ChipsFebruary 18, 2026via Microsoft Research
Project Silica’s advances in glass storage technology
Why it matters
As AI training datasets and model weights grow exponentially, durable cold storage becomes critical infrastructure. Project Silica's glass-based encoding offers orders of magnitude better longevity and cost efficiency than current tape/disk solutions—a foundational layer for scaling AI compute economics.
Key signals
- Project Silica encodes data in borosilicate glass
- 10,000-year data preservation capability
- Lower media cost vs. existing solutions
- Simplified writing and reading systems
- Published in Nature journal
- Microsoft Research initiative
The hook
10,000-year storage. Microsoft's Project Silica just cracked the infrastructure problem that could reshape AI model archival.
Project Silica introduces new techniques for encoding data in borosilicate glass, as described in the journal Nature. These advances lower media cost and simplify writing and reading systems while supporting 10,000-year data preservation.