> For the complete documentation index, see [llms.txt](https://tools.osintnewsletter.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://tools.osintnewsletter.com/osint-tools/image-whisperer.md).

# Image Whisperer

| **Image Whisperer** | **Quick Overview**                                                                                                             |
| ------------------- | ------------------------------------------------------------------------------------------------------------------------------ |
| URL                 | <https://imagewhisperer.org/>                                                                                                  |
| What it does        | Spots signs of AI manipulation and synthetic media by scanning images for artifacts, inconsistencies, and generation patterns. |
| How to use it       | Simply upload an image, let the tool analyse, then review flags and confidence indicators.                                     |
| Cost                | Free (for 2 searches). Paid for further searches.                                                                              |
| Account required    | No.                                                                                                                            |
| Cookies             | Mostly session management & analytics trackers.                                                                                |
| Ownership           | Created and developed by Henk van Ess, an OSINT specialist and journalist from the US.                                         |
| Use in Reporting    | Quick first-pass check for suspected AI-generated or manipulated images, ideal for triage before deeper forensic work.         |

### What does Image Whisperer do?

Image Whisperer helps journalists, researchers, and investigators quickly assess whether an image might be fake or manipulated. Instead of overwhelming technical jargon, it highlights visual “tells” like unnatural textures, lighting inconsistencies, or structural oddities often produced by generative models.

Note: It’s not to be treated as definitive truth but it’s a strong early warning system.

**The lowdown**: It’s a quick, no-friction AI image detector that’s ideal for early-stage verification.

### How to Use

**1. Simply upload or drag-and-drop an image and run the analysis.**

<img src="/files/2Ndqh9c0KUjyVtxuB5xx" alt="" height="223" width="602">

**2. Review results, checking flagged areas, confidence scores, and visual cues.**

<img src="/files/JFMFtBxc2VWyctrCevIY" alt="" height="545" width="602">

You can also download a full verification report as below:

<img src="/files/W9OvSM9KHUPIBlx93jiZ" alt="" height="301" width="602">

### Cost

* [ ] Free
* [x] Partially Free
* [ ] Paid

Free for 2 searches. Paid for further searches.&#x20;

## Data Processing

### Account Required:

* [ ] Yes
* [x] No

### Cookies:&#x20;

Cookies used are mostly session management and analytics trackers.&#x20;

One cookie (ph\_\*\_posthog) is from an analytics tool (likely PostHog) that tracks anonymous user behavior and referral sources, while others (tics\_session\_id, detectai\_session, iw\_fp) help maintain sessions and identify repeat visits without necessarily tying data to a real identity.<br>

### Use in Reporting

Image Whisperer is useful for:

* Quickly screening viral images before publishing.
* Flagging potential AI-generated visuals in breaking news.
* Supporting fact-checking workflows.
* Adding context when reporting on misinformation or synthetic media.

As a real-world example,[ Image Whisperer was one of the tools cited by BBC Verify](https://www.linkedin.com/posts/searchbistro_aiverification-misinformation-journalism-activity-7420082420252708866-BrNO/) in the Kamchatka snowstorm AI hoax in Russia in 2026 when debunking the content.

<table data-header-hidden><thead><tr><th width="259"></th><th></th></tr></thead><tbody><tr><td><strong>Capabilities</strong></td><td><strong>Limitations</strong></td></tr><tr><td>Detects likely AI-generated imagery.</td><td>Not 100% accurate. False positives and negatives can happen.</td></tr><tr><td>Highlights visual inconsistencies including textures, shadows and anatomy issues.</td><td>Struggles with high-quality or lightly edited AI images.</td></tr><tr><td>Fast and accessible with no technical expertise needed.</td><td>May not provide detailed metadata or provenance analysis.</td></tr><tr><td>Useful as a triage tool in verification workflows.</td><td>Doesn’t replace forensic tools or human judgment.</td></tr><tr><td>Export full verification reports.</td><td>Limited transparency on underlying detection methods.</td></tr></tbody></table>

### Summary

Image Whisperer is typically used in the analysis and enrichment phase of the OSINT workflow, ideal for surfacing non-obvious clues, but each key finding should still be verified using independent methods.

### Ownership

Image Whisperer is created and developed by [Henk van Ess](https://www.linkedin.com/in/searchbistro/), an OSINT specialist and journalist from the United States focused on digital verification.

### Ethical Considerations

* Avoid presenting results as definitive proof.
* Be transparent about uncertainty in reporting.
* Combine with other verification methods (reverse image search, metadata, context).
* Be cautious when labeling content as “AI-generated” to prevent misinformation.
* Consider privacy when uploading sensitive or unpublished images.

### Related Tools:

* AI or Not
* Hive Moderation AI Detector
* InVID Verification Plugin
* FotoForensics
* Google Reverse Image Search

#### Sources

<https://imagewhisperer.org/>&#x20;

<https://imagewhisperer.org/about>&#x20;

<https://gijn.org/resource/guide-detecting-ai-generated-content/>&#x20;

<https://www.linkedin.com/in/searchbistro/>&#x20;

<https://www.linkedin.com/posts/searchbistro_aiverification-misinformation-journalism-activity-7420082420252708866-BrNO/>