form 10-K\"), and reflect the Company's expansion into the purchase, ownership,\nand operation of GPU computing infrastructure deployed in data center facilities. These risk factors should be read together with the\nrisk factors and other information contained in the 2025 Form 10-K and our subsequent filings with the U.S. Securities and Exchange Commission.\nTo the extent the following is inconsistent with the risk factors in the 2025 Form 10-K, the following supersedes those risk factors.\nAny of the following risks could materially and adversely affect our business, financial condition, results of operations, and prospects,\nand the trading price of our common stock could decline. Additional risks and uncertainties not currently known to us or that we currently\ndeem immaterial may also impair our business operations.\n\nRisks Related to Our Ownership and Operation of GPU Computing Infrastructure\n\n**Our expansion into owning and operating GPU computing infrastructure is capital-intensive\nand will require substantial and growing capital expenditures, and any inability to obtain capital on acceptable terms may adversely affect\nour business.**\n\n** **\n\nWe have historically pursued an asset-light operating model under which we did not own GPU computing\nhardware within physical data center facilities and instead provided access to GPU compute capacity primarily through infrastructure made\navailable by the Aethir network. We have now expanded our business to purchase, own, and operate GPU computing hardware. This owned-asset\nmodel is substantially more capital-intensive than our prior model and will require significant and growing capital expenditures to procure,\ndeploy, maintain, upgrade, and expand our infrastructure. We expect to fund these expenditures through a combination of customer deposits\nand prepayments, cash from operations, and equity and debt financing, which may not be available to us on favorable terms, or at all.\nIf adequate financing is not available when required, we may be unable to acquire the hardware and infrastructure necessary to fulfill\nour customer commitments or execute our growth strategy. If we raise additional funds through equity or convertible securities, our existing\nstockholders may experience substantial dilution, and any such securities may have rights, preferences, and privileges senior to those\nof our common stock.\n\n**The GPUs and related infrastructure we will now own are subject to rapid\ntechnological obsolescence, and our results of operations depend on our ability to accurately estimate their useful lives and to\navoid impairment of these assets.**\n\n** **\n\nUnlike our prior model, in which we did not own the underlying compute hardware, we now bear the full\neconomic risk of the GPUs and related equipment we purchase. GPU technology is advancing rapidly, and newer generations of GPUs that offer\nmaterially better performance, efficiency, or total cost of ownership are introduced frequently. As a result, the GPUs and related infrastructure\nwe will now own may become obsolete, decline in value, or generate lower pricing and utilization than we anticipate before the end of\ntheir expected useful lives. We must make estimates regarding the useful lives of our computing equipment and our ability to redeploy\nthat equipment beyond the term of any initial customer contract, and we cannot guarantee that these estimates will prove accurate. If\nour assumptions regarding useful lives, residual values, redeployment, or utilization prove incorrect, or if events or changes in circumstances\nindicate that the carrying amount of our infrastructure may not be recoverable, we may be required to accelerate depreciation or record\nmaterial impairment charges, which could materially and adversely affect our reported financial results.\n\n**A substantial portion of our compute revenue is expected to be derived from a limited\nnumber of customers and contracts, and the loss of, or non-performance by, any such customer would adversely affect our business.**\n\n** **\n\nWe expect that, for the foreseeable future, a substantial portion of our compute revenue\nwill be concentrated among a small number of customers and contracts. This concentration exposes us to heightened counterparty\ncredit risk and to the risk of non-payment or non-performance, including in the event a customer experiences financial difficulty,\ninsolvency, or bankruptcy. Although our contract is structured on a take-or-pay basis and secured with a deposit, prepayment, and\nmonthly in-advance payments, we cannot assure you that a customer will perform its obligations, that the definitive agreement will\nbe enforceable in accordance with its terms, or that a customer will exercise any renewal option. The\nloss of, a default by, a dispute with, or a significant reduction in spending by any one of our major customers, or our inability to\nreplace such revenue on comparable terms, could have a disproportionate adverse effect on our business, results of operations, and\nfinancial condition.\n\n**Our compute business depends on the operation of dedicated data center facilities,\nincluding the availability of reliable power and cooling, and operational failures at these facilities could materially disrupt our ability\nto serve customers.**\n\n** **\n\nOur expanded business depends on deploying and operating owned GPU infrastructure within\ndedicated data center facilities. The performance, availability, and delivery of our services depend on numerous factors, many of which\nare outside our control, including the continued availability and functioning of power and cooling systems, the success or failure of\nredundancy, disaster recovery, and business continuity systems, and decisions or failures by the third-party owners and operators of\nthe facilities in which our infrastructure is installed. Such data centers and associated infrastructure are also subject to risks of\ndamage, interruption, or destruction from power outages, equipment failures, fires, floods, natural disasters, physical or cybersecurity\nattacks, human error, and other events. Because the deployment under our largest contract to date is concentrated in a single facility,\nany prolonged outage, capacity constraint, or other disruption affecting that facility could prevent us from meeting contracted service\nlevels, expose us to service credits, penalties, or termination rights, and materially and adversely affect our business.