Hyde: AI Bullshit Detector

Hyde: AI Bullshit Detector

Verify ChatGPT, Claude, Gemini, Perplexity in real time.

5 followers

Hyde checks if AI is lying to you - in real-time, as you chat. ChatGPT and Claude make up fake sources, statistics, and facts. You don't notice because they sound confident. Hyde catches these lies instantly. Two characters watch every AI response: Pocket Devil spots the bullshit, Clown keeps it fun. They can either warn you or argue with the AI directly while you watch. It's like having a fact-checker sitting next to you, but automatic. Works on ChatGPT, Claude, Gemini, Perplexity, DeepSeek.
Hyde: AI Bullshit Detector gallery image
Hyde: AI Bullshit Detector gallery image
Hyde: AI Bullshit Detector gallery image
Free Options
Launch Team
OS Ninja
OS Ninja
Explore and Learn Open Source using AI
Promoted

What do you think? …

Yuka Kust
My brother Vitaly's birthday was this month. We've been vibeсoding together since October nonstop, so I decided to build him something with my own hands. I thought about what pisses me off most. Remembered this moment: I ask AI, "Did you write all the code? Everything works? Logic is solid?" AI confidently answers: "Yes, absolutely, everything's done perfectly." Cool, let's run it. Nothing works. I exhale. Spend 1.5 hours finding holes in the logic. Fix