The most important number in Csquare's IPO filing is not the $1.35 billion it hopes to raise. It is the $734 million in revolving credit facility debt the company plans to pay off with the proceeds.
Brookfield Corp.-backed Csquare is not going public to fund a land grab for AI-powered data center development. It is going public to refinance the bridge loans that funded its creation. The IPO is a capital structure event dressed as a growth story.
Csquare plans to offer 50 million shares at $23 to $27, valuing the company at up to $4.2 billion. Brookfield will retain a 67.1% stake. The proceeds will first repay the full $734 million revolver, then a $75 million promissory note held by Brookfield, and finally a portion of the company's $4.3 billion in asset-backed loans.
This is not a company selling equity to fund expansion. It is a company selling equity to replace floating-rate bank debt with permanent capital. The revolver and asset-backed loans were the financial scaffolding that allowed Brookfield to assemble Csquare through acquisitions, including the $1 billion cash purchase of 10 data centers and the merger of Evoque and Cyxtera into Centersquare last October. That scaffolding was never meant to be permanent.
The market signal is clear: even for a Brookfield-backed hyperscaler with 389 megawatts of power across 64 sites, the cost and structure of debt matter. Csquare lost $66 million in the first quarter, nearly double the $34.9 million loss a year earlier. Revenue rose 16% to $270 million, but the losses are widening as interest costs and depreciation from the acquisition binge compound.
Public equity markets are being asked to absorb the risk that bank lenders and private credit funds are increasingly reluctant to carry. The IPO is a transfer of risk from the debt stack to the equity stack. The banks get paid off. Brookfield gets a liquid public currency for its stake. Public shareholders get a business that is still losing money at an accelerating rate, with $4.3 billion in asset-backed loans still on the books.
Who benefits? Brookfield most directly. It retains control, gets its promissory note repaid, and unlocks a public market valuation for a portfolio it assembled through debt-funded acquisitions. The banks that provided the revolver get their capital back. The asset-backed lenders get a partial paydown, reducing their exposure.
Who is exposed? The public market investors who buy the IPO. They are underwriting a business that is pre-profit, capital-intensive, and dependent on continued hyperscaler demand for AI workloads. Csquare's customer concentration is low by design, with no single client above 7% of revenue, but that diversification comes at a cost: the company lacks the anchor tenant economics that make some data center REITs so predictable.
The timing matters. Brookfield is bringing this IPO into a market that has shown strong appetite for AI infrastructure stories, but also one where investors are increasingly discriminating between platforms with clear paths to profitability and those still burning cash. Csquare's prospectus leans heavily on its early-mover advantage on AI, its 2-millisecond latency reach to 92% of the U.S. population, and its diversified customer base. The financials, however, tell a different story.
This is the kind of transaction that appears when a sponsor has assembled a platform using expensive, short-dated debt and needs to replace it with cheaper, longer-dated capital before the debt matures or the interest burden crushes the income statement. The IPO is not a vote of confidence in the data center market's future. It is a vote of confidence in the idea that public equity markets are still willing to fund capital-intensive infrastructure stories at a premium to book value.
The next thing to watch is the pricing. If Csquare prices at the low end or below the range, it will signal that public investors are demanding a discount for the debt overhang and the operating losses. If it prices at the high end, it will confirm that the AI infrastructure narrative still commands a premium, even for a company that needs the IPO primarily to fix its balance sheet.
Either way, the deal is not proof that data center IPOs are back. It is proof that even the best-sponsored platforms need to refinance their construction debt, and the public market is the only source of capital large enough to do it.