In this episode of Good Question, we explore a fast-emerging challenge in fraud detection: AI-generated documents. Recent updates to ChatGPT’s image generation capabilities have introduced powerful new tools for fraudsters. Using an autoregressive image model (as opposed to diffusion-based approaches), these systems can now generate entire documents from scratch (pixel by pixel) with highly realistic formatting and texture.

Join Inscribe CEO & Co-founder Ronan Burke, Senior Research Scientist Martina Pugliese, and host Brianna Valleskey as they dive into how new generative image models are being used to create realistic-looking receipts, invoices, and even bank statements — and what that means for risk teams. We cover:

- How autoregressive image models work (and why they’re more convincing than ever)

- Which types of documents are easiest to fake (and which still pose challenges for AI)

- What forensic and semantic signals Inscribe uses to detect synthetic documents

- How Inscribe’s AI agents are evolving to stay ahead of this threat

- Why this isn’t hype — but also not a panic moment

From synthetic receipts to near-realistic invoices, this episode explores what’s real, what’s hype, and what’s coming next in the arms race between generative AI and fraud detection.

#GenerativeAI #DocumentFraud #fintechpodcast #Fintech #FraudDetection #LLMs #genai #RiskManagement