Jabil Inc. has entered into a collaboration with HyperLight Corporation, United Microelectronics Corporation (UMC), and Wavetek to accelerate the deployment of thin‑film lithium niobate (TFLN) photonics in hyperscale AI data‑center interconnects. The partnership combines Jabil’s manufacturing expertise, HyperLight’s TFLN technology, UMC’s semiconductor foundry capabilities, and Wavetek’s photonic integration solutions to bring high‑bandwidth, low‑latency optical links to the next generation of AI workloads.
The collaboration targets the growing demand for optical interconnects that can deliver the bandwidth and energy efficiency required by AI clusters. TFLN offers low loss, an ultrafast electro‑optic effect, and a compact footprint, allowing data‑center operators to reduce power consumption and laser requirements while scaling inter‑connect speeds. By integrating TFLN into rack‑level modules, the partners aim to provide a scalable solution that can be manufactured at data‑center volumes.
Jabil’s FY2025 results underscored the importance of AI to its business. Net revenue reached $29.8 billion, with core diluted EPS of $9.75. The Intelligent Infrastructure segment grew 62% year‑over‑year in Q4 FY2025, driven largely by AI‑related services. For FY2025, Jabil projected AI‑related revenue of $6.5 billion, a figure that reflects the company’s recent guidance update. In Q1 FY2026, revenue rose 18.7% to $8.3 billion, but diluted EPS of $1.35 missed consensus estimates of $2.72, while core diluted EPS of $2.85 beat the $2.55 estimate. Management raised FY2026 revenue guidance to $32.4 billion and core diluted EPS to $11.55, citing continued AI demand and operational leverage.
"We are powering the buildout of AI infrastructure," said Jabil CEO Mike Dastoor. "Fiscal 2026 is off to an excellent start, with Q1 performance ahead of expectations across revenue, core operating margins, and core EPS." Jabil’s General Manager and Vice President of Photonics, Jason Wildt, added, "Hyperscale and AI customers must deploy at scale. They are looking for optical technologies that can be manufactured, integrated, and deployed reliably at data‑center volumes," and that the collaboration "brings together advanced photonics, proven manufacturing at scale, and system integration capabilities, enabling TFLN‑based solutions to be deployed at the rack level for AI and hyperscale customers." HyperLight CEO Mian Zhang noted, "TFLN provides a significant advantage to AI data center networking through reduced power consumption and lower laser requirements. TFLN's fundamental material benefits only grow as optical interconnect speeds scale. With Jabil driving system integration and manufacturing execution, and with UMC and Wavetek supporting scalable device production, we are moving TFLN from innovation into practical, high‑volume deployment for hyperscale data centers today."
The partnership positions Jabil ahead of other electronic manufacturing services that are still building out their photonics capabilities. By leveraging HyperLight’s TFLN platform and UMC’s foundry scale, Jabil can offer a differentiated, energy‑efficient interconnect solution that meets the bandwidth demands of AI workloads. The collaboration also strengthens Jabil’s Intelligent Infrastructure segment, which has become a key growth engine as cloud and AI providers expand their data‑center footprints. With the partnership, Jabil is better positioned to capture a larger share of the lucrative AI data‑center supply chain and to reinforce its engineering‑led moat in a market where speed, reliability, and cost efficiency are critical.
The announcement signals a strategic shift toward high‑margin, high‑growth AI infrastructure. By integrating advanced photonics into its manufacturing portfolio, Jabil is aligning its capabilities with the evolving needs of hyperscale data centers. The partnership is expected to accelerate the commercial deployment of TFLN modules, enhance Jabil’s competitive positioning, and support the company’s broader objective of expanding its Intelligent Infrastructure footprint in the AI era.
The content on EveryTicker is for informational purposes only and should not be construed as financial or investment advice. We are not financial advisors. Consult with a qualified professional before making any investment decisions. Any actions you take based on information from this site are solely at your own risk.