AI image fraud will cost $40 billion next year - can these international standards help?

2026-07-23 · Source: News and Advice on the World's Latest Innovations | ZDNET · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy · Depth: Novice, short

Summary

International standards bodies, including the IEC, ISO, and ITU, are advancing efforts to combat the growing credibility crisis in digital imagery caused by AI-generated deepfakes. This issue is projected to escalate US fraud losses from \$12.3 billion in 2023 to \$40 billion by 2027, according to Deloitte estimates. New standards, announced at the UN's AI for Good conference, aim to provide tools for end users and companies to distinguish authentic content. Key among these is JPEG Trust, which includes Part 2 for trust profile snippets and reporting templates, and Part 3 for media asset watermarking. These standards are designed to empower users with verification means rather than relying on creators to label fraudulent content. Additional initiatives include an Originator profile framework, a Vocabulary for expressing content preferences for AI, and H.MMAUTH for multimedia content authentication via digital signing.

Key takeaway

For policy makers and AI security engineers addressing digital content authenticity, recognize that relying on creators to label AI-generated fraud is ineffective. Instead, prioritize the adoption and integration of emerging international standards like JPEG Trust, Originator profiles, and H.MMAUTH. These frameworks empower your organization and end-users with the necessary tools to independently verify content authenticity, thereby mitigating the projected \$40 billion fraud losses by 2027 and restoring public trust in digital media.

Key insights

International standards are emerging to provide user-side verification tools for distinguishing real from AI-generated content amidst a growing credibility crisis.

Principles

• Fraudsters won't label AI content.
• Trust in content is context-dependent.
• Metadata provides crucial content context.

In practice

• Utilize JPEG Trust Part 2 for workflow profiles.
• Implement JPEG Trust Part 3 for watermarking.
• Employ Originator profiles for content provenance.

Topics

• AI Image Fraud
• Deepfakes
• Content Authenticity
• JPEG Trust
• International Standards
• Digital Watermarking

Best for: CTO, Executive, AI Product Manager, AI Security Engineer, Policy Maker, Legal Professional

Related on AIssential

• A viral Reddit post alleging fraud from a food delivery app turned out to be AI-generated — AI tools enable sophisticated online hoaxes, challenging traditional fact-checking and content verification methods.
• Microsoft's research argues AI media authentication doesn't work reliably, yet new laws assume it does — Current AI media authentication methods are unreliable, requiring combined approaches and facing legislative overreach.
• AI fraud is exploding. Here’s the countermove. — AI-driven document fraud necessitates AI-enhanced forensic analysis for effective detection and prevention.
• The Real Iranian Women Protesters Trump Made Look Synthetic — AI-enhanced and generated content erodes trust in human rights documentation, obscuring real cases amidst manufactured narratives.
• We Committed Fraud with OpenAI's New Image Model (and Called Mum) - EP99.38 — AI image generation has reached a level of realism enabling highly convincing forgeries, posing significant societal and economic challenges.
• OpenAI is making it easier to check if an image was made by their models — OpenAI uses C2PA metadata and Google's SynthID watermarking to signal AI-generated images.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by News and Advice on the World's Latest Innovations | ZDNET.