INNOVATION 创新科技 28 | JNA January/February 2025 transforming the diamond industry and propelling the industry forward,” he continued. AI implications AI is another transformative technological force in the jewellery industry. At Jewellery & Gem WORLD Hong Kong in September 2024, a session on “Practical Applications of Artificial Intelligence (AI) in the Jewellery and Gemstone Industry” delved into fresh AI innovations in diamond and coloured gemstone testing, jewellery design and marketing. Diamond solutions firm Sarine Technology Group launched the first tech-based AI-driven polished diamond grading system in 1992 and, over the last decade, has invested more than US$100 million in developing new technology and services for the diamond trade. David Block, CEO of Sarine, explained, “Most diamonds are graded manually by a gemmologist in a grading lab. Now, technology is being used – which we believe will revolutionise how diamonds are graded over the next five to 10 years.” A major challenge in using AI tech in diamond grading is data collection. Block said copious amounts of data are needed to effectively train and teach a system. Sarine first used AI grading for round diamonds, so it collected massive amounts of diamonds in different colours, cuts and clarities to create a representation of the round diamond population in the world. “One of our biggest advantages is having 100 million diamonds that go through our system annually and that equals a great deal of data. There is potential to create new products and services using those data,” remarked Block. Coloured gems AI technology is also making waves in the coloured gemstone sector. In 2023, Gübelin Gem Lab unveiled GemtelligenceTM, an AI-driven software that determines the origin of rubies, sapphires and emeralds, as well as heat treatments in rubies and sapphires. Daniel Nyfeler, managing director of Gübelin Gem Lab, said the development of Gemtelligence was driven by a desire to transform gemmological expertise into readily accessible technology that can provide accurate analytical data absent inherent inconsistencies that human experts could make. Gemtelligence’s algorithm, based on the latest deep learning architecture, recognises significant patterns among various data modalities, which, compared to gemmologists, provides heightened precision and robustness. “This presents another advantage. We can reduce the amount of data that we feed into the software, and it will still perform better than human experts,” added Nyfeler. “It can also skip this one task, which is the most expensive and most labour-intensive process: Microscopy.” The software is also capable of developing new features and recognising new patterns, which allow gem labs to shorten the pipeline for some gemstones and come up with a more effective way to test gems, he noted. Design customisation Jewellery design is also witnessing a renaissance of sorts, with the advent of generative AI technology. Payal Shah, founder of jewellery brand L’Dezen, introduced an AI jewellery design platform, L’Dezen By You.AI, in partnership with Algoneering, an emerging AI technology solutions company focusing on the jewellery sector. “This collaboration empowers anyone to create personalised jewellery, transforming creative ideas into custom designs in real-time,” explained Shah. “This platform helps you bring your vision to life with no advanced skills required.” Users input prompts such as product category and style, to name a few, into the system to generate a customised jewellery design. The application takes about two minutes to come up with designs based on the prompts. “You can explore different gemstones, materials and metals, among others, to spark new ideas for your custom design,” explained Shah. “The AI helps translate your ideas into reality. The magic behind the platform is its simplicity. You input your ideas, and the AI generates multiple designs – at least four in real time.”
RkJQdWJsaXNoZXIy NjAxOTU=