Q1 gross margins beat because they released reserves, not because they fixed the fabs. Then they guided Q2 down. That is not a turnaround. That is financial engineering.
Thesis: Intel is a construction project masquerading as a tech stock. The thesis rests entirely on 18A delivering TSMC parity in 2025. Until then, the P&L is noise. The 'AI PC' narrative is a volume play to fill older fabs, not a margin driver. The separation of Foundry is transparency theater until an external customer signs a meaningful check. You are paying for the privilege of watching Gelsinger spend capex. The trade is binary: 18A works and this is a multi-bagger, or it fails and the government bails out the shell. No edge in guessing right now.
Verdict: HOLD — Conviction: MEDIUM
Catalyst: Release of the 1.0 PDK for Intel 18A in Q2 and subsequent yield data from the Clearwater Forest ramp.
Key Risk: Generation-Skipping Execution Risk. If 18A yields are soft, the entire '5 nodes in 4 years' narrative collapses, leaving them with high-cost fabs and no pricing power.
The Tell: Zinsner admitting the Q1 margin beat was largely due to 'sell-through... on previously reserved material' which 'pulled some of the benefit... into Q1.' They borrowed from the future to beat the present.
Friction Level: HIGH_FRICTION — Bulls see a margin trough and a straight line to 18A leadership. Bears see a 'Say/Do Gap' where accounting adjustments mask structural uncompetitiveness and missing external foundry customers.
Report not found
The report data is no longer available. Please return to the archive.