\n\n**Our business could be harmed if we are unable to secure sufficient power, or by increases\nin the cost of power or the imposition of new regulatory requirements on data center power consumption.**\n\n** **\n\nOperating owned GPU infrastructure requires access to substantial, reliable, and cost-effective electrical\npower; our largest deployment to date requires 4.8 megawatts of committed power capacity alone. The rapid expansion of AI and large-scale\ndata center development has significantly increased electricity demand in certain markets, and policymakers, utilities, and regulators\nare increasingly scrutinizing the impact of data centers on ratepayers, grid reliability, and the environment. We may face power outages,\nshortages, capacity constraints, interconnection delays, or significant increases in the cost of securing power, any of which could limit\nour ability to operate or expand our infrastructure. In addition, governments may impose new requirements on data center operators, including\nobligations to fund grid upgrades, procure dedicated generation, enter into long-term capacity arrangements, accept curtailment during\nperiods of grid stress, or satisfy additional permitting, carbon reporting, or cost-allocation requirements, and may restrict, condition,\nor delay new data center development. The global energy market has experienced significant volatility and inflationary pressure, and we\nexpect power costs to remain volatile and unpredictable. Any of these developments could increase our operating costs, impair our ability\nto serve customers, delay our growth, and materially and adversely affect our business.\n\n**We depend on a limited number of suppliers, and primarily on NVIDIA, for the GPUs and\nother hardware we purchase, and any supply disruption, delay, or price increase could impair our ability to deploy infrastructure and\nfulfill customer commitments.**\n\n** **\n\nOur ability to acquire and deploy owned GPU infrastructure depends on our ability to procure GPUs\nand related hardware in sufficient quantities, on acceptable terms, and within timeframes consistent with our customer commitments. We\nsource GPU hardware primarily from NVIDIA, which is currently the dominant supplier of GPUs used for AI training and inference, and we\ndo not manufacture any hardware ourselves. Reliance on a limited number of suppliers exposes us to a range of risks, including limited\navailability of the latest-generation components, lack of control over production costs, delivery, and pricing, extended or unpredictable\nlead times, the potential for binding price or purchase commitments at above-market rates, supplier prioritization of other customers,\nand shifts in market-leading technologies away from those offered by our current suppliers. Our suppliers in turn rely on complex networks\nof third-party suppliers, including semiconductor foundries such as Taiwan Semiconductor Manufacturing Company, and any disruption affecting\nthese upstream suppliers, whether due to geopolitical factors, capacity constraints, or natural disasters, could affect the availability\nand cost of the hardware we require. The loss of or significant disruption to our access to NVIDIA GPU supply and related hardware, or\nmaterial price increases or extended delivery lead times, could delay our deployments, including our targeted third-quarter 2026 deployment,\nreduce our available capacity, and materially and adversely affect our business.\n\n**We may be unable to deploy our owned infrastructure on the timelines we have committed,\nand delays in deployment could result in penalties, lost revenue, or reputational harm.**\n\nOur largest contract to date contemplates a dedicated cluster purpose-built to the\ncustomer's specifications, with a targeted deployment start in the third quarter of 2026. The procurement, integration,\nconfiguration, and commissioning of large-scale GPU clusters and associated storage, networking, power, and cooling infrastructure\nis complex and subject to numerous potential points of failure, including hardware delivery delays, facility readiness, the\navailability of data center equipment such as switchgear, power distribution units, and cooling equipment, the availability of\nskilled labor, and dependence on third-party facility operators and contractors. Our forward-looking statements regarding deployment\nare subject to risks relating to the execution and enforceability of the definitive agreement, hardware supply chain constraints,\nand facility readiness. If we are unable to deploy contracted infrastructure on the agreed schedule, we may be subject to service\ncredits, penalties, delayed or reduced revenue, customer disputes, termination rights, or reputational harm, any of which could\nmaterially and adversely affect our business.\n\n**If customer demand is insufficient to utilize the capacity we build, or if we are unable\nto redeploy infrastructure following the expiration or termination of a contract, we may not realize the expected returns on our capital\ninvestments.