A new gemological laboratory is entering the market—but this time, the headline isn’t just who is behind it.
It’s how they’re choosing to present it.
Swiss International Gemlab (SIG), founded by gemologists and co-founders Willy Bieri, CSO, Lawrence Hahn, COO, and Matthias Alessandri, CTO—each bringing between 14 and 21 years of hands-on colored stone laboratory experience, including senior leadership roles—officially launched April 1, 2026, with operations in Lucerne and Hong Kong. The group plans its first public appearance at GemGenève this May, positioning itself as an independent, global lab focused on colored stone reports.
What sets SIG apart—at least at the outset—is its decision to highlight AI-assisted analysis as part of its model.
AI in Gemology: Already Here, Just Not Always Visible
Artificial intelligence is not new to gemological laboratories.
In the diamond sector, GCAL, working alongside Sarine Technologies, has incorporated AI-driven systems to support color grading and clarity grading. Those efforts are built on large, standardized datasets—an environment where consistency and repeatability are easier to define.
In colored stones, the situation is more complex—but advanced data analysis is already part of the landscape.
Leading professional gem laboratories including GIA, SSEF – Swiss Gemmological Laboratory, Gübelin Gem Lab, AGL – American Gemological Laboratories, and GGTL Laboratories have spent years building extensive reference collections and databases, using computational tools to analyze chemical and spectroscopic data in support of identification, both country and color origin determination, and treatment analysis.
Gemological determinations may involve machine learning, pattern recognition, and large-scale data modeling—but they have largely remained behind the scenes—just not something they call AI, or use in self-promotion.
They are tools used to support the gemologist, not features promoted to the trade.
AI Moves Into View
That is where SIG takes a different approach.
Rather than treating advanced data analysis as behind-the-scenes lab work, SIG is presenting AI as a visible part of its analytical process—what it describes as “AI-assisted” evaluation.
At this stage, however, few technical details have been made public. There is no clear information yet on the size or structure of the datasets behind the system, how the AI has been trained, or how its results are validated against established gemological methods.
Those aren’t criticisms. They’re simply the next questions.
In comments provided to The Roskin Gem News Report, the founders emphasized that AI at SIG is not intended to replace the gemologist, but to support the process. Internally, they describe it as a “co-pilot”—a system designed to assist with consistency, efficiency, and quality control, while the final determination on every report remains with an experienced senior gemologist.
A Familiar Team, Entering a New Conversation
All three founders previously worked at GRS (GemResearch Swisslab), where they were involved in colored stone gem identification, origin determination, treatment analysis, fieldwork, and senior final grading.
GRS has developed a strong following in parts of the trade, particularly around its use of commercial color terminology such as “Pigeon Blood” for ruby and “Royal Blue” for sapphire—terms that resonate strongly in the marketplace.
At the same time, those terms—and how they are applied—have been the subject of ongoing discussion within the gemological community, particularly around consistency, definition, and the balance between commercial language and gemological precision.
That background may shape how the trade evaluates SIG’s approach—especially as it introduces AI into one of the most nuanced areas of gemology: color description and origin determination.
The Foundation Behind the Data
While SIG is positioning AI as part of its model, the use of advanced data analysis in gemology has been advancing for years.
At major laboratories, that work has long been built on deep research and carefully constructed reference datasets—systems that may involve complex computational analysis but are rarely presented as “AI” to the trade.
At GIA’s Tucson presentations earlier this year, Wim Vertriest put it plainly.
“Research is the foundation for all of these things,” he said, referring to origin determination work. “We don’t just go out and guess these things.”
He described three essential pillars behind that work: experienced staff, proper instrumentation—and reliable samples.
That last point is critical.
“If you collect a bag of gems, all you have is a bag of gems.”
In other words, data is only as good as the material behind it. Instruments can generate information, but interpretation—and ultimately trust—depends on expertise, calibration, and well-documented reference material.
SIG notes that its own approach is built on these same foundations: experienced gemologists, advanced instrumentation, and a growing reference collection—placing AI alongside these elements, not above them.
(Vertriest’s comments are drawn from our earlier Tucson feature. You can read more HERE.)
A Familiar Idea, Reframed
In many ways, SIG’s model echoes earlier developments in diamond analysis, where AI was introduced not to replace the gemologist, but to support consistency and repeatability.
That point has already been demonstrated. At GCAL, working with Sarine Technologies, AI systems are trained on large datasets of known color grades.
As GCAL president Angelo Palmieri explained, when tens of thousands of diamonds of known grades are fed into the system, it begins to recognize those grades through machine vision and image processing. Over time, “it started to learn GIA grading.”
But even with that level of data and repeatability, the role of the gemologist remains central.
“Our graders play a pivotal role in this process,” Palmieri notes. A stone is only finalized when “the machines and graders agree.”
In that sense, the technology itself isn’t new—the difference is how it’s being presented.
What the Trade Will Be Watching
For wholesalers, retailers, and appraisers, the key issue is not whether AI belongs in gemology.
That move into the future is already underway.
The real question is how it is presented—and how it performs in practice.
- How consistent are the results?
- How transparent is the methodology?
- How well do the findings align with established expectations in the trade?
In colored stones, where value often rests on subtle distinctions of hue, color origin, country of origin, and treatment, those answers carry significant weight.
Wait and See?
AI is already part of the modern gemological toolkit.
What is new is how it is being positioned.
With SIG, AI moves from a quiet and increasingly necessary analytical tool used behind the scenes to a visible part of the lab’s identity.
Whether that shift gains traction will depend not on the presence of AI itself, but on the degree of transparency, consistency, and trust that develops around it.
Tap here for more information about SIG, the Swiss International Gemlab.
