SAP Concur: AI-Generated Receipts Challenge Human Audits, Verify Detection 18x Higher

Key points

• AI-generated receipts are visually realistic, mimicking logos, fonts, itemized entries, VAT lines and totals, and can have metadata such as EXIF fields, timestamps and geolocation stripped or faked, removing conventional red flags.
• SAP Concur's AI-powered auditing solution Verify has helped customers identify thousands of AI-generated receipts, with a detection rate roughly 18 times higher than earlier checks focused only on known online receipt generators.
• Chris Juneau, senior vice president and head of Product Marketing at SAP Concur, said about 1% of reviewed receipts have been flagged as potentially AI-generated, created by image generators including ChatGPT, Gemini and Stable Diffusion.
• Juneau said AI is not increasing the frequency of expense fraud but is changing how it occurs; with nearly 70% of expense transactions including an attached receipt, manual review is insufficient.
• Effective defenses pair metadata forensics with machine-learning detectors trained on large bodies of confirmed real and AI-generated receipts, kept current through continuous retraining and partner-sourced intelligence.
• Operational fallout includes overwhelmed audit teams, heavier accounts-payable workloads, undermined corporate travel program integrity, and erosion of employee trust or legal exposure from mishandled investigations.
• Finance, travel and compliance teams are adapting by tightening submission policies, making e-receipts mandatory and integrating automated tools into policy enforcement.
• A risk-based approach is recommended: automated scoring flags only higher-risk submissions for human review, supported by clear policies, transparent communication, documented decisions and easy appeals or clarifications.