**\n\n** **\n\nThe owned-asset model requires us to commit substantial capital to acquire and deploy infrastructure,\noften in advance of, or in reliance upon, specific customer contracts. Our expected returns depend on sustained customer demand and high\nutilization of the capacity we build. If a customer reduces its usage, does not renew or terminates its contract, or if we are otherwise\nunable to redeploy or resell capacity on economically attractive terms following the expiration of an initial contract term, we may experience\nunderutilized capacity, stranded assets, reduced margins, or impairment charges. Because our infrastructure is purpose-built and concentrated,\nand because GPUs are subject to rapid obsolescence, we may be unable to repurpose assets for other customers or workloads without incurring\nadditional cost or delay. Any failure to achieve sufficient utilization of our owned infrastructure could materially and adversely affect\nour business, results of operations, and financial condition.\n\n**We expect to incur indebtedness and to use secured or asset-backed financing structures\nto fund our infrastructure, and our leverage could adversely affect our financial condition and flexibility.**\n\n** **\n\nTo fund the acquisition of GPU infrastructure, we may incur substantial indebtedness and\nmay pursue secured financing arrangements, including asset-backed, equipment-financing, or other collateralized structures, in which our\nGPUs and related assets serve as collateral. Companies in our industry carry significant indebtedness and finance GPU purchases through\ndelayed draw term loans, original equipment manufacturer financing arrangements, and similar structures secured by the depreciable cost\nof GPU servers. A substantial level of indebtedness could require us to dedicate a significant portion of our cash flow to debt service,\nincrease our vulnerability to adverse economic and industry conditions, limit our ability to obtain additional financing, restrict our\noperational and strategic flexibility through restrictive covenants, and expose us to the risk of acceleration or foreclosure on pledged\nassets in the event of a default. The management of a more complex capital structure, including multiple layers of secured and unsecured\ndebt with differing covenants, maturities, and priorities, could increase our financial and operational risks and heighten the risk of\ndisputes among creditors. Rising or volatile interest rates would increase the cost of any floating-rate indebtedness, and we may be required\nto enter into interest rate hedging arrangements that may not be effective.\n\n**We have a limited operating history operating an owned-infrastructure GPU business,\nwhich makes it difficult to evaluate our business and prospects.**\n\n** **\n\nWe have only recently expanded into purchasing, owning, and operating GPU computing infrastructure,\nand we have a limited operating history under this business model. Our prior compute model was asset-light and distributed, and the owned-infrastructure\nmodel requires different capabilities, including the procurement and lifecycle management of hardware, the operation of dedicated data\ncenter deployments, the management of large multi-year take-or-pay contracts, and the management of capital-intensive financing. Our limited\nexperience delivering and managing longer-term, large-scale customer contracts may expose us to cost overruns, underutilized capacity,\nperformance obligations, service-level commitments, and other contractual liabilities. As a result, our historical results are not indicative\nof our future performance, our future results may be difficult to predict and may fluctuate significantly from period to period, and you\nshould consider our business and prospects in light of the risks and uncertainties frequently encountered by companies operating in new\nand rapidly evolving capital-intensive markets.\n\nUpdates to Existing Risk Factors\n\n**The energy and environmental demands of data centers and GPU compute infrastructure\nmay constrain the growth of the compute market and result in increased regulatory costs or operational limitations.**\n\n** **\n\nData centers are significant consumers of electrical power, and this level of energy consumption\nhas attracted increasing scrutiny from regulators, utilities, and environmental groups, which may result in additional restrictions, permitting\nrequirements, carbon reporting obligations, or energy surcharges that increase the cost of GPU compute infrastructure. Because we will\nnow own and operate GPU computing infrastructure deployed in dedicated data center facilities that require substantial committed power,\nincluding 4.8 megawatts of dedicated power for our largest deployment to date, constraints on available power capacity, increases in the\ncost of power, and new regulatory or environmental requirements directly affect our operating costs and our ability to expand. In addition,\nreputational and environmental, social, and governance concerns relating to the energy and water footprint of AI compute infrastructure\ncould adversely affect our business relationships, our access to capital, and our ability to obtain permits and approvals.\n\n**Geopolitical tensions and trade restrictions, particularly between the United States\nand China, could disrupt GPU supply chains and limit our addressable market.**\n\n** **\n\nThe global GPU compute market depends heavily on complex international supply chains,\nincluding semiconductor manufacturing concentrated in Taiwan and South Korea, and geopolitical tensions between the United States\nand China have already resulted in restrictions on the export of certain advanced semiconductors, including certain NVIDIA GPU\nproducts. Because we will now own GPU hardware sourced primarily from NVIDIA, geopolitical tensions, tariffs, economic sanctions,\nand export controls directly affect the cost, availability, and delivery lead times of the GPUs and related components we acquire.