Artificial Intelligence is shaping the future of gemology, which is something I’ve been interested in studying for several years now. As someone who has a passion for gems and minerals, jewelry making, and learning, gemology is a natural evolution as I continue on my own learning journey.
Although I haven’t enrolled in any university yet, I keep up with news and trends. I aspire to attend the Gemological Institute of America (GIA) and become a graduate gemologist someday. It’s been a dream of mine for a long time to learn gemology. I think learning gemology will be fun, exciting, and mentally stimulating for me.
It is my interest in this field that has kept me up to date on new advances, trends, tools, and of course, gems and minerals themselves. And seeing AI being used in gemology is something I find intriguing, especially since I’ve written about AI for both CleanTechnica and Teslarati and even attended Tesla’s first AI Day event in 2021.
What Is AI?
AI has been shaping our future for quite some time. We use it on our phones and gadgets. An AI model developed by Cedars-Sinai was found to be 87% accurate in predicting how a woman would deliver within four hours of being admitted to the hospital.
AI is used in autonomous vehicles. It’s used in data security, astronomy, transportation, and travel. Have you ever used a GPS to navigate around a city? It’s kept me from getting lost while exploring new areas more often than once.
AI is used to automate tasks and save time while increasing output or results. It’s even used in writing, although I am not using ChatGPT to write this. I’ve yet to experiment with ChatGPT.
What Is Gemology?
Gemology is, in a nutshell, the study of natural and artificial gemstone materials. It’s considered a geoscience and a branch of mineralogy. I often get asked by friends and family what a certain type of stone is, and I can usually identify them with a glance if the mineral is something I’ve handled or studied.
However, cut and faceted gemstones are a bit harder to identify. Some gemstones are easy to identify with a glance, but many require special tools, especially if all you have is a gemstone and no other factual information. Anyone can bring a cut and faceted gem and say that they were told it was a diamond, but without some type of proof like a gem report, you would need to study the gem with the right tools to determine its identity.
GIA is probably the most recognizable gemology institute in the world and is considered the golden standard by many in the industry. It was founded in 1931 with the goal of protecting buyers and sellers of gemstones with strict standards used to evaluate gemstone quality. In 1953, GIA developed its International Diamond Grading System and the 4Cs:
- Carat Weight.
How AI Is Being Applied To Gemology
In October 2022, GIA announced that it developed an AI-based solution for grading the clarity of a diamond. GIA and IBM collaborated on this tool, and AI-powered software is able to find small imperfections within a stone. These imperfections are known as inclusions, and they are not necessarily a bad thing, but they are key to determining whether or not a stone is lab-made or naturally formed.
IBM developed the system using GIA’s standards and decades of data points to teach and train the algorithms to locate and identify the inclusions. After locating and evaluating the inclusions, the algorithm then grades the gemstone for clarity.
For those worrying that AI would put a gemologist out of a job, well consider the fact that there are millions, probably billions of gemstones in the world. GIA’s own executive vice president, Tom Moses, pointed this out, noting that the institute analyzes millions of diamonds.
“The Institute is uniquely positioned to leverage AI to drive our consumer protection mission forward,” he said.
GIA’s Chief Operating Officer, Pritesh Patel, expects AI to grade 70-80% of the millions of diamonds annually that GIA grades. However, the most difficult grading cases will still be handled by GIA-trained gemologists. In this case, AI adds efficiency, reliability, and consistency to the gem grading industry.
This is just one way AI is being applied to gemology. In 2020, the Gübelin Gem Lab announced a partnership with CSEM, a Swiss research and development center, to automate the process of determining a gemstone’s authenticity and origin.
In 2023, Gübelin Gem Lab celebrated its 100th anniversary with the launch of Gemtelligence, a result of that above-mentioned partnership. Gemtelligence is not only able to consistently analyze colored gemstones (a gemological definition of gemstones that are not diamonds), but it determines the origins of the precious colored gemstones (sapphires, rubies, and emeralds) using AI and deep learning.
This is a huge milestone and the first time an AI has been able to do this. Professional Jeweller gives a deep dive into how this deep learning architecture is applied to gemology.
AI isn’t going away, and it will continue to shape our world and our lives. This can be a good thing but also a bad thing—a double edge sword. However, if used correctly, it can enhance our lives. I do agree with Elon Musk’s take:
“I think we need to regulate AI safety, frankly. Think of any technology which is potentially a risk to people,” — @elonmuskTweet
“I think we need to regulate AI safety, frankly. Think of any technology which is potentially a risk to people, like if it’s aircraft or cars or medicine, we have regulatory bodies that oversee the public safety of cars and planes and medicine. I think we should have a similar set of regulatory oversight for artificial intelligence, because I think it is actually a bigger risk to society.”
As for gemology, I think AI could enhance this field and protect buyers and sellers as GIA was founded to do. However, AI should be seen as a tool—a gun, knife, pen—a tool that can be used for good or ill, depending on one’s intentions. However, unlike an ink pen, AI could do more damage if used with ill intentions.