Corning vs. Honeywell International: Which Technology Stock Is a Better Buy in 2026?

Corning reported FY2025 revenue of nearly $15.6B, up 19.1%, net income of about $1.6B, and $1.4B of free cash flow, positioning it for AI, data-center and optical-connectivity growth. Honeywell generated larger FY2025 revenue of $37.4B, up 7.8%, with $4.7B of net income and $5.4B of free cash flow, while trading at lower valuation multiples of 25.5x forward P/E and 1.9x sales versus Corning's 47.6x and 7.9x. The article favors Honeywell as the more defensive choice but views Corning as a higher-risk, speculative AI-infrastructure opportunity given customer concentration, cyclicality and leverage.

Analysis

The relevant question is not AI exposure but incremental optical-connectivity intensity per dollar of hyperscaler capex. GLW monetizes a narrower bottleneck—fiber density, connectivity upgrades, and specialty materials—so it can outperform when data-center buildouts shift from GPU procurement toward network completion. That sensitivity also makes its earnings multiple vulnerable: at roughly 48x forward earnings, even a modest deceleration in optical-fiber orders or a pause in carrier spending could produce material multiple compression before any annual revenue shortfall is visible.

HON’s apparent valuation discount should not be read as a clean defensive arbitrage. Portfolio separations can unlock a higher-quality aerospace/automation multiple over 6-18 months, but the near-term setup includes stranded-cost, execution, and cash-use uncertainty; investors should demand evidence that remaining-segment margins and order growth improve rather than merely accepting pro forma management targets. Aerospace aftermarket resilience provides downside support, while industrial automation is the more macro-sensitive leg if global PMIs weaken.

Consensus likely overstates GLW’s direct read-through to NVDA/AMZN capex. Fiber demand is influenced by network architecture, customer inventory, carrier budgets, and deployment timing; capex announcements do not translate one-for-one into GLW shipments. Conversely, the market may underappreciate a 2026-27 optical upgrade cycle if AI clusters require increasingly dense interconnects, creating upside only if management converts growth into operating leverage rather than absorbing it through energy, materials, and capacity costs.

Near term, this is a monitoring event rather than a broad AI-infrastructure signal. The key 1-3 month catalysts are GLW optical-communications backlog/order commentary and HON separation disclosures; the 6-18 month determinant is whether GLW sustains margin expansion through the connectivity cycle versus HON demonstrating cleaner organic growth post-repositioning.

Key Decisions for Investors

• Maintain a modest long GLW only on pullbacks or following independently confirmed optical-communications order acceleration; size as a high-beta satellite position, not a core AI holding. Thesis is invalidated by two consecutive quarters of segment order deceleration, margin erosion, or evidence that major customers are digesting inventory.
• Prefer a 6-12 month pair trade: long HON / short GLW in roughly beta-neutral dollars if GLW’s valuation premium remains near current levels without upward EPS revisions. The trade captures HON portfolio-unlock optionality and GLW’s elevated expectations; stop if GLW raises full-year earnings materially on verified fiber demand or HON separation costs impair guidance.
• Before adding HON, wait for separation filings that quantify dis-synergies, leverage allocation, and stand-alone margin targets. A favorable setup requires pro forma free-cash-flow conversion and aerospace aftermarket growth to offset any automation slowdown; absent that disclosure, treat the lower headline multiple as incomplete information.
• Track VZ and other carrier-capex commentary alongside hyperscaler networking spend. A divergence—strong cloud capex but weak carrier fiber orders—would favor reducing GLW exposure, because its end-market diversification does not eliminate telecom-cycle risk.