Hyperscale Data (GPUS), a former Bitcoin mining company that has shifted its business toward artificial intelligence infrastructure, is warning that it may not have enough cash to complete a major data center project in Michigan. The company needs another $74.8 million by March, despite having sold its entire Bitcoin treasury for approximately $74 million since late July.
The financial pressure raises questions about whether Hyperscale can deliver on an AI infrastructure agreement it has promoted as potentially worth $1.2 billion. In a regulatory filing published Wednesday, the company said its available funds and financing options would not be sufficient to cover its planned spending and obligations over the next 12 months.
At the end of September, Hyperscale had just $17 million in unrestricted cash, compared with $231 million in current liabilities. Management warned that these conditions raised substantial doubt about the company’s ability to continue operating.
Hyperscale Sells Its Remaining Bitcoin as Cash Reserves Shrink
Hyperscale confirmed last week that it had sold its remaining 211 BTC for approximately $16 million. The transaction completed the company’s exit from its Bitcoin treasury, bringing total sales since late July to around 1,113 BTC worth approximately $74 million.
The liquidation has not resolved its funding challenges. Its unrestricted cash balance fell from $36.8 million at the end of June to $17 million by September 30, a decline of more than 50%. The company’s short-term liabilities exceeded its current assets by approximately $128 million, adding to concerns about its ability to meet upcoming financial commitments.

Hyperscale also reported a net loss of roughly $62 million during the first nine months of 2026. Management is now considering several options to secure additional funding, including new debt, equity financing, asset sales and other transactions.
The company already has a program allowing it to sell common shares directly into the market, but that program is currently suspended. The latest warning follows a similar disclosure in August, indicating that the company’s financial difficulties have persisted while it attempts to complete its business transformation.
$1.2 Billion AI Agreement Depends on Completing Michigan Data Center
Hyperscale’s financial situation is closely tied to its attempt to convert a former Bitcoin mining operation in Dowagiac, Michigan, into an AI computing facility.
In June, the company signed an agreement with an unnamed California-based AI infrastructure provider to supply 20 megawatts of computing capacity. Hyperscale has presented the contract as potentially generating $1.2 billion in revenue. However, that figure is conditional. It assumes the customer exercises two additional five-year extensions following the initial 10-year agreement and that Hyperscale fulfills its contractual obligations.
The construction and equipment costs for the project are estimated at $122.1 million. The company had spent approximately $47.3 million through October, leaving a substantial amount of work and financing still required.
Hyperscale expects to deliver the first 10 megawatts of capacity in December, followed by another 10 megawatts by the end of March. Completing the remaining infrastructure will require approximately $74.8 million in additional spending between November and March.
This creates a tight timeline for a company already facing limited liquidity. Any delays in securing financing or completing construction could complicate its ability to meet the project’s schedule and realize the potential value of the agreement.
Bitcoin Miners Face Growing Financial Demands as They Pivot to AI
Hyperscale’s situation highlights the financial challenges facing Bitcoin miners that are repositioning their businesses around artificial intelligence and high-performance computing.
Mining companies possess infrastructure that can be attractive to AI operators, particularly access to electricity, power connections and existing industrial facilities. However, converting mining sites into data centers requires substantial investment in construction, equipment and financing.
Riot, meanwhile, expects to spend around $5.5 billion on its Anthropic-related project. The company could face an annual financing cost equivalent to 12% through rent deductions if it cannot obtain funding on acceptable terms.
These examples demonstrate that the AI opportunity comes with considerable capital requirements. While long-term infrastructure contracts can offer miners a route toward recurring revenue, companies must first secure the resources needed to build and operate the facilities.
Can Hyperscale Data secure the additional $74.8 million needed to complete its AI infrastructure project by March? Does the growing demand for AI data centers offer Bitcoin miners a sustainable alternative to mining, or could the cost of the transition create new financial risks?
Editorial Process for bitcoinist is centered on delivering thoroughly researched, accurate, and unbiased content. We uphold strict sourcing standards, and each page undergoes diligent review by our team of top technology experts and seasoned editors. This process ensures the integrity, relevance, and value of our content for our readers.

