5 mins read
Dateline: Incode Technologies, product launch, 14 October 2026
Incode Technologies, the San Francisco identity verification company, launched TrustGraph on 14 October 2026, a peer-to-peer network that lets banks, marketplaces and AI companies match fraud and trust signals across institutions without pooling customer records, and says early users detected up to 53% more fraud than with their existing models. The release, Fintechbits notes, does not say how many institutions sit on TrustGraph at launch, how many took part in the early access program behind its up to 53% figure, or what false-positive rate came with the extra detections. TrustGraph builds on technology from Identiq, which Incode acquired in June 2026, according to the launch release.
Deal: Built on technology from Identiq, acquired by Incode in June 2026Market: Financial institutions, marketplaces and AI companies
How TrustGraph works
TrustGraph uses secure multi-party computation to compare one-time cryptographic material between participating institutions. When the same actor shows up at two banks, the second bank gets a signal without either bank seeing the other’s customer records. There is no central data lake, and each institution applies its own rules to the signal it receives. Incode says the network is meant to supply intelligence, and each member decides what to block. Technical details are on Incode’s site.
Signals are anchored to verified biometric identities, and that anchor is what lets the network recognise a repeat fraudster who turns up with a new name, document or device. Inside a single institution, the same matching links applications that share a device, document details or near-identical addresses so investigators can treat a fraud ring as one case.
It also works in the other direction. A customer who has already verified successfully elsewhere on the network can carry a positive trust signal, which Incode says helps institutions approve more legitimate applicants.
Fintechbits analysis
Why Incode says the timing is right
Incode’s case is that fraud has become networked faster than defences. It projects that agentic fraud, where AI agents build synthetic identities and retry after every rejection, will make up 60% of all fraud attempts by the end of 2026. That is Incode’s own projection, not an independent estimate.
Ricardo Amper, Incode’s founder and chief executive, says: “Fraud is networked. Defenses are still siloed.” Navjeet Grewal, managing director of verification and digital credentials at Interac, Canada’s payments and identity network, is quoted in the release backing the network approach.
Fintechbits has followed the same arms race at other vendors, including Binance.US and Socure and Bron, FaceScan and iProov.
Fintechbits analysis
Why TrustGraph’s value depends on who joins
A shared fraud network is worth roughly what its members bring to it. The idea behind TrustGraph is sound, and the cryptography answers the usual objection that banks cannot share customer data with competitors. But the network effect does not start until enough institutions are on it, and the release does not name a single participant.
Incode has a strong base to recruit from. It says its customers include eight of the top 10 US banks and the top three global neobanks, and the Identiq technology was built for exactly this use. If even a few of those banks switch TrustGraph on, it becomes hard for a rival consortium to catch up.
The 53% figure needs context before anyone relies on it. “Up to” marks the best result, not the typical one. Without a participant count or a false-positive rate, nobody can tell whether TrustGraph caught more fraud or simply flagged more customers. Banks buying it will ask for both, and they should.
Frequently asked questions
What is Incode TrustGraph?
TrustGraph is a privacy-preserving network from Incode that lets institutions match fraud and trust signals across participants without pooling or exchanging customer records. It uses secure multi-party computation and is anchored to verified biometric identities.
How does TrustGraph protect customer data?
Customer records stay with the institution that holds them. TrustGraph compares one-time cryptographic material between peers, with no central database, and each institution decides how to act on the signals it receives.
What does the TrustGraph launch release leave out?
Fintechbits notes that the release does not say how many institutions are on TrustGraph, how many were in the early access program behind the up to 53% figure, or what false-positive rate came with it. Those three numbers decide whether the network is useful at scale.
What to Watch Next
The milestone that matters is the first named bank on TrustGraph, particularly one of the eight top-10 US banks Incode counts as customers. A published participant count, or an independent false-positive figure, would turn the 53% claim into something buyers can test.
Why this matters
The release, Fintechbits notes, does not say how many institutions are on TrustGraph, how many were in the early access program behind its up to 53% detection claim, or what false-positive rate came with it. The design answers banks’ old objection to sharing fraud data with rivals, because no customer records leave the institution. Incode’s customer list, including eight of the top 10 US banks, gives it a head start. The network only works once those banks join.
Primary sources
businesswire.com | PRESS RELEASE | https://www.businesswire.com/news/home/20261014354134/en/Incode-launches-TrustGraph-the-first-trust-network-built-for-the-agentic-era
incode.com | COMPANY WEBSITE | https://incode.com/platform/trustgraph
