Innerworks: fighting deepfake fraud with synthetic threat intelligence

AI‑generated voice and video fraud have exploded, costing businesses billions.

Innerworks has built what it calls “Synthetic Threat Intelligence” to stay ahead of deepfakes and automated scams. The London start‑up raised €3.7 million in seed funding led by AlbionVC, with participation from investors including Digital Currency Group, Firestreak Ventures and Founders Capital. Innerworks’ platform simulates AI‑driven fraud scenarios, using generative models to probe a client’s defences, and then deploys detection algorithms that achieve 97 % accuracy

Co‑founders – veterans of fraud‑detection and cyber‑forensics – argue that existing tools, built for conventional threats, can’t keep up with deepfake bots. By generating artificial threat models, Innerworks allows banks and digital‑first businesses to spot attacks before they occur.

The company claims its approach can reduce fraud losses by up to 80 %. The seed funding will be used to hire machine‑learning researchers, expand sales to financial‑services clients and develop zero‑knowledge‑proof techniques that ensure customer data privacy. Partnerships with European banks will help refine models on real‑world data.

AlbionVC says synthetic threat intelligence could become a cornerstone of cybersecurity strategies by 2026.

Analysts note that regulators are pressuring banks to address deepfake risks after a spate of voice‑cloning scams. Innerworks plans to broaden its models to cover AI‑generated phishing and social‑engineering scripts.

The company also aims to build an industry‑wide knowledge base of emerging attacks to share insights across its customer network.

Photo: Innerworks founder in London office (WebProNews).