As the countdown begins to the 2025 CIBJO Congress in Paris on October 27, the organisation’s Technology Committee has unveiled its tenth and final pre-congress Special Report — an insightful exploration of how artificial intelligence (AI) is reshaping the jewellery landscape.
Chaired by Stéphane Fischler, the report examines AI’s expanding footprint in the sector, offering a balanced perspective on its promises and pitfalls. It seeks to move beyond hype, encouraging stakeholders to view AI as a practical tool rather than a disruptive threat.
“AI itself is neither positive nor negative — its impact depends entirely on how it’s applied,” said Fischler. “The companies that integrate AI thoughtfully, ethically, and strategically will boost efficiency while keeping human judgment at the core of every decision.”
Acknowledging jewellery’s deep roots in heritage, artistry, and emotional value, the report urges the trade to cultivate a more experimental mindset. Fischler emphasizes the need for “safe-to-fail” environments, where innovative ideas can be tested and refined without fear of missteps. “From generative design and market forecasting to consumer engagement, AI thrives when used as a collaborator in creativity,” he noted.
However, the report also sounds a note of caution against excessive automation. Fischler reminds readers that luxury jewellery derives its soul from craftsmanship and storytelling — aspects that algorithms cannot replace. “AI should amplify human artistry, not overshadow it,” he said.
The report brings together perspectives from leading voices in the global jewellery community, including Elle Hill (Hill & Co.), Mahiar Borhanjoo (De Beers Group), Thomas Baillod (BA111OD), David Block (Sarine Technologies), Daniel Nyfeler (Gübelin), and Emmanuel Piat (Maison Piat) — each contributing insights into how technology can coexist with tradition to shape the future of fine jewellery
Disclaimer: This information has been collected through secondary research and TJM Media Pvt Ltd. is not responsible for any errors in the same.
