Intel's Comeback Hits Its First Air Pocket: -12% in a Week on Terafab Doubts, the OpenAI Scare, and a Valuation With No Room for Error

Intel's Comeback Hits Its First Air Pocket: -12% in a Week on Terafab Doubts, the OpenAI Scare, and a Valuation With No Room for Error

The Intel comeback (see prior revisions: +250% YTD, fifth-best S&P 500 performer, $650B market cap two weeks ago) has run into its first real correction. The market view shows INTC at $104.70, down 12.3% over five days with RSI at 29 — oversold — and the market cap down to $553B, roughly $100B of value erased in a week. The stock is still +26.1% above its 200-day SMA and +8.0% over three months, so this is a correction inside an uptrend, not a reversal of it. Yet.

Driver 1 — Terafab/TSMC: the foundry thesis gets a rival at the table. Intel joined Musk's Texas Terafab project April 7 as the only named manufacturing partner, with Musk confirming Terafab would use Intel's 14A (1.4nm-class) process. That exclusivity assumption broke this week. Per hwbusters, Musk confirmed the TSMC talks in four words: "Just discussions, but something may come of it." The risk, per TheStreet citing CNBC: "If TSMC joins, Intel's process could be dropped in favor of TSMC's own technology." Tim Culpan's reporting suggests the most likely TSMC outcome would be TSMC owning and operating the new factory itself. CEO Lip-Bu Tan responded that Intel "will keep working with Musk on Terafab" (Reuters, Oct 7). The stakes: Intel Foundry generated $5.8B last quarter, up 31% YoY — mostly internal — and 14A's outside-customer validation is the core of the foundry-turnaround bull case.

Driver 2 — the OpenAI revenue scare hit Intel disproportionately. In Thursday's Oct 8 chip selloff (see The AI Trade Passes Its First Live Stress Test: OpenAI's $20B Revenue Whipsaw), Intel was among the biggest casualties — down more than 5-6% in a single session — because its rally was built partly on the agentic-compute demand narrative that an OpenAI revenue gap undermines directly¹.

Driver 3 — macro and mean reversion. MarketBeat's Oct 9 recap: "Intel shares fell 2.2% to $104.70 amid a broader selloff in growth-oriented semiconductor stocks driven by higher Treasury yields and rising oil prices. The stock's sharp year-to-date rally makes it particularly vulnerable to valuation concerns and profit-taking." The same piece flags the near-term setup: analysts see the Oct 29 Q3 report turning on gross margin — prior guidance ~42%, only modestly above Q2's 41.8% — rather than revenue, which has recently exceeded guidance. One quirk in the fundamentals worth flagging: the June quarter showed an $11.03B net loss and negative EBITDA alongside the $0.30 EPS beat — a large one-time, likely non-cash item that GAAP/non-GAAP divergence explains; worth drilling in the 10-Q before Oct 29.

Why this matters: the bull case bought Intel as a foundry-validation story at 63x forward earnings (per the pre-pullback reporting); the bear case now has three concrete attack vectors — Terafab dilution, AI-demand credibility, and a demanding multiple with -19.8% TTM profit margin. The RSI-29 washout into an Oct 29 print with a ~42% GM bar is a classic setup: the stock needs the foundry margin trajectory to justify the multiple, and Terafab was supposed to be the proof that 14A has a marquee outside customer. "Just discussions" was enough to take $100B off the table.