Intel’s tick-tock isn’t coming back, and everything else I just learned

Today, on the company’s Q3 2025 earnings call, where Intel saw its first earnings in nearly two years due primarily to those lifelines, CEO Lip-Bu Tan and CFO David Zinsner explained how the company doesn’t have enough chips yet. It’s currently seeing a shortage that it expects to peak in the first quarter of next year — in the meantime, leaders say they will prioritize AI server chips over some consumer processors as they address supply and demand.

“We expect CCG [Intel’s consumer chips] Be down modestly and DCAI [Intel’s server chips] “We must be aggressive because we prioritize the capacity of server shipments over entry-level client parts,” Intel says. Tan revealed today that Intel will also launch new AI-powered GPUs every year, following Nvidia and AMD in changing its traditional pace to meet the huge demand for AI servers. It’s not clear what that might mean for those hoping for more gaming GPUs from Intel.

While all eyes are on Intel’s new Panther Lake and its 18A process to show the world that it can still make the most powerful consumer PC chips and manufacture them in-house, the company has confirmed that it will only launch one Panther Lake SKU this year and slowly roll out the others in 2026. Here’s another possible reason for that: Zinsner hinted today that Panther Lake will be a “very expensive” product initially, and Intel will have to push its current product instead, Lunar Lake chips, “for at least the first half of the year.”

While Intel has repeatedly disputed the notion that its 18A operation had poor returns, the company admitted to investors and analysts today that it’s not poised for a huge financial hit either: The returns are “enough to address supply but not where we need them to be to achieve the right level of margins,” Zinsner says, suggesting it could be 2026, or even 2027 for an “acceptable level of returns” in This regard.

For now, Intel will “work closely with customers to maximize our available production, including adjusting pricing and mix, to shift demand toward products for which we have supply and for which we have demand” — which sounds like gouging the prices it charges PC makers to keep Intel inside their PCs and direct them to Lunar Lake parts rather than hot new parts. Tan confirmed today that he will not invest in more production capacity unless there is “committed external demand,” and Zinsner says investments in production capacity next year “will not significantly change the outlook.”

Intel says the 18A will be a “long-life node” that will power “at least the next three generations of client and server products.” If you were hoping for a return to the TikTok days where Intel alternated between shrinking its chips and launching new architectures every generation, that’s not happening here.

But that doesn’t mean Intel will cancel its next node, the Intel 14A, as it warned. Tan suggested today that customers had stepped in to rescue 14A, and Intel said the company was “pleased and more confident” with it, and Zinsner says it was not only a “good start” but was better than 18A at this stage “in terms of performance and returns.”