Franklin Templeton’s Sara Araghi urges Nvidia to clarify capital deployment plans

The day before Nvidia’s fiscal second-quarter earnings report, one of Wall Street’s more prominent voices in AI infrastructure investing delivered a pointed message: growth numbers alone are not going to cut it anymore.

Sara Araghi, portfolio manager at Franklin Equity, told Bloomberg Television on August 25 that Nvidia needs to present specific plans for how it intends to deploy capital and sustain its spending trajectory.

What the market is actually saying

Nvidia’s forward price-to-earnings ratio sits at roughly 21x for the next 12 months. For context, that is a relatively modest multiple for a company that has spent the past two years posting some of the most dramatic revenue growth in semiconductor history.

Araghi’s concern is precisely that gap between what Nvidia has delivered and what it can credibly promise to deliver going forward. Meeting quarterly earnings targets, she indicated, is no longer the whole conversation. The question is what returns those investments are actually generating.

Nvidia’s fiscal Q2 2027 earnings are due after market close on August 26, with revenue expectations around $92 billion.

The $500 billion infrastructure play

Part of what Araghi is pressing on relates to a significant strategic move Nvidia made in early August 2026. The company announced partnerships with BlackRock and Goldman Sachs aimed at raising more than $500B to finance AI computing infrastructure.

Under some of those arrangements, Nvidia would reportedly provide residual-value support on certain transactions, meaning the company is not just selling chips into this ecosystem but attaching its balance sheet to parts of the financing structure.

She described demand for AI compute as being at an early stage, a view that aligns with most estimates of how far along enterprise AI adoption actually is.

Why this conversation matters beyond one earnings call

The BlackRock and Goldman partnerships mobilize third-party capital to fund infrastructure buildout, rather than loading that burden entirely onto Nvidia’s own balance sheet. But they also create complexity: residual-value commitments, financing structures, and off-balance-sheet exposures that require careful explanation to investors who are used to evaluating Nvidia as a straightforward chip designer and seller.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.