\nIncreasing use of tariffs and export controls has impacted, and may in the future impact, the availability and cost of GPUs and\nother components, and expansion or reinterpretation of U.S. export controls covering advanced computing hardware could limit the\navailability of components or require reconfiguration of our deployment plans. Any such disruption could increase our procurement\ncosts, delay our deployments, including our targeted third-quarter 2026 deployment, and materially and adversely affect our compute\nbusiness and our ability to execute our strategy.\n\n**Demand for GPU compute is highly concentrated, and a slowdown in AI-related spending\nor the development of excess industry capacity could adversely affect our business.**\n\n** **\n\nA substantial portion of current and projected demand for GPU compute infrastructure is driven\nby a small number of large technology companies and government-sponsored AI programs, and any significant reduction in their capital expenditures\ncould have a disproportionately negative impact on the broader GPU compute market. In addition, a substantial portion of our own compute\nrevenue is now expected to be derived from a limited number of customers and contracts, including our recently announced approximately\n$260 million enterprise engagement. A slowdown, deferral, or reprioritization of AI-related customer spending, or the development of excess\nindustry capacity if anticipated AI workloads do not materialize, could result in pricing pressure, reduced utilization, longer sales\ncycles, contract renegotiations, or impairment charges, any of which could be magnified by the capital-intensive, owned-asset nature of\nour expanded business.\n\n**Advances in AI model efficiency could reduce demand for GPU compute, adversely affecting\nthe value of our compute business and our owned infrastructure.**\n\n** **\n\nA key driver of demand for GPU compute is the scale required to train and run AI models,\nand consistent advances in AI model efficiency — such as new architectures, training techniques, or algorithmic improvements that\nachieve equivalent or superior results using significantly less compute — could substantially reduce demand for raw GPU compute\ncapacity. Because we will own GPU hardware rather than relying solely on a distributed network, a significant and sustained reduction in\nGPU compute demand could reduce the utilization, pricing, and resale or redeployment value of our owned infrastructure, and could require\nus to recognize accelerated depreciation or impairment charges, in addition to adversely affecting the value of our ATH treasury holdings.\n\n**Security breaches and other disruptions affecting our infrastructure or the facilities\nin which it is housed could compromise sensitive information and expose us to liability.**\n\n** **\n\nOur business requires that we collect and store sensitive data, and our information technology\nand infrastructure are susceptible to attacks by hackers, viruses, employee error, malfeasance, or other activities. In addition, our\nowned GPU computing infrastructure and the data center facilities in which it is deployed are subject to physical and cybersecurity risks,\nincluding attacks by outside parties (whether private or state-backed), human error, malfeasance, insider threats, system vulnerabilities,\nand inadequate security controls, any of which could result in service outages, unauthorized access to or loss of customer data and workloads,\nor damage to our infrastructure. Our enterprise customers contract for dedicated infrastructure in part to ensure that their proprietary\ndata remains within a controlled facility boundary, and any physical or cybersecurity incident affecting our infrastructure or the facilities\nin which it is housed could expose us to service-level penalties, contractual liability, loss of customers, regulatory exposure, and reputational\nharm.\n\n**If our information technology and communications systems, or the infrastructure and\nfacilities on which our compute business depends, fail or experience a significant interruption, our business could be materially and\nadversely affected.**\n\n** **\n\nThe efficient operation of our business is dependent on information technology and communications\nsystems, the failure of which could disrupt our business and result in decreased revenue and increased overhead costs. Our expanded business\nfurther depends on the continuous operation of owned GPU computing infrastructure housed in third-party data center facilities, and the\navailability and performance of that infrastructure depend on power, cooling, network connectivity, redundancy systems (including N+1\nredundant power), and the performance of the third-party operators of the facilities in which our equipment is installed. The failure\nof any of these systems or services, including any failure of redundancy or disaster recovery measures, could prevent us from meeting\ncontracted service levels and could materially and adversely affect our reputation, business, and results of operations.\n\n**Our expansion into owned GPU infrastructure has materially increased our capital requirements\nand our dependence on external financing.**\n\n** **\n\nWe have a history of negative operating cash flows and have funded our operations in\npart through at-the-market and private placement equity financings, with a significant portion of our liquidity held in ATH, a\ndigital asset whose market price has exhibited substantial volatility. Our expansion into purchasing and owning GPU computing\ninfrastructure has materially increased our capital expenditure requirements and our dependence on external financing, and our\nliquidity needs are now driven in part by the substantial upfront and ongoing costs of acquiring, deploying, maintaining, and\nexpanding owned hardware and data center capacity. Although our largest contract to date is supported by a customer deposit,\nprepayment, and monthly in-advance payments on a take-or-pay basis, these amounts may be insufficient to fund our capital\nrequirements, and our reliance on volatile sources of liquidity, including the price of ATH, may further constrain our ability to\nfund these commitments."}