Dr. Kamata's Grain-of-Salt Laser Chip Breaks Power-Size Barrier

Without light, there would be no life on earth – and without photonics, the science and technology of controlling light, no modern technology. 🤔 On May 16, we celebrate the International Day of Light – a UNESCO initiative honoring the role of light in science, culture, and our daily lives. At ZEISS, light is at the heart of everything we do. Since 1846, we have been shaping the science of light – photonics – across all our business areas to drive innovation: ZEISS Semiconductor Manufacturing Technology: Extreme ultraviolet (EUV) light with wavelengths of just 13.5 nanometers enables the structuring of the tiniest chip features, the foundation for powerful smartphones and AI systems. ZEISS Microscopy: Fluorescent light makes individual proteins visible in living cells, helping researchers understand diseases at the molecular level. ZEISS Industrial Quality Solutions: Structured blue light captures components in 3D with micrometer precision, ensuring top quality from turbine blades to car body parts and supporting quality assurance of precision mechanics in photonics applications with various technologies. ZEISS Medical Technology: Photonics enhances diagnostics, surgery, and therapy. From imaging technologies to minimally invasive procedures, light is used in a wide range of medical applications. ZEISS Vision: Through precise refraction of light, advanced lenses can not only improve but also protect vision and visual health and even support a more natural development of the eye. ZEISS Planetariums: Star projectors show you the starry night sky in brilliant optical detail, and digital projections allow you to travel through time and space – all fully immersive, inspiring fascination for science and research. Light connects us all. According to UNESCO and EU initiatives, photonics is considered a key enabling technology for sustainable development, digital infrastructure and modern industry. #ZEISS #Innovation #InternationalDayOfLight #Photonics

The optical industry in 2026 is being fundamentally rewired by three converging macro trends: the AI infrastructure boom, the accelerating global myopia crisis, and the commercial maturation of smart eyewear. First, AI’s insatiable compute demands have transformed advanced optics from a niche component into a critical bottleneck solution, with optical interconnects and silicon photonics now essential for scaling low-latency, high-bandwidth data centers—projected to grow at a 28.4% CAGR through 2030 as companies like NVIDIA, Lumentum, and Coherent accelerate in-package optical I/O deployment to bypass copper’s thermal and speed limits. Simultaneously, the myopia epidemic has shifted from a clinical concern to a multi-billion-dollar care imperative; the WHO and Brien Holden Vision Institute estimate that nearly 5 billion people will be myopic by 2050, with daily disposable myopia-control lenses and orthokeratology seeing triple-digit adoption growth in APAC and North America as regulatory pathways streamline and consumer awareness outpaces traditional single-vision correction. Finally, smart eyewear has crossed the chasm from prototype to prescription-ready mainstream, driven by Meta’s Ray-Ban Meta line (which saw 203% YoY growth in key European markets through early 2025) and EssilorLuxottica’s aggressive optical-first design strategy that seamlessly integrates AI audio, spatial computing, and health monitoring into everyday frames—paving the way for a $15B+ smart eyewear market by 2030. For professionals across optics, photonics, and vision care, 2026 isn’t just about better lenses or faster chips; it’s about recognizing that sight, data, and human-computer interaction are now inextricably linked, and those who align R&D, clinical practice, and retail strategy around this convergence will define the next decade of vision innovation. #OpticalIndustry #VisionCare #AIInfrastructure #MyopiaManagement #SmartEyewear #Photonics

The first generation of our (MSEIS) is now online—marking an exciting step toward next-generation chip-scale imaging. By integrating MEMS-enabled precise spatial modulation with image sensing, MSEIS breaks through conventional pixel-size limitations and unlocks significantly enhanced resolution. This is just the beginning—more innovative and impactful MEMS applications are on the way.

A new device generates phonons, tiny units of sound, at ultra-low temperatures, opening the door to “sound lasers” for communication, imaging, and next-generation technologies. https://lnkd.in/dqB3EU67

Diffractive Optical Element Market Size Worth $ 463.93 million by 2032 | CAGR: 9.8% The diffractive optical elements market is growing steadily due to increasing demand for advanced optical systems in various industries. These elements are used to manipulate light in applications such as imaging, laser systems, and telecommunications. Rising adoption in consumer electronics and medical devices is a key driver. Technological advancements are enabling improved efficiency and miniaturization. Growing use in augmented and virtual reality applications is also boosting demand. High precision manufacturing requirements can limit scalability. However, increasing research and development activities are supporting market expansion. The market outlook remains positive. 𝐑𝐞𝐚𝐝 𝐌𝐨𝐫𝐞: https://lnkd.in/dG8_2-e9 Polaris Market Research & Consulting, Inc. #DiffractiveOptics #OpticalEngineering #PhotonicsTech #LaserOptics #OpticsMarket

Photonics doesn’t pivot on technology. It pivots on market pull. Three years ago, photonics startups raising serious capital were building photonic compute. Today, the ones getting acquired are building photonic interconnect. -> The technology didn't suddenly change. The buyer did. Lightmatter started as a photonic AI accelerator. Today its public messaging emphasizes Passage, the photonic interconnect platform. Celestial AI built photonic compute, repositioned around Photonic Fabric for interconnect, and was acquired by Marvell Technology for up to $5.5B. Same teams. Same silicon photonics expertise. Different commercial context. Different valuation rationale. The same pattern played out in dual-use defense. EU defense spending rose from €218B in 2021 to €343B in 2024. European defense tech VC funding rose from €200M in 2021 to €2.6B in 2025. Photonics companies selling lasers and optics into industrial markets suddenly found defense customers as their most urgent buyers. Same lasers, same optics. Different cycle, different multiples. This is one of the core lessons in photonics investing: -> A technology can sit underpriced for years until the right market pulls it into relevance. In my experience, the majority of photonics companies now in AI interconnect or dual-use defense were not native to those categories. They pivoted in. That's not a weakness. It's a signal. The teams that read the pivot early and moved fast are now inside the wave. -> Four ingredients separate companies inside the wave from the ones left behind: 1. Reading market pull early and adjusting the company roadmap swiftly. 2. Full-stack architecture, not a single component. Layers get repriced before parts. 3. Manufacturability and volume readiness. Acquirers pay premium for crossing the prototype-to-production gap. 4. Named customers with paid pilots or design-ins. Commercial pull, not technical push. -> A third pivot may be forming - Physical AI. Humanoid robotics, mobile manipulation, autonomous industrial systems could pull photonics the same way AI infrastructure did: ● Sensing the outside world: event-based cameras, compact LiDAR, hyperspectral, VCSELs. ● Inference at the edge: on-device optical compute, short-reach interconnect. These technologies spent years searching for traction in automotive, industrial, and consumer markets that never pulled hard enough. -> Now robotics may create the buyer. NVIDIA is again positioning itself as one of the architects. The platform pattern that pulled optics into AI infrastructure is starting to appear in robotics. The companies positioned across the full sensing stack, with manufacturability and named pilots, will be repriced first. Not because the technology changed. Because the buyer did. Which photonics category gets pulled hardest if Physical AI delivers - sensing, edge inference, short-reach interconnect or something else? #photonics #deeptech #venturecapital #physicalAI #patternrecognition

Unlocking Hidden Signal Layers of Radar Sensors for AI Models After IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2026 in Barcelona, I am happy to share the final poster of our demo: SensingSP™: Digital Twin for 4D Imaging Radar Design, Simulation, Signal Processing, and AI Start from high-level sensing requirements such as maximum range, range resolution, angular resolution, field of view, update rate, velocity resolution, and detection probability, and translate them into detailed radar parameters such as FMCW waveform, bandwidth, chirp slope, PRI/PRF, ADC sampling, transmit power, antenna configuration, and MIMO strategy. After the radar is designed, the digital twin performs dynamic ray tracing and propagation modeling. For each radar acquisition, it computes path length, path-length change, delay, Doppler shift, phase, and amplitude consistently over time and across TX/RX channels. The demo also covers adaptive radar detection, array processing, and radar array design, including examples inspired by #TexasInstruments and #HUBERSUHNER radar configurations, as well as newly designed arrays and their performance in 2D angular imaging, namely azimuth and elevation, as two of the key dimensions in 4D+ imaging radar. A key motivation is radar AI. Most commercial radar sensors expose mainly the final detection point cloud. However, many useful signal layers exist before detection, including raw ADC data, range profiles, range-Doppler maps, range-angle maps, beamforming outputs, and range-azimuth-elevation representations. SensingSP, the radar engine at the heart of LuxTwin, makes these hidden radar signal layers accessible inside the digital twin. This allows AI models to be studied not only on sparse point clouds, but also on richer pre-detection radar representations. The full ICASSP demo series: https://lnkd.in/eUZbUfhB https://lnkd.in/eQRRVvE8 https://lnkd.in/eRHndB5u https://lnkd.in/eF-MiW9U https://lnkd.in/eSYudkdW https://lnkd.in/et2y6Tpv https://lnkd.in/edtrh92f https://lnkd.in/ewBxEb2q https://lnkd.in/ezWZxHWA https://lnkd.in/eXzjkPSS Earlier related post: Hands-On Radar Simulation with Blender   https://lnkd.in/e4m9P-fJ Many thanks to everyone who visited the demo and discussed radar digital twins, 4D imaging radar, signal processing, and radar AI with us. I especially enjoyed the long discussions with Shahrokh Valaee and Pu (Perry) Wang. This work was carried out with Bhavani Shankar M. R., Björn Ottersten, and Thomas Stifter, within the SPARC Research Group at SnT, University of Luxembourg, in collaboration with IEE S.A., Luxembourg. #ICASSP2026 #Radar #SignalProcessing #DigitalTwin #4DImagingRadar #FMCWRadar #MIMO #RadarAI #AutonomousDriving #AI #SensingSP #SnT #UniversityOfLuxembourg

WATCH: 210ms vs. <16ms — The SADA Telesurgery Video Demo ⚡ In telesurgery, "average" speed is a vanity metric. The only number that matters is the worst-case spike—the Tail Latency. The Problem: Best-Effort OS Latency On standard Linux/Android, AI inference and control loops run as "best-effort." When a system hits a 210ms spike, the robotic instrument lags behind the surgeon’s hand. The Result: Overshooting, tissue damage, and unacceptable patient risk. This is the #1 barrier to scaling remote surgery. The Solution: SADA’s Deterministic Zone With Source-Aware Deterministic Architecture (SADA), the vision → NPU → actuator path runs in a reserved, shielded lane. The Result: Every control loop is guaranteed to complete in <16ms. No jitter. No stutter. Just surgeon-level precision, remotely. The Difference:  ❌ Legacy Path: 210ms lag = Tissue damage = Clinically non-viable. ✅ SADA Path: <16ms guaranteed = Perfect sync = Clinically viable. Core Insight: Real-time AI isn't about how fast it runs usually. It’s about how it performs at its worst. For robotics, drones, and surgery, the worst-case is the only number that matters. SADA makes that worst-case <16ms. Inventor: Gandhi Davuluri Status: U.S. Patent Pending (No. 64/041,227) Interested in the architectural breakdown? Message me for the Technical Whitepaper (available under MNDA). #Robotics #Telesurgery #EdgeAI #EmbeddedSystems #Innovation #SADA #RealTimeAI Qualcomm NVIDIA Arm Broadcom Texas Instruments Intel

Why Is the Photonics Market Powering the Next Wave of Global Technology Innovation? 🔬💡 From AI data centers and high-speed telecom to LiDAR, AR/VR, medical imaging, and semiconductor manufacturing — photonics is becoming the foundation of next-generation digital infrastructure. 📈 The Global Photonics Market is projected to grow from USD 839.4 Billion in 2026 to USD 1.73 Trillion by 2035, expanding at a CAGR of 8.43% during the forecast period. 🚀 Key Factors Driving Market Expansion: ✅ Asia-Pacific accounts for 45% of global market share ✅ Visible-light devices represent 50% of photonics deployment ✅ LEDs hold 32% of total product share worldwide ✅ Silicon materials dominate with 40% market share ✅ Data communication captures 43% of silicon photonics usage ✅ Rapid growth in AI, telecom, and high-speed optical interconnects ⚡ Emerging Trends Reshaping the Industry: • Rising adoption of silicon photonics in data centers • Growing demand for LiDAR and autonomous vehicle sensing • Expansion of AR/VR and micro-LED display technologies • Increasing deployment of optical communication networks • Strong growth in medical imaging and healthcare photonics 🌍 Regional Market Insights: 🇨🇳 Asia-Pacific leads global production and semiconductor integration 🇺🇸 North America dominates silicon photonics innovation 🇪🇺 Europe expanding automotive and industrial photonics adoption 🇸🇦 Middle East increasing renewable and smart-city photonics investments 🏆 Top Players in the Photonics Market: Hamamatsu Photonics France Intel Infinera Innolume GmbH Finisar Corporation Philips IPG Photonics OSRAM Hewlett Packard Enterprise Molex IBM 👇 Sample Link is available in comment 👇 💡 As demand for faster connectivity, advanced sensing, semiconductor innovation, smart displays, and AI-powered infrastructure continues to rise, photonics technologies are set to become one of the most critical enablers of the global digital economy. #Photonics #SiliconPhotonics #OpticalCommunication #Semiconductor #LiDAR #DataCenters #Telecom #ARVR #AIInfrastructure #MedicalImaging #Optoelectronics #TechnologyInnovation #MarketResearch #SmartTechnology #FutureTech

**Global Photonics Giants: Who Is Shaping the Future of Light-Based Technologies?** Photonics is quietly powering the modern world — from AI data centers and telecom networks to healthcare, defense, and advanced manufacturing. Today, a few key regions are emerging as **global photonics giants**, each with distinct strengths: 🌏 **China** A manufacturing powerhouse with rapid growth in optical communications, LiDAR, lasers, and photonic chips. Massive domestic demand and strong policy support are accelerating its shift toward innovation leadership. 🇺🇸 **United States** Home to cutting-edge R&D in silicon photonics, quantum optics, and advanced semiconductor technologies. Strong ecosystem of startups, academia, and tech giants. 🇩🇪 **Germany** A leader in precision optics, industrial lasers, and photonics-enabled manufacturing. Known for engineering excellence and high-end optical systems. 🇯🇵 **Japan** Renowned for high-quality optical components, imaging systems, sensors, and materials science innovations. 🇰🇷 **South Korea** Dominates in display technologies (OLED, microLED) and is rapidly advancing in semiconductor-linked photonics. 🇹🇷 **Turkey** An emerging player, particularly strong in defense electro-optics, fiber lasers, and optical systems for UAVs and surveillance. 🔍 **What’s the big picture?** Photonics is becoming the backbone of the **AI era**, enabling faster data transfer, smarter sensing, and more efficient energy systems. The next decade will likely be defined by: • Integrated photonics & optical chips • AI-driven optical networks • Quantum photonics • Advanced sensing & imaging The real question is not just who leads today — but **who will define the standards and breakthroughs of tomorrow**. #Photonics #DeepTech #Innovation #AI #Semiconductors #Optics #GlobalTech