Geological Logging with AI in Hours

Key Takeaways

First-pass (“quick”) geological logging, the initial lithology and alteration read on fresh core, is one of the more mature AI use cases in exploration.
Datarock Core and Seequent Central’s logging module both support AI-accelerated first-pass logging directly from core photography.
Datarock has processed over 30 million metres of core since 2020, turning what would be months of manual logging into hours.
This is a “Yes”: already deployed, production-grade AI, not a research prototype.

TL;DR

For fast geological logging with AI, photograph core as it arrives and run the images through a tool like Datarock Core. You get a structured first-pass log of rock types, fractures and structures within hours. Geologists then review and refine that draft instead of logging from a blank sheet.

What Is Quick Geological Logging?

Geological logging records what comes out of a drill hole: lithology, alteration, mineralization, structures and fractures, interval by interval. Quick logging is the first pass. It happens as soon as core reaches the shed, long before detailed petrography or assay results are back.

That speed matters because geologists need an immediate read to make real-time decisions. They decide whether to extend a hole, change a target or reprioritize the next rig move. Historically that meant a geologist walking the core shed with a hand lens. Today, it increasingly starts with a camera.

How Do I Automate Geological Logging With AI?

Automated first-pass logging follows three steps: capture, extract and share.

Capture consistent core imagery

The AI can only log what the camera sees. Therefore, standard lighting, scale bars, tray labels and wet or dry conditions matter as much as the model. A reliable core photography setup is the first investment. It also leaves a permanent visual record that anyone can re-check later.

Extract a first-pass geological logging draft

Datarock Core extracts geotechnical and geological information directly from standard RGB core imagery. It picks out rock type boundaries, fracture frequency, discing and structural features. The result is a structured, auditable log delivered within hours of the core arriving at the shed.

It plugs into existing core-shed workflows rather than replacing them. Geologists still validate and interpret, but they start from an AI-generated first draft instead of a blank log sheet.

Seequent Central’s logging module offers a complementary angle. It keeps quick-logged data centralized, version-controlled and immediately visible to the rest of the geoscience team in 3D. As a result, a quick log done at the rig doesn’t sit disconnected from the block model until someone re-enters it by hand.

How Reliable Is AI Geological Logging?

The evidence base here is real, not speculative. Datarock’s fracture-detection AI has been validated against expert manual logging in published case studies, including at the Carrapateena copper-gold mine. The technology is in production use at major mining companies, not just pilot programs.

Still, a first pass is a first pass. Photos show the core surface, so subtle alteration, fine-grained mineralogy and textures seen only under a microscope need other tools. Detailed follow-up comes from automated petrography and lab methods. Once quick logs exist, informed core logging keeps the interpretation consistent with the evolving deposit model across the whole program.

FAQs

How fast is AI geological logging compared with manual logging?

AI geological logging delivers a structured first-pass log within hours of core reaching the shed. Manual logging of the same core can take days or weeks. Datarock reports processing over 30 million metres of core since 2020.

What does AI geological logging capture from core photos?

It captures rock type boundaries, fracture frequency, discing and structural features from standard RGB imagery. Those are features visible on the core surface. Microscopic textures and detailed mineralogy still need other methods.

Do geologists still need to log core?

Yes. The AI log is a first draft that geologists validate, correct and interpret. The change is where they start: from a structured draft rather than a blank sheet.

What is the difference between quick and informed logging?

Quick logging focuses on getting a usable first read fast, often while the rig is still drilling. Informed logging focuses on interpreting each new hole against the current model and earlier holes. Most programs benefit from both.

Try It With Geocluster

Getting from “AI logged the core” to “here’s what it means for the deposit model” still takes synthesis. You have to weigh logging output against prior geological knowledge, analog deposits and exploration targets. That is the research work Geocluster is designed to accelerate, so check it out if you’re building this into your exploration pipeline.