Trust in inspection results is critical when decisions affect safety, availability and long-term asset health. That’s why we regularly ask our clients and technical reviewers to assess their confidence in the outputs of our AI-based damage detection and classification system, as presented through our web portal.

Across the most recent confidence surveys, including two involving experienced blade specialists reviewing data at specialist-portal level, the feedback shows consistent strengths:

• High confidence in report clarity and completeness, with scores clustering in the upper range, supporting efficient technical review and decision-making.
  • Strong confidence in damage detection and classification for the most common blade damage types, particularly erosion and open bond lines, which are critical drivers for maintenance planning.
  • Positive confidence in repair prioritisation recommendations, especially among Blade Experts working with detailed damage data.
  • Consistent confidence in repair decision support, indicating that our outputs are considered reliable inputs rather than black-box results.

Importantly, surveys of Blade Experts reflect feedback from highly experienced structural and materials specialists, using the Specialist part of the portal. While surveys on the End User Portal reflect the feedback from Repair and Asset Owner companies with experience in Repair and Cost management. Their combined input confirms that AI-supported inspections can be trusted not only at Blade Expert level, but also at Operator level.

The surveys also provide clear guidance on where to continue improving: notably expanded damage classification options, reinforcing the value of continuous feedback in strengthening confidence further.

We see this ongoing dialogue with users as essential: confidence is not claimed, it is earned, inspection by inspection, dataset by dataset.