On the latest episode of the Data Center Frontier Show podcast, DCF Editor in Chief Matt Vincent speaks with Melissa Kalka, M&A and private equity partner, and Kimberly McGrath, real estate partner at Kirkland & Ellis, about how capital, power, and deal strategy are changing in the AI data center era.

Their core message is clear. Capital is still flowing into digital infrastructure, but the market has become far more disciplined. Investors are no longer simply chasing land or growth stories. They are digging deeper into platform quality, delivery track record, contractual structure, and above all, power certainty.

That last point now sits at the center of nearly every transaction. As AI workloads push development from 20 MW and 48 MW deals toward 100 MW, 500 MW, and even gigawatt-scale campuses, power availability has become the first screen in diligence. A site may have land and entitlements, but without credible access to power, it may struggle to attract customers, financing, or buyers.

The conversation also underscores how AI has changed the asset class itself. Data centers are no longer being evaluated strictly as real estate. They are increasingly underwritten as a hybrid of real estate and infrastructure, with longer hold periods, shared campus systems, and more complex capital stacks.

That dynamic is driving new financing structures, including more private credit activity, more infrastructure-style investment, and growing interest in open-ended and perpetual vehicles for long-term ownership.

Powered land, meanwhile, has emerged as an asset category of its own. In a market where development pipelines remain robust and hyperscalers are pursuing massive capacity expansions, sites with large increments of secured power are drawing intense interest.

Kalka and McGrath also explain that customer contracts now function as a key part of financing infrastructure. Lease and colocation agreements are being negotiated with greater attention to lender expectations, long-term revenue stability, and risk allocation around power delivery and development timing.

For developers and operators, one of the biggest lessons is that structure matters early. Projects need to be organized from the outset in ways that make them financeable, investable, and divisible as platforms mature. Just as important, these deals now require extraordinary coordination across legal, real estate, regulatory, financing, environmental, and community stakeholders.

The episode offers a timely look at a market moving out of its speculative phase and into a more demanding period defined by execution. In the AI era, the winners will not simply be those who raise capital fastest, but those who can align capital, contracts, land, and power into a credible path to delivery.

From Land Grab to Capital Discipline: Kirkland & Ellis Explains How AI Is Transforming Data Center Finance

The Data Center Frontier Show

From Land Grab to Capital Discipline: Kirkland & Ellis Explains How AI Is Transforming Data Center Finance