Cango receives over $75M in equity investments to boost AI compute push as ‘supercycle’ takes off

• Cango received $10.5 million from an equity investment from EWCL in addition to $65 million cumulative investment from entities owned by members of its leadership.
• The firm launched a new Dallas-based subsidiary, EcoHash Technology, to lead development of its global AI compute network.
• Cango’s roadmap includes near-term hardware deployment, mid-term software orchestration, and long-term global AI platform expansion.

Cango announced that it received over $75 million in fresh equity investments to fuel the growth of its AI infrastructure business.

According to a statement by the firm, it has closed the previously announced Class B equity investment of $10.5 million from Enduring Wealth Capital Limited (EWCL), 

The firm also secured equity investments to the tune of $65 million from entities wholly-owned by Cango’s chairman, Xin Jin and company director, Chang-Wei Chiu, bringing the total capital raised to $75.5 million.

The fresh capital signals strong support from company leadership to strengthen its balance sheet and also contribute to its AI business expansion.

CEO Paul Yu stated that Cango also made a treasury adjustment to strengthen its balance sheet and reduce financial leverage with a $305 million sale from its BTC holdings to settle debt obligations. This adjustment is expected to increase the company’s capacity to fund its strategic expansion into AI compute infrastructure.

In a letter to its shareholders, the NYSE-listed company mentioned that it has established EcoHash Technology LLC, a new subsidiary that it fully owns, based in Dallas, Texas, to focus on advancing its AI compute initiatives.

Cango sets up infrastructure to power AI transition 

As mentioned in earlier reports, Cango made major strides to end 2025, with its Bitcoin treasury valuation exceeding its stock market capitalization as of early January 2026. 

At the time, it also announced plans to create a globally distributed AI-compute network, and this latest development is a step in that direction for the company. 

CEO Paul Yu stated that the firm is trying to cover the “Power Gap,” a disconnect between rising AI compute demand and existing grid capacity, and that has informed its latest strategic transformation. 

Yu’s sentiments echo opinions shared by Nokia CEO Justin Hotard, who told Reuters that “I fundamentally think we’re at the front end of an AI supercycle, much like the 1990s with the internet.”

Cango plans to leverage its globally distributed mining infrastructure, which spans more than 40 sites across North America, the Middle East, South America, and East Africa, to deliver scalable, low-latency compute capacity to meet what it describes as long-tail inference demand from small and medium-sized enterprises.

While it is committing resources to expand its AI inference compute service to its existing sites, the firm also plans to become an enabler for mid-to-small-sized BTC mining operators seeking a low-cost, modular pathway to diversify their infrastructure to tap into the AI supercycle. 

Internal moves lined up for expansion goals

To lead Cango’s AI operations is Jack Jin, who was recently hired as the chief technology officer of its AI business subsidiary. Before his role with Cango, Jin managed large-scale GPU systems and orchestration platforms at Zoom Communications.

Cango is also assembling a dedicated team that will be joining the new CTO in guiding technical execution.

The company ticked off another box after it successfully completed a technical demonstration that validated its core hardware innovation, which is standardized, plug-and-play compute nodes designed for fast deployment across its existing infrastructure. 

This will enable the firm to offer on-demand compute capacity by drawing on power from current mining operations. Cango expects the sites to become operational in relatively short timeframes.

The company stated that conversion to AI-ready infrastructure requires limited upgrades, creating the potential to keep developing more revenue streams while addressing enterprise demand for accessible AI compute.

Cango’s transition roadmap

Cango transition is following a three-phase development roadmap, which are the near, medium, and long-term phases, and according to Yu, early results are already emerging.

In the near term, the focus remains on standardization and the efficient deployment of containerized GPU compute nodes as plug-and-play solutions for quick rollout.

The medium-term phase focuses on software-defined orchestration. Cango is developing a proprietary in-house platform to manage and integrate its distributed compute capacity, evolving its role from operator to what it terms an ecosystem enabler. As an enabler, the company will be helping mid-to-small-sized BTC mining operators to diversify their infrastructure into AI at relatively lower costs.

This architecture is designed to offer its global footprint as an integrated, enterprise-grade network without the typical infrastructure complexity associated with distributed computing at scale.

The long-term vision extends to building a mature global AI infrastructure platform by activating underutilized power across its mining ecosystem. The company management expects this strategy to bring in recurring revenue streams from platform services and compute agreements designed to be durable across market cycles.

Yu acknowledged the multi-year nature of the transition, stating, “The transition from mining to AI compute will continue to develop over multiple years, and we are still in the early stages. While the path ahead requires sustained effort, our roadmap is clear, and we are committed to prudent, step-wise execution.”

The announcement follows Cango’s completion of several foundational milestones in 2025, including acquiring and enhancing the hashrate efficiency of 50 exahash per second of on-rack machines, securing an initial 50 megawatts of energy infrastructure, divesting legacy operations, and completing its transition to a direct NYSE listing.

Cango still maintains strong interests in its AutoCango.com international used car export business alongside its digital asset operation.