Proven where failure
was never an option.
A national oil major has run our AI on its offshore drilling rigs since 2018, today across ~half of a national fleet. In a controlled field trial, it cut tracked unsafe events by ~80%.
Offshore rigs monitored, ~half a national drilling fleet
Error rate across 18,000 events processed every day
Of raw footage filtered out. Only the 0.23% that matters surfaces
Reduction in tracked unsafe events over the field trial
The deployment in numbers.
Capabilities running on site today.
Controlled measurement. Verified reduction.
A controlled field trial — baseline collected before deployment, outcomes measured over the trial period — recorded the following reductions in tracked unsafe events.
| Event type | Before | After | Reduction |
|---|---|---|---|
| Distracted operator | 53 | 4 | ~92% |
| Human-machine segregation breach | 8 | 0 | Eliminated |
| Missing PPE / improper uniform | 23 | 8 | ~65% |
| Excessive speed | 22 | 9 | ~59% |
| Total tracked events | 106 | 21 | ~80% |
How it runs on a drilling fleet.
On each rig, an edge device processes every camera feed locally. No footage leaves the rig; detection is on-premise, low-latency, and independent of satellite bandwidth. The fleet management layer aggregates alerts, performance metrics, and safety data across all 13 rigs into a single dashboard for the central safety team.
See the same results on your cameras.
Book a demo and we’ll show you the detectors running on a live deployment. Then scope your site with a fully-costed Discovery phase.