Vrify AI platform to revolutionise exploration

Since its launch six years ago, Vrify has revolutionised how companies visualise their projects. Now, with the launch of an artificial intelligence (AI) platform, it aims to do the same for mineral exploration, chief executive Steven de Jong told Mining Journal.

Vrify brought the corporate presentation, the main investor promotional tool of most companies, into the digital age by enabling virtual mine tours and 3D modelling of deposits, allowing company executives to give a personalised and active walk-through of their projects.

Through its AI exploration platform, Vrify aims to make exploration more efficient and effective, increase the hit rate of drilling, and make exploration more accessible to non-specialist investors. The benefit of using AI in other industries can be significant.

"AI negates the need for so much technical jargon as it gives a predictive model that investors can understand. In terms of an AI premium, the returns in the industrial sector in 2023 were 25%, but AI-enabled industrial companies delivered 125%. We will see an AI premium come into the mineral sector and it will become common to have a slide in each deck about how a company is leveraging AI to improve ROI."

"We are getting 10% use value out of existing geological data sets than what we could be getting. If investors can use AI to audit companies' exploration programmes, it will reduce the risk profile. It is a new way to look at existing projects and see if you missed anything," said de Jong.

De Jong uses the oil and gas sector as a benchmark for comparison, a sector that attracts large investments and where drilling has a much higher hit rate. Investment presentations for oil and gas projects typically contain a quantification of the potential and the risk expressed as an estimated target size, as well as the chance of success.

Bringing this concept to mineral exploration, this would be an independent third-party statement akin to saying, there is potential to find a 5Moz gold deposit, with a 15% chance of success.

The Vrify platform uses AI to interpret a company's data for a target area to produce a prospective model with a 3D block model heat map and a confidence level expressed as a prospectivity score. The platform is led by geoscientists who won the Integra gold rush challenge eight years ago, with a project that combines machine learning and geoscience.

"Institutional resource investors are asking whether AI can make more efficient use of resources, reduce the leakage from exploration budgets and increase the chance of discovery. It is hoped the use of AI will enable the normalisation of the information presented to give investors the ability to compare mineral investment opportunities better.

"If you take 10 random exploration companies co and go through their presentations, one has geophysics, another has another type of geophysics, and another has something else. We are big believers that the industry needs to go in a direction where all that information sits in a standardised technical appendix."

"If investors and companies learn to trust a brand like Vrify to analyse the raw data, investors will not need to do this themselves. We are going to position ourselves as a standard in this space," said de Jong.

Vrify has a strong brand platform to build upon, with 170 clients using its deposit visualisation platform, and four initially signed up for the AI platform, and four more about to be added.

In this context, Vrify is repeating what it did in the visualisation space, where companies essentially used the same paper presentation to communicate about their projects that generations of companies before them did. In the exploration space, companies are in the habit of deploying the same tools over the same areas that have previously been explored expecting to see something that a previous operator missed.

"The new platform will be a catalyst and deliver AI-driven targets to clients that they can talk about and drill. There is talk about a lack of discoveries in the sector, that you need to go deeper or further away. I have never believed in that, as so many discoveries are made near big mines. Look at Great Bear Resources [in the Red Lake district of Ontario, Canada]. Finding a new way to look at old data will discover all kinds of nuggets," said de Jong.

The proof of the pudding will be if it works. De Jong said that the initial indications are that it does. Its initial four clients are exploring entirely different geological environments, including orogenic gold, copper porphyry and lithium pegmatites, with Vrify seeing the machine-learning aspects of the platform improving the tool from one client to the next.

"We train the model on a known area and filter out different data sets to determine the leading indicators as to why it is a target. We are finding that the predictive model can identify faults and fold structures that the geologists did not know existed."

"Some of the predictive models we have created and back-tested by withholding drill data are highly accurate with not a lot of data. There is so much low-hanging fruit within existing data sets. It can completely turn around how exploration is done," said de Jong.