Expectations that Intel would shake off its slump and show signs of a comeback have sharply cooled within a few months.

After President Donald Trump announced that the U.S. government would buy a 10 percent stake in Intel, investors expected demand to surge. That helped Intel shares, which had been tumbling, turn higher. Reports that SoftBank and Nvidia would invest in Intel also lifted the stock, which rose more than 120 percent in five months.

The Information reported that Intel had traded at a lower valuation than Taiwan's TSMC, the world's largest contract chipmaker, but has recently jumped to about 20 times projected EBITDA. That is far higher than TSMC's 12.5 times.

Given that Intel is expected to see revenue decline in both 2025 and 2026 and has not posted notable results in the AI chip market, the situation is seen as reflecting heightened expectations for the future.

But the rise did not last long. On Jan. 22, local time, Intel shares plunged 17% as soon as the company released fourth-quarter results, wiping out $46 billion in value.

Investor confidence fell on concerns that Intel had not properly prepared for increased demand. The Wall Street Journal recently reported that Intel failed to respond adequately to a surge in orders for processors used in AI data centres after it cut capacity at older production lines for several months.

Stacy Rasgon (스테이시 라스곤), a semiconductor analyst at Bernstein, said, "The stock soared on sentiment and tweets." He said, "In theory, it should be ready to take advantage of this demand, but it isn't. It's really unfortunate."

Intel drew support from President Trump, secured a $2 billion investment from SoftBank and signed a custom chip contract with Nvidia. At that point, expectations were high that Intel would now show progress.

Investors watched the fourth-quarter results hoping Intel would show advances in its money-losing foundry business or gain more from the AI boom reshaping the technology industry, but expectations quickly turned to disappointment.

The fourth-quarter release showed Intel's operational problems remain serious, which was seen as fuelling selling. The WSJ reported that Intel still has not secured a single customer for 14A, its next-generation chipmaking technology.

Executives inside Intel are reportedly urging patience over winning new customers for the 14A manufacturing process. 14A is a key technology Intel is counting on to catch up in AI computing, but full-scale mass production appears unlikely until at least 2028 or 2029.

The current situation is also linked to Intel taking a conservative stance in a chicken-and-egg dilemma. Intel chose to delay investment in new facilities until it secures customers. That contrasts with TSMC, which is pouring money into expanding chipmaking facilities in the United States.

Critics say Intel could have benefited significantly as CPU demand surged in AI data centres, but missed the opportunity because it lacked supply capacity.

The WSJ reported that since the second half of last year, companies such as OpenAI, Amazon Web Services and Google have begun to realise deploying AI models requires more powerful, high-performance CPUs than they originally thought.

Intel's challenges are also tied to the legacy of former CEO Pat Gelsinger (팻 갤싱어), who left the company at the end of 2024. Under Gelsinger, Intel moved aggressively to invest in foundry infrastructure, pouring large sums into it even without securing customers. During that period, Intel was lagging in the AI chip race.

In that environment, CEO Lip-Bu Tan (립부 탄), who took the helm last year, scrapped plans for a multibillion-dollar chip manufacturing facility in Europe and delayed plans to build a semiconductor plant in Ohio, in addition to a 15% workforce cut. Intel is also limiting spending on older technology and taking a more cautious approach to expanding leading-edge chip production capacity as it cuts costs, the WSJ reported.