Intel's Foundry Story Reaches a Pivotal Moment: A Potential SK Hynix Deal Meets a $20 Billion Cash Burn

Intel's foundry advances on 18A/14A, but losses persist. SK Hynix evaluates Intel as HBM4E base-die second source, while $20B capital raise fuels transformation.

The semiconductor industry has a habit of serving up irony in generous portions. For Intel, the current paradox is stark: its manufacturing technology is hitting internal milestones faster than planned, while its foundry division is bleeding cash at a pace that would test the patience of even the most committed turnaround investor.

That tension was on full display last week when CFO David Zinsner told the Deutsche Bank technology conference that the 18A process is exceeding internal targets and defect reduction on 14A is progressing ahead of schedule. The same presentation brought news that 2026 capital expenditure guidance is being lifted from $18 billion to $20 billion. Progress, it seems, does not come cheap.

A Potential Customer With Real Weight

The market's attention, however, has been captured by something potentially more consequential than process-node updates. Reports from the Herald Economy and other outlets over the weekend suggest SK Hynix is evaluating Intel Foundry as a second source for the base dies used in its upcoming HBM4E memory generation, including Intel's proprietary EMIB packaging technology.

The logic is straightforward. SK Hynix currently relies on TSMC's 12-nanometer process for the base dies of its HBM4 chips — the carrier layer beneath the memory stacks themselves. With TSMC's base-die costs running three to four times higher than the actual memory core, the appeal of a multi-vendor supply chain is obvious. A second source gives SK Hynix leverage on pricing and reduces single-supplier risk.

The stock market's reaction was muted at best. Intel shares slipped in initial trading, while SK Hynix and TSMC both lost ground at their home exchanges amid broader weakness in Asian tech stocks driven by geopolitical tensions and a wider sell-off in technology names. Investors appear to be taking a wait-and-see approach until the evaluation phase produces an actual contract.

The Numbers That Ground the Narrative

For all the promise of a potential SK Hynix partnership, the foundry division's financial reality remains sobering. The segment posted an operating loss of $2.1 billion in the second quarter of 2026. Bernstein analysts project cumulative operating losses exceeding $10 billion by the end of 2027 before the necessary scale for profitability comes into reach.

That math is now central to how the market prices Intel. On Wednesday, the stock fell 2.1 percent, making it the weakest performer in the Dow Jones. The question hanging over the shares is no longer whether Intel can close the technology gap — that part appears to be on track — but how much patience investors will extend to a company still deep in its investment phase.

HBM4E itself remains in its early stages, with 12-layer samples in testing since June rather than in mass production. A potential SK Hynix order would be a reputational win, but it would not yet move the needle on the balance sheet. The distinction between evaluation and volume business is one the market is clearly keeping in mind.

A Capital Raise With Consequences

Intel's recent $20 billion capital increase — upsized from an initial $15 billion — closed just over a week ago, with 210,526,315 shares sold at $95 each, yielding net proceeds of approximately $19.7 billion. Since the transaction closed, the stock has gained 3.2 percent, suggesting the market broadly welcomes the fresh liquidity.

Bank of America reaffirmed its buy rating on August 12 but trimmed its price target from $160 to $145. Analyst Vivek Arya cited the dilution from the new shares, which he estimates will reduce earnings per share by 4 to 5 percent. The trade-off is clear: Intel is buying time for its expensive transformation, but the cost is a diluted earnings picture.

CEO Lip-Bu Tan has put his own money behind the bet, purchasing roughly $10 million worth of shares at the $95 issue price in early August. It is a vote of confidence in the proposition that the technological catch-up will ultimately outweigh the near-term dilution.

A Stock Caught Between Euphoria and Fatigue

The share price tells a story of remarkable gains and recent exhaustion. At €77.23, the stock sits about 15 percent below its 50-day average of €90.38, reflecting the correction since the quarterly results and the capital increase. The Friday close of €77.27 was roughly 38 percent below the 52-week high, with a relative strength index of 41.2 suggesting the stock is closer to oversold than overbought.

Yet the longer-term picture remains extraordinary. The shares are up 272 percent over twelve months and 146 percent year-to-date, still far above the 52-week low of €20.30. The big move has already happened; what remains to be seen is whether the foundry story can be backed up with actual orders.

Institutional Signals and Strategic Bets

Beyond the SK Hynix speculation, there are other signs that major players are positioning around Intel's story. A 13F filing analysis from August shows that Intel, alongside SpaceX, accounts for roughly 80 percent of Nvidia's $63.4 billion US equity portfolio — with the position nearly unchanged in share count but more than tripled in value. That is not evidence of operational strength, but it is a notable signal that a central industry player is holding a strategic position.

HSBC analyst Frank Lee doubled his price target to $200 in July, explicitly building the foundry valuation into his model. It is a bold wager that today's losses will become tomorrow's market share. Whether that calculation holds will only become clear when the new manufacturing processes enter volume production and customers follow.

The broader picture is one of structural transformation rather than quick turnaround. Recent personnel moves, including the appointment of Dean Jarnac as chief sales officer, record-level investments, and technical progress on 18A and 14A all point to a company in the midst of a fundamental rebuild. Zinsner himself acknowledged on the conference call that foundry losses will persist and that an intensive investment cycle lies ahead — a notably candid admission for a company whose rival AMD has reached a 34 percent market share in servers.

The core question remains whether a company burning billions to become competitive again in three to four years can keep the market's patience intact. The potential SK Hynix deal is not a guarantee of a foundry turnaround, but it is credible evidence that Intel is once again being taken seriously as a second source in semiconductor manufacturing. The opportunity outweighs the risk of renewed disappointment — provided the evaluation becomes a contract in the coming months. Until then, the stock remains what it has been for months: a bet on future promise, with a considerable amount of confidence already priced in.