AI's time machine: How Cisco, Intel, and Corning are reviving fears of the tech crash

Cisco, Intel and Corning are staging comebacks more than 25 years after the dot-com crash, as investor enthusiasm around artificial intelligence drives a renewed rally in once-battered internet-era names. The article highlights a sentiment-driven rotation rather than a fresh fundamental catalyst, suggesting a modest but notable impact on the individual stocks and related tech positioning.

Analysis

The key second-order effect is that this is not just a nostalgia trade; it is a barbell rotation into perceived “picks and shovels” of the AI capex cycle. The market is implicitly saying legacy networking, enterprise compute, and specialty glass/interconnect still have operating leverage to AI buildout even if they are not the platform winners, which can pull capital away from software names whose monetization is still unproven. That matters because these stocks are lower-expectation and often under-owned, so even modest fundamental revisions can drive outsized multiple expansion over the next 1-2 quarters.

The risk is that the rally becomes technically self-reinforcing before the fundamentals catch up. If AI-related orders do not translate into visible backlog growth, margin inflection, or better guidance within the next 1-2 earnings cycles, the move can unwind quickly because positioning likely shifted faster than estimates. The biggest reversal trigger is a broad re-rating from “AI beneficiary” to “old-cycle hardware company,” especially if capex normalization or inventory digestion shows up in distributor/channel data.

Competitive dynamics favor suppliers with installed bases and switching costs, but the real winners may be adjacent beneficiaries not named here: optical components, semiconductor equipment, and contract manufacturers exposed to AI server buildouts. By contrast, the losers are likely higher-multiple AI software names if the market continues preferring cash-flowing infrastructure over long-duration growth stories. The consensus may be missing that these are not pure AI compounders; they are cyclical assets with a narrative overlay, which creates good upside if the story persists but also a sharp air-pocket risk if the flow fades.

Key Decisions for Investors

• Long CSCO vs. short IWM for 4-8 weeks: express a quality/value rotation plus AI-infrastructure exposure; target 8-12% relative outperformance if the market keeps favoring cash-generative beneficiaries.
• Buy INTC on weakness into the next earnings cycle with a 1-2 quarter horizon; risk/reward is attractive only if management can show credible AI/server share gains or foundry traction, otherwise trim quickly on guidance misses.
• Long GLW as an underappreciated AI hardware proxy for 2-3 months; use it as a cleaner way to own the interconnect/optics supply chain, with upside if AI capex breadth expands beyond GPUs.
• Pair trade: long CSCO/GLW basket, short a basket of high-multiple AI software names for 1-3 months; thesis is that capital is rotating toward tangible AI enablers with better near-term cash flow conversion.