In an interview with The New Jeweller magazine, David Block, CEO of Sarine Technologies, said, “When we look at grading reports, we see that the 4Cs does not provide sufficient information for customers today. There can be a 20-40% spread in the reports that similar diamonds receive.”
The issue of inconsistency in diamond grading has long been a scourge in the diamond industry. It is well known that if you submit the same diamond to a number of gem labs, you are likely to receive different 4Cs gradings from each, sometimes significantly different.
The Problem with 4Cs Grading
The international, industry-wide grading system for diamonds - known as the “4Cs” - was first developed in the 1950s. For the past 70 years, it has been the common language in the diamond trade, in wholesale, retail and among lay consumers.
The 4Cs - Cut, Color, Clarity and Carat Weight - each provide a grading ‘score’ that together helps buyers and sellers understand the rarity and value of the diamond. For example, a diamond with a large carat weight and excellent clarity and color will naturally be more valuable than a smaller diamond with a lower clarity or a poorer cut. When comparing two very different diamonds, the 4Cs may be enough to reflect each diamond’s relative position on the 4Cs scale. However, the diamond trade is far more complex than that.
In the real world, over 140 million carats worth of rough diamonds are mined each year. Approximately half of these are gem quality and will go on to become polished diamonds. When 4Cs grades are assigned to millions of diamonds of all sizes and appearances, at numerous grading facilities all over the world, by thousands of gemologists every day, it becomes clear that maintaining accuracy and standardization across the board is going to be impossible.
A Human Pursuit: How Diamonds Traditionally Receive 4Cs Grading
4Cs diamond grading is a very human pursuit. Grading of diamonds has traditionally been carried out by experienced gemologists who manually inspect the inventory, using visual aids such as the magnifying loupe. With the help of standard 4C grading charts, they decide the diamond’s grades in each category.
A skilled, professional gemologist certainly can provide a solid and accurate analysis for 4Cs grading. However, no human being can deliver standardized consistency in any endeavor requiring visual judgement 100% of the time. People, no matter how experienced, are subject to natural tendencies, such as fatigue, and no person is capable of perfectly objective perception, visual or otherwise. There are so many factors at play - physical, cognitive, physiological and emotional. After all, people are not machines.
In an opinion piece in IDEX Online earlier this year, Albert Robinson discussed the advent of artificial intelligence in the diamond industry, in response to Sarine’s launch of its AI-based diamond grading lab, the Sarine Lab. According to Robinson, “[But] the advent of AI is likely to change the reality of the diamond industry just as it is impacting other businesses. If an autonomous car can be developed to pick you up, deliver you to your destination and then continue to the next customer, then a machine that can inspect a diamond based on certain parameters does not seem like too much of a stretch.”
AI Diamond Grading: Beyond Human Capability
Machines and computers have taken over many tasks traditionally done by humans, enabling results that are more accurate, time efficient and cost efficient. This is happening to some extent in the diamond grading industry too. As Sarine CEO David Block continues in The New Jeweller profile, “There has been a big breakthrough in grading using AI. I believe this will continue to develop.”
In fact, Sarine has been leading the movement towards AI-based and automated diamond grading since the launch of the world’s first technology of its kind, the Sarine Clarity and Sarine Color, in 2016. These technologies are based on machine learning - the more diamonds scanned and graded by the device, the more knowledge it accumulates, the more sophisticated its algorithm-based methodology becomes. Based on the accumulation of diamond data, the grading outputs become more and more accurate and consistent, enabling standardized, ever improving grading results.
That’s not to say that AI grading technologies are set to replace gemologists - rather, AI is a powerful tool in the hands of a skilled gemologist, finally enabling a provably consistent, objective approach to diamond grading.
Don’t miss the next installment of this special blog series about AI and technology in diamond grading, Part 2: Clarity Grading. Scroll to the top of this page to subscribe to The Diamond Blog and get all the latest diamond news delivered straight to your inbox.