Free AI Image Detector: Was This Photo Made by AI?

Tested properly, people are far worse at spotting AI-generated images than they assume about themselves. Confidence does not track accuracy here. The visual tells everyone learned - hands, text, over-smooth skin - are exactly the failures each new model version targets first, so relying on them means relying on a list that goes out of date with every release.

Meanwhile the volume moved. AI images are now routine in marketplace listings, dating profiles, rental ads, insurance claims, review photos and news feeds. Most of them are harmless. The problem is that harmless and fraudulent look identical, and the only thing separating them is evidence you cannot see by looking.

Midjourney, across all versions, produces the most photographic output of the major generators. Its aesthetic tuning shows up as a characteristic smoothness: skin, fabric and foliage carry less random variation than a real camera would record. Hands improved dramatically from v5 to v6, so hand-counting is no longer the tell people still repeat.

DALL-E 3 renders text better than almost anything else, which removed the old giveaway of garbled signage. What it did not remove is its close prompt adherence, which tends to produce compositions that are slightly too centered and too obliging. Images made in ChatGPT since May 2026 also carry the invisible OpenAI watermark, which FauxLens reads.

Stable Diffusion is the awkward one, because it is open. SD 1.5, SDXL, SD3, LoRAs, community checkpoints and heavily fine-tuned derivatives all exist in the wild, and a custom checkpoint drifts from the base model's signature. What often survives is the metadata: front-ends such as AUTOMATIC1111 and ComfyUI write the generation settings into the file, and when that text is intact it is the kind of generator signature a FauxLens scan reads.

Flux.1 came from the original Stable Diffusion authors and uses a diffusion transformer rather than a U-Net. So classifiers trained purely on older diffusion output perform poorly on it. Flux renders readable in-image text, which defeats another traditional visual tell. Dev, Pro and Schnell differ measurably from each other: Schnell is distilled and takes fewer sampling steps, which leaves more compression-like artifacts than Dev.

Also covered: Google Imagen, Adobe Firefly, and the video generators on the video pages.

The honest caveat is that this list is a snapshot. New model versions ship faster than any detector retrains, so a brand-new generator always has a window where accuracy against its output is lower. FauxLens does not try to name the model from the pixels. It names a maker only when a watermark points to OpenAI or Google, or when the file's own credentials or metadata name the tool.

The confidence percentage shows how strongly the evidence supports the verdict. It is not a calibrated probability: a confidence of 92% does not mean the image is definitely AI-generated. It means the evidence strongly points in that direction. The evidence lines show what drove the call. A content credential or watermark is direct evidence, while a pixel score is a judgment. If the evidence says an AI tool altered only part of a real photo, it may have been upscaled or partly repainted rather than made from nothing. Use this to understand the nature of the AI involvement, not just its presence. If you receive a high-confidence result that contradicts other evidence you have, read the evidence lines closely and consider whether the image was significantly processed before you received it.

Romance scams are the most financially damaging use case. Scammers generate consistent AI personas using Midjourney or Flux (a military officer, a successful engineer, an overseas contractor) and sustain multi-month relationships that cost victims tens of thousands of dollars each. Romance scams are among the costliest frauds reported to the FTC every year. Because these AI personas have never been photographed, reverse image search cannot catch them. Forensic pixel analysis is the only reliable detection method.

In hiring fraud, North Korea-linked threat actors have been documented by the FBI using real-time deepfake video filters to fraudulently obtain employment at US technology companies, gaining access to internal systems and source code. Security researchers and hiring teams across multiple industries are now reporting a significant rise in suspected deepfake candidates in video interviews.

In insurance fraud, AI-generated damage photos (flooded basements, crumpled fenders, fire-damaged roofs) are submitted to claims departments. Industry analysts and insurers warn that AI-fabricated claim documentation is a growing source of fraud losses that manual review alone cannot address.

In journalism, AI-generated photographs have been published and later retracted, typically because they were authenticated by eye rather than by forensic tools.

Heavily re-compressed images (those downloaded from social media platforms like WhatsApp, Instagram, or Twitter) have been through multiple JPEG compression cycles. Each cycle strips metadata and throws away fine pixel detail the detection model relies on. Small images have the same problem, because there is simply less to judge. Heavy re-compression is the single biggest reason a scan comes back inconclusive rather than wrong, which is why the original file is worth asking for.

Adversarially processed images are the hardest case. Techniques like adversarial perturbation specifically designed to fool AI detectors exist in research contexts and are beginning to appear in sophisticated fraud operations. Our multi-signal approach is more resilient than single-signal detectors. But a determined adversary with access to our detection pipeline could potentially reduce our confidence.

Hybrid images, those that start as real photographs with AI inpainting applied to specific regions, show mixed signals. The detection model recognises this case and says so in the evidence line (an AI tool altered part of the image), and the call gets harder when the repainted area is small. The overall verdict may be "Likely AI-Generated" with moderate confidence rather than a high-confidence determination.

When confidence is low (below 70%) and the stakes are high, do not rely on FauxLens alone. Cross-reference with reverse image search. Request the original uncompressed file. Ask for additional context about the image source. For legal or high-value financial decisions, engage a certified digital forensic examiner who can apply chain-of-custody procedures and provide documented expert analysis.