AI detection is a serious problem because of scams on one side and false accusations on the other. But while building our AI detector, we realized that if people are scared or bored by your tool, they won't use it, especially when there are plenty of other options on the market. And if they don't use it, all that serious tech doesn't matter.
One of the very first decisions for our product was to make it fun and easy to use. We'd already tested different AI detectors, and all of them were very academic, clean, and serious. Not saying that's bad, but we thought, what if AI detection felt less like a security checkpoint and more like having a curious friend look over your shoulder? That's why we created this fun logo, mascot and playful interface. Later on, we though, why not use the flame from the logo and the cat from the main page as a couple of cute AI detectives? They now show up in the interface, our social media and in an animated series (I'll talk about it in the next post). They make the whole topic of AI detection feel like and adventure, not like an interrogation.
Aside from our little heroes, the interface itself also shows visual evidence of AI, inviting you to look closer, notice details, and learn something new about AI detection while checking content, instead of tons of textual explanations. We wanted it to say "hey this part of yuor content looks suspicions, see?"
All of it because we want to solve not only tech problem, but emotional as well. This way we hope people want to check if something's real or AI.
Shepherd
Congrats on the launch! The fact that you show why something was flagged (not just a yes or no verdict) is a great differentiator. The visual heatmaps and highlighted fragments are also a smart move for building trust.
I'm keen to know - do you have plans to offer a browser extension so people can check content directly while scrolling social media?
@yangfanyun Thank you!
We realized that people don’t need just a single % They want to understand what exactly was flagged so they can draw their own conclusions. That’s why we show highlighted fragments and visual heatmaps instead of a simple yes/no verdict.
And yes, a browser extension is in our roadmap. The goal is to let users check content directly while scrolling social media, in one click.
Scade.pro
@yangfanyun thank you so much Yang, yes we have it in our vision, but currently we are focused on improving our detectors quality
congrats on the launch ! I like the user-friendly interface. Do you see a way where your tool could be integrated into applications (social-media platforms, e-commerce stores, ...) so that we users won't have to drag-an-drop every piece of content manually but would be visually informed within the applications about the nature (fake, true) of teach piece of content they see ?
@julienergan Thanks!
Yes. In the near term, we’re adding link-based checks (paste a URL), a browser extension, and other convenience features. Longer term, we plan to offer integrations (API/SDK) so platforms can show in-app signals like highlights/heatmaps instead of a simple fake/true label.
Scade.pro
@julienergan we really want it to be this way, although we just launched. We are working on API and we'd be happy to collaborate with online stores and platforms for sure
@tereza_hurtova Thanks for the thoughtful questions 🙌
Primary users today:
Individuals checking suspicious media (scams, deepfakes, AI text)
Businesses that need verification inside their workflows
And yes, it is an arms race. As generative models improve, detectors must constantly retrain and evolve.
It’s very similar to antivirus software: threats evolve -> protection updates.
Detection isn’t a one-time solution, it’s a continuous process.
Our edge comes from continuous model updates, proprietary datasets, and multimodal detection methodology.
Congrats on the launch! Sounds like this could be really useful for school plagiarism detection. I wonder if it can also detect AI generated images from a partial / cropped image though?
Scade.pro
@ler_ws thank you Wilson! yes it can, if the image is completely generated
Awesome! So is it able to detect video reels on FB and other social platforms while im scrolling?
Scade.pro
@shahmeer_baloch1 Thank you Shahmeer! Not yet, rn you can upload a video file and get results, but soon well add a field for a link and it will be much more convenient.
Hopefully one day we'll be able to show you that something you are looking at is Fake right on Facebook or Instagram :)
@shahmeer_baloch1 Thanks!! we wish to do so, but not yet :)
Congrats on the launch! How does it work and how accurate is it?
Scade.pro
@janschutte Hi Jan, thank you for your question.
We work in 4 formats so I'll break down my answer:
for Text we analyze patterns like perplexity and burstliness of the speech. They show predictability of the text and variation in sentence complexity, overuse of linking words, suspiciously perfect formating etc. Humans are messier than AI and less predictable in word choices and syntax. For text you get a confidence score how likely it's writtern by human or AI and highlighted fragments that are likely AI
for Audio we analyze waveforms and spectograms for missing breathing, micro-variations, pitch monotony and some cutoffs that human voices have usually. @olga_scry will add more for images and video
@janschutte @nastassia_k Thank you for joinging our launch day! And for your question!
Speaking of images we scan pixel level data, some noise artifacts, and AI models fingerprints. Our heathmap shows what is likely AI generated or heavily AI edited.
For videos we do frame-by-frame analysis, we notice lip sync errors, blinking anomalies, melting backgrounds, even skin color changes. isFake flags suspicious frames that are more likely AI
@janschutte @nastassia_k this is an example of a heathmap in AI image detector
@nastassia_k @olga_scry Thanks you for the thorough response!
Fakeradar
My iPhone uses AI to enhance images when recording videos. Is it labeled as AI?
@artem_anikeev Thank you for your question, Artem 🙂
We did face challenges with iPhone photos and videos because Apple applies strong AI enhancements. We’ve since improved the model, and it now handles iPhone media much more accurately.
Filters (Instagram, TikTok, editing apps) can still influence the score, since they alter image patterns.
We also provide a percentage score and heatmaps, so you can interpret the result yourself — there’s a clear difference between 75% and 95-100%.
Fakeradar
@olga_scry Say hi to Fire and Kitten from their colleagues ;)
Scade.pro
@olga_scry @artem_anikeev Is and Fake are here to say hiii and meow :D