JunkDNA.AI was founded inside BMSachs Group — a multi-vertical incubator focused on long-tail personal-data infrastructure. The thesis: identity should propagate, not be re-entered. The founding observation: no single regulator, vendor, or platform was incentivized to fix the cross-institution synchronization problem, because every player only sees their slice. That’s the gap we built around.
We took the company name from genomics. In the early decades of human-genome research, roughly 98% of human DNA was dismissed as “junk” — non-coding sequences with no apparent function. We now know that “junk” is what makes you, you. It carries regulatory information, evolutionary history, the subtle context that distinguishes one human from another. The 2% on a driver’s license is just the surface; the curated 98% is everything else.
Personal digital identity works the same way. The fragmented, scattered, “junk” details of your life — addresses, employer, providers, subscriptions, credentials — are where the actual value sits. And like genomic junk DNA, they only become useful when something can finally read across them. That something is what we’re building.
By April 14, 2026, the company filed its Non-Provisional Utility patent (Application No. 19/647,711) and the corresponding PCT International Application (PCT/US26/23713) covering the orchestration kernel. Behind those formal filings sits a structured continuation, continuation-in-part, and divisional strategy targeting eight major vertical extensions — property management, labor & trades, healthcare, escrow, messaging, data portability, compliance fabric, and verified marketplace governance. The roadmap from here is sector-by-sector deployment of that stack through 2029.