Records kept for decades often become impossible to open once the software that created them is retired. Preservica is now pitching its digital preservation platform at enterprises in regulated industries facing that problem.
The Boston-based company announced that it is expanding into enterprise data archiving. Its latest feature, AI-powered natural language search, is in preview with early access customers ahead of general release in early 2027.
Preservica's Active Digital Preservation keeps records in current formats while retaining the originals. It checks their integrity and logs every preservation action, so records remain usable without the original systems.
Natural language search lets users query the archive in their own words. Results respect existing permissions, and every result is an actual record rather than AI-generated text.
The company says its AI runs inside the platform and records are never used to train models. Earlier this year it added AI capabilities for PII detection, OCR, audio-visual transcription and metadata quality control.
Preservica cites a global construction and engineering firm that consolidated long-term records from 11 business units across SharePoint, cloud storage and file shares. A single search now surfaces records across the estate.
It also cites a 2026 BARC study of 225 data, analytics and AI leaders. In that study, 70% said less than half of their organisation's unstructured data is discoverable and usable for analytics or AI.
"A bridge, a power plant or a rail line can stay in service for generations, but the software behind its drawings and inspection footage will be replaced many times over," said Mike Quinn, CEO of Preservica.
Target sectors include construction, energy, manufacturing, transportation and financial services.