How significant is the impact of the "glass bridge"? Wall Street: FAU manufacturers face disruptive pressure, while AI optical module companies are relatively safe for now.

How significant is the impact of the "glass bridge"? Wall Street: FAU manufacturers face disruptive pressure, while AI optical module companies are relatively safe for now.

```

The official debut of Corning GlassBridge "glass bridge" technology has attracted significant attention within the AI optical communications industry chain, but evaluations from Wall Street are much more moderate than market concerns: fiber array unit (FAU) manufacturers face real long-term disruptive pressures in the development path of CPO, while the impact on AI optical transceiver module companies will be limited in the next one to two years, leaving the overall industry relatively within a safe zone.

On June 24, 2026, Corning officially released GlassBridge at the AI Data Center Optical Communication and Interconnect Technology Conference in Seoul. This is a fiber-to-photonic integrated circuit (PIC) connection platform based on wafer-level ion exchange (IOX) waveguides, which directly couples optical fibers to PICs and, by using wafer-level passive alignment, enhances the density and scalability of fiber interfaces. It is seen as an evolutionary alternative for iFAU (intelligent fiber array unit) within the CPO architecture.

According to Morgan Stanley, existing CPO solutions (including Quantum and Spectrum series) have already been mass-produced and finalized, and are expected to remain unaffected by GlassBridge; Citi Research also points out that the 2H27 CPO solution for the Rubin Ultra Kyber platform will complete design confirmation in the second half of 2026, and is similarly unlikely to be replaced by new technology. Both institutions judge that the probability of GlassBridge causing substantial commercial impact in the next one to two years is extremely low.

The core concerns of both institutions focus on a longer time horizon: As CPO and NPO architectures are gradually commercialized between 2028 and 2030, GlassBridge will compete with incremental FAU in new deployment projects. The ongoing uncertainty in GlassBridge’s commercialization timetable will cause the stock prices of companies with high FAU business option value to continue to face volatility pressure.

GlassBridge Technology Analysis: Wafer Process Challenges Traditional Precision Assembly

Corning GlassBridge embeds wafer-level IOX waveguides into the glass connector body, bridging the size mismatch between silicon photonic chip waveguides and fiber cores, achieving high-density direct coupling of fibers to PICs, and supporting detachable system integration.

According to Morgan Stanley, when the number of fibers increases significantly in traditional FAU, assembly complexity rises sharply and scalability becomes limited; GlassBridge offers a wafer-level passive alignment solution with higher density and scalability, providing a supplementary or potentially alternative path to existing FAU solutions.

It is noteworthy that GlassBridge did not appear suddenly. Morgan Stanley points out that related news had circulated in the market as early as September 2025, and the technology had already been included in Corning’s $10 billion photonics business plan announced on Analyst Day.

Limited Recent Impact: Switching Costs and Certification Cycles Form Double Barriers

Citi Research emphasizes that there is substantial uncertainty for GlassBridge before it completes qualification certification, reliability testing, and yield ramp-up. It must also pass the CPO/NPO evaluation period as well as onsite reliability verification in high-speed AI cluster systems, and is not expected to cause significant impact in the next one to two years.

On the ecosystem adaptation side, adopting GlassBridge requires PIC designers to reconstruct optical interface layouts, redesign spot size converters, and modify bump structures. Citi Research believes these switching costs create strong technical inertia, further limiting GlassBridge’s short-term penetration.

Morgan Stanley likewise points out that the actual commercialization timetable for GlassBridge remains unclear, and warns that this will create volatility in the valuation for companies with significant FAU business exposure.

AI Optical Module Companies Are Relatively Safe: NPO Applications Provide Hedging Buffer

For AI optical transceiver module companies, Morgan Stanley offers a relatively optimistic assessment: GlassBridge can be used in both CPO and NPO architectures, and broader NPO applications are expected to partially offset potential risks on the CPO side. Therefore, the impact on AI optical transceiver module companies is expected to be quite limited.

Citi Research’s assessment of Eoptolink confirms this logic: as an optical transceiver module integrator, GlassBridge for Eoptolink is more a neutral upstream component replacement option, with limited direct substitution impact. The core competitiveness of Dongshan Precision and Accelink comes from active optical chips, which are also basically unaffected by GlassBridge.

FAU Manufacturers Face Disruptive Pressure, Risks for Companies Like TFC Are Relatively Manageable

Companies located in the FAU segment of the supply chain face more direct long-term replacement risk. Morgan Stanley clearly notes that FAU manufacturers are more likely to experience disruptive impact from GlassBridge as CPO evolves.

Taking Tianfu Optical Communication as an example, Citi Research believes its long-term replacement risk in iFAU business is effectively controlled: TFC’s growth engine has shifted from FAU to active optical engines, and Citi expects this segment to contribute over 70% of revenue between 2026 and 2028. Diversified revenue structure effectively hedges GlassBridge’s potential impact. T&S Communications focuses on MT ferrules and fiber distribution boxes (passive optical components). Citi Research points out that due to business decoupling risk with Corning, the company is unlikely to benefit from GlassBridge’s adoption, nor will it face direct impact.

Outlook After 2030: Passive Optical Components May Face Paradigm Shift

Looking at a longer time horizon, Citi Research raises the possibility of structural disruption. As CPO and NPO are gradually commercialized between 2028 and 2030, GlassBridge will mainly compete in new deployment projects with incremental iFAU, without affecting the existing installed base.

After 2030, if wafer-level photonic packaging matures in yield and insertion loss performance, Citi Research believes the passive optical components industry could see a true paradigm shift—from physical micro-assembly to semiconductor-level process integration.

Currently, Morgan Stanley and Citi Research both characterize GlassBridge as a technological evolution direction with highly uncertain timetable. Its most direct short-term market impact is creating valuation volatility for companies with large FAU business exposure, rather than immediate fundamental impact.

Risk warnings and disclaimersThe market has risks, and investment should be made cautiously. This article does not constitute personal investment advice, nor does it take into account the special investment objectives, financial situation, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular circumstances. Investments based on this article are at your own risk. ```