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.
Hey Product Hunt! 🐱
I'm Olga, and our team of cybersecurity researchers and ML experts who built isFake AI.
We were tired of watching trust collapse in real time, that's why we created the world's first (and only fun) multimodal AI detector.
AI content now looks identical to reality. That shift broke trust everywhere: big outlets publish AI hurricane videos, great teachers wrongly accuse students on genuine essays, grandparents wire money to synthetic voices of their grandkids. We couldn't just watch this happen.
And no, we don't want to ban AI, we use it ourselves. But we realized all of us need a way to check truth without bias or regret, and that tool simply didn't exist.
So we made isFake AI...
...multimodal. We didn't want to build another text checker. Real misinformation doesn't respect format boundaries, so we built AI detection for text, images, code, audio, and video. All in one place
...visual and transparent. We believe you deserve to understand the "why." That's why we don't just give you a verdict AI or NOT. We highlight suspicious fragments so you can see what triggered the alert.
...fast. We know trust needs speed. Even with videos :) Upload and get your answer in seconds because waiting kills verification in the moment you need it most.
...informative. We wanted everyone to verify viral content instantly, so we built free AI content library (Explore page) where you can check trending posts without even running the detector yourself.
... fun. We have a pretty simple interface and 2 little helpers who make transparency entertaining: 🔥 Is (a spark of digital wisdom) and 🐱 Fake (a curious cat). They're the heroes of our animated series where they catch celebrity deepfakes, unmask romance scams, expose AI voice clones, and crack jokes while solving internet mysteries with isFake AI. Meet them in our Demo video.
What our users do with it
People use it for protecting themselves from scams and financial losses, teachers verifying work fairly without false accusations, journalists stopping disinformation, companies safeguarding their reputation.
🎁 One more little special for the Product Hunt community:
Early adopters get 40% off for the first 6 months
Use promo code: PHFOUNDERS
Limited to the first 100 users
We want to build it for you, so please share your feedback, we are here to answer any questions.
@olga_scry congrats on the launch, how does it compare to GPTZero and other detectors? Wishing you all the best with it.
@mogabr Thank you 🙂 Direct comparisons can be tricky since we’re trained on our own proprietary datasets and models. Our main difference is that we are multimodal, not just text, but also image, video, and audio detection.
we don't want to be just an AI detector, we’re building toward detecting misleading and scam content in general, not only identifying whether something was generated by AI
@olga_scry @mogabr thank you, I'll answer from the positioning and brending side and Olga will reply from the tech side.
We are the only fun AI detector :D other tools are very academic, it's not bad, but I feel like they create a distance between them and their audience. We want our users be curious about AI and AI detection, not fearful of it. Hopefully our easy-to-use interface, design, playful tone of voice, branding and everything around product helps us with this :)
@maria_anosova thanks Maria! it depends on the video length, usually it takes under 30 sec, we constantly improve the speed
@maria_anosova on behalf of Is and Fake, huge thank you for finding them cute :D
Love your mascots! Curious: what inspired the choice of fire for your logo? 🙂
I tested your tool on a piece of content that was generated by AI based on insights / ideas provided by a human. The tools hasn't flagged it as AI-generated. What criteria do you use to identify if it's AI content?
@alina_petrova3 thank you! Great question 🙂
fire represents “stress-testing” content, and and also our idea of burning away misleading or harmful synthetic content.
AI-generated fragments are highlighted in red, and for media we use visual heatmaps to show where synthetic signals are detected
@alina_petrova3 About your test, we know AI is used almost everywhere now.
And if content is strongly shaped by human thinking and editing, AI signals become much weaker, and the score decreases accordingly. our detector is mainly focused on clear synthetic generation like mass auto-generated articles, spam, scam content, and low-effort outputs. In hybrid cases (AI + real human input), a low AI score is expected and aligned with our philosophy :)
@alina_petrova3 Thank you so much Alena. Olga will say more about tech part and I'll tell a little story about the fire. When I just joined the team, one of my first questions was what is the name of their cat on the main page. And during this discussion guys called him Fake, and said, well, we also have a flame in logo, maybe we can call him Is and have like a couple of mascots "Is and Fake". And this way the couple of our cute detectives was born :D
we even have a cartoon series with them on youtube
@alina_petrova3 @nastassia_k yep, love them :)
@nastassia_k So cool 😀
According to what parameters does the tool know the exact percentage of AI usage?
@busmark_w_nika Hi Nika, thanks for your question.
The percentage is calculated using trained neural networks that analyze patterns typical of AI-generated content.
Each detector (text, image, video, voice) evaluates statistical signals and generation artifacts in the file you upload and returns a probability score of AI usage. The result depends on the specific content.
We wrote articles on how AI detection works in every format, sharing in case you are interested for a detailed explanation:
Text – https://isfake.ai/how-to-detect-ai-generated-text
Image – https://isfake.ai/how-to-detect-ai-generated-image
Video – https://isfake.ai/how-to-detect-ai-generated-video
Audio – https://isfake.ai/how-to-detect-ai-generated-voice
@busmark_w_nika Nika, thank you for reviewing our product :) it depends on the format of content, @olga_scry shares how it works under the hood
Looks nice. Is there a browser extension planned? Would love to just right-click an image and check it on the spot — that would be most convenient.
Or maybe even a desktop app?
@ovarious Thank you 🙂We’ve just launched recently, so right now we’re focused on expanding and improving our core detectors.A browser extension is definitely in our roadmap, as for the desktop app we'll discuss it with our team, thanks for your idea!
Congrats on the launch! I can see this being a game-changer for KYC and online verification. I noticed you have a web platform, but are there any plans to release a browser extension?
@new_user__289202369678691c937c656 thank you :) it's in our roadmap, but as we’ve just launched, so mainly focus on improving the quality of our ai detectors
@olga_scry It's just fascinating to view ProductHunt from this perspective :) I wonder how much of that space is now dominated by AI.
@new_user__289202369678691c937c656 thank you!
Congrats on the launch! Have you tested it against the latest image generators?
@builov84 thanks. yes, but it's a really moving target, cause new models drop daily, so for us it's like an ongoing process all the time.
@builov84 thanks, good question. Constantly working on it
Important problem to tackle — trust really is the bottleneck of this AI era. I like the multimodal approach and especially the visual highlighting instead of just giving a binary verdict.
Curious how you handle edge cases where content is partially AI-generated and partially real (e.g., edited videos or mixed media)? Would love to understand how nuanced the probability scoring gets.
@hossein_r thank you for your question. that's one of the trickiest problems in AI detection, so we work alot on it. Because of this we don't do binary AI/Human so what you get as a result is a probability score with highlighted fragments, specifically because mixed content is super common.Actually partially AI is a new norm online, rather than edge case, lots of people edit photos with AI, we even have Ai enhancements in iPhone camera. That is why we build the scoring and visual highlights of particular fragments that are likely AI-generated