I built a small project over the weekend and wanted to share it here.
It’s called AuthenticImage — a tool that analyzes an image and tells you if it’s AI-generated, edited/manipulated, or unsafe.
Link: https://authenticimage.site/
The main idea was simple:
AI-generated images are everywhere now, and there isn’t a straightforward way for regular users (or even developers) to check if a photo is real, fake, or modified. So I stitched together a few detection models into a unified pipeline and built a fast, no-signup interface.
No account required, completely free during early development
I quietly put it online to see if anyone would find it. Unexpectedly, around 100 people used it on the first day, even without any marketing or posts. That made me think there might be an actual need for a lightweight image authenticity checker.
A few things I’m trying to understand:
Who needs this the most? (journalists, marketplaces, dating apps, educators?)
What matters more: accuracy, speed, or explainability?
Should this evolve into an API, a browser extension, or a standalone consumer product?
Right now it’s just an early MVP. I’d appreciate:
brutally honest feedback
failure cases
feature suggestions
accuracy expectations
thoughts on whether this is worth turning into a serious project
I’m happy to share how the pipeline works internally if anyone’s curious.
Hi HN,
I built a small project over the weekend and wanted to share it here. It’s called AuthenticImage — a tool that analyzes an image and tells you if it’s AI-generated, edited/manipulated, or unsafe. Link: https://authenticimage.site/
The main idea was simple: AI-generated images are everywhere now, and there isn’t a straightforward way for regular users (or even developers) to check if a photo is real, fake, or modified. So I stitched together a few detection models into a unified pipeline and built a fast, no-signup interface.
What it does right now:
Detects AI-generated content (Stable Diffusion, Midjourney, etc.)
Detects manipulated/edited images
Flags unsafe content (nudity, violence, gore, etc.)
Runs in ~100ms on average
No account required, completely free during early development
I quietly put it online to see if anyone would find it. Unexpectedly, around 100 people used it on the first day, even without any marketing or posts. That made me think there might be an actual need for a lightweight image authenticity checker.
A few things I’m trying to understand:
Who needs this the most? (journalists, marketplaces, dating apps, educators?)
What matters more: accuracy, speed, or explainability?
Should this evolve into an API, a browser extension, or a standalone consumer product?
Right now it’s just an early MVP. I’d appreciate:
brutally honest feedback
failure cases
feature suggestions
accuracy expectations
thoughts on whether this is worth turning into a serious project
I’m happy to share how the pipeline works internally if anyone’s curious.
Thanks!