Arm heads into its fiscal second-quarter report with two stories pulling in opposite directions. Its data center business and the new ArmAGI silicon effort are gaining momentum, while its core smartphone royalty stream is being squeezed by rising memory prices. Last quarter management raised its sights on artificial intelligence and lowered them on handsets in the same call. The stock has lost 13.4% since then, so the market has so far given more weight to the handset worries.

Wall Street expects non-GAAP earnings of $0.48 a share on revenue of about $1.37 billion, which would be growth of roughly 20.7% from a year ago. Management guided to $0.43 to $0.51 in EPS and $1.33 billion to $1.43 billion in revenue. That puts the revenue consensus slightly below the midpoint and the EPS estimate a touch above it. When the guidance was issued, the Street sat at $0.44, so estimates have drifted higher since. The Earnings Whisper of $0.51 sits at the top of the guided range, which suggests the informal bar is a clean beat rather than a result that simply lands inside guidance. Hitting $0.48 would still leave earnings well below the roughly $0.60 Arm earned two quarters ago, a reminder of how much heavier spending has weighed on profits.

The royalty line is the most important test. Management cut its full-year royalty growth outlook from about 20% to the high teens and guided this quarter to only low-to-mid-teens growth. It also said memory-driven weakness had spread from budget phones into mid-range and premium devices. That spread is the key issue, because premium handsets carry Arm's richest per-unit royalties. If royalty growth comes in at the low end of the range, or if management signals further erosion in upper tiers, the cut will look like the start of a trend rather than a one-time reset. Royalties holding in the mid-teens, with cloud strength offsetting phones, would support management's view that diversification is working.

The offset is data center. Royalties there have more than doubled year over year for three straight quarters, helped by NVIDIA's Vera entering production and AWS planning tens of millions of Graviton 5 cores. A fourth quarter of that kind of growth would be strong evidence that the mix shift toward the data center is real. Licensing should also be strong: management guided to roughly 30% growth, above its earlier framing, after a record first quarter. On ArmAGI, the company said it had secured manufacturing capacity for an initial $1 billion of demand, against a pipeline above $2 billion, and that its confidence in delivering had improved. This quarter should show whether shipments are actually moving, not just whether more customers have signed on.

Margins are where good news could still disappoint. Non-GAAP operating margin fell from 49% to about 41% last quarter as R&D stayed elevated. First-generation ArmAGI chips carry gross margins in the high 30s to low 40s, well below Arm's traditional licensing economics. Faster silicon revenue could therefore dilute blended margins even as revenue rises. Investors will want some evidence that spending growth is levelling off, or that operating leverage is coming back into view.

Sentiment has only edged from mildly bearish to roughly neutral, and the shares have trailed the S&P 500 by more than 17 percentage points. At about $223, the stock trades just below its 200-day moving average and sits much closer to the $208 post-earnings low than the $337 high. That setup lowers the emotional bar, but it does not lower the numerical one. The report comes down to a single question: can data center royalties and ArmAGI shipments grow fast enough to cover a handset slowdown that is reaching premium phones? If they can, the two-track story holds; if royalties slip again and margins do not stabilize, the market's skepticism will look justified.