🥸 Fakes, Frauds, and Forgery
Scammers Level Up, Fight Fire with Fire, and AI Reads Your Data
"AI is poised to make high-fidelity forgery inexpensive and automated, leading to potentially disastrous consequences for democracy, security, and society."
—Oren Etzioni, CEO of the Allen Institute for Artificial Intelligence
The AI Breakdown
Forgery Gets a Facelift

Last week, we marveled at ChatGPT's new image generation capabilities. This week, we're shifting gears to explore how this shiny new tool might be a double-edged sword for dealerships.
Photo Fakery Made Easy
OpenAI’s latest image upgrade has made it astonishingly simple to create ultra-realistic images on demand. Driver’s licenses, insurance cards, accident photos—AI can spin them up in seconds. And yes, reports are already circulating of people using these tools to forge everything from receipts to insurance claims.
When everything can be faked, visual trust goes right out the window. Just like deepfakes made videos suspect and AI chatbots muddied email authenticity, this new image generation leap is turning even basic documents into a guessing game.
Dealerships in the Crosshairs
Day to day, dealerships rely on visual proof. IDs, insurance papers, service records—they’re routine parts of the transaction. But with AI generating these documents on demand, it’s getting harder to tell what’s real.
A fake driver’s license could sneak through and unlock financing. AI-made accident photos could inflate claims. Fraud, once a manual hustle, is now quickly scalable
Protective Measures for Dealerships
To safeguard against the emerging threats posed by AI-generated fraud, dealerships can consider the following strategies:
Implement Advanced Verification Systems: Adopt multi-factor authentication and biometric verification methods to ensure the authenticity of customer identities.​
Leverage Decentralized Identity Solutions: Explore options like Web3-based identity verification systems that provide tamper-proof credentials, reducing reliance on potentially falsifiable documents.​
Educate Your Team: Regular training sessions can help employees recognize signs of AI-generated documents and understand the latest fraud tactics.​
Stay Vigilant: Regularly monitor transactions and be on the lookout for red flags. Anomalies in purchasing behavior or document inconsistencies should prompt deeper investigation.
Prompt of the Week
Fight fire with fire by asking OpenAI 4o to help you detect digitally generated or altered images. This prompt flips the script by turning AI into your forensic sidekick.
Upload an image and let OpenAI run through a full inspection—checking for odd lighting, sketchy reflections, irregular color patterns, and more. You’ll get a step-by-step visual breakdown and a final verdict on whether the image is likely to be fake.
Keep this one handy for anything from customer-submitted docs to fishy online listings. AI fraud might be scaling up, but this prompt will help you stay one step ahead.
We are building a forensic photo analysis suite. When I upload a photo and say "check for digital alteration," you will run step-by-step Python code to detect signs of AI generation, compositing, colorization, and digital alterations. Perform and display each result individually to avoid memory errors. Step 1: run local contrast variation tests using Canny edge detection and multiple Gabor filters. Step 2: check reflections and highlights consistency using FFT noise pattern analysis and gradient magnitude lighting checks. Step 3: test for colorization by analyzing the B channel and checking saturation and hue irregularities. Step 4: perform patch-based anomaly detection using variance maps. Step 5: run histogram analysis. Always display each result visually before moving to the next step. At the end, provide a probability rating of low, medium, or high for whether the image is photoshopped, composited, digitally altered, or AI-generated, and explain your rating based on the successful analyses. Lastly, offer to export the final report as a PDF. If you understand, say "Upload a photo and say 'check for digital alteration.'"
Fresh Finds for Auto Pros
Identity Verification: Onfido
For auto dealers expanding into digital retailing or remote service payments, Onfido helps verify customer identities during online transactions, finance applications, and trade-in verifications. It’s a smart way to prevent fake IDs, synthetic identity fraud, and protect against scams when customers upload documents for financing or vehicle purchases.
Real-Time Fraud Detection: Featurespace
Featurespace specializes in live fraud monitoring using adaptive behavioral analytics. Perfect for dealers handling high volumes of online payments or in-house financing, it can flag suspicious activity early (like potential account takeovers or abnormal transaction patterns) keeping your sales and service departments a step ahead of fraudsters.
Bot Prevention: DataDome
Battles bots across websites, mobile apps, and APIs. Its capabilities are particularly beneficial for auto dealerships aiming to safeguard their digital platforms from malicious activities such as web scraping, credential stuffing, and payment fraud.​
Hear from the Experts
We caught up with Doug Betts, Global Head of Automotive at J.D. Power, during NADA 2025 to get his take on how AI is reshaping the data game in automotive.
Doug emphasized that AI is unlocking insights from data that was once too overwhelming to process. Now, thanks to AI’s ability to read, organize, and categorize vast information, dealers and OEMs can make smarter, faster decisions.
Check out the interview for more on how Betts sees AI turning automotive data from noise into knowledge.
Bits and Bytes
Amazon has upgraded its AI video model, Nova Reel, with the ability to generate videos up to two minutes long. 🎥
A recent study suggests that AI bots may be able to deliver mental health therapy with as much efficacy as—or possibly even more than—human clinicians. đź§
Objection! An AI avatar tried to argue a case before a New York court (but the judges quickly shut it down). 🧑‍⚖️
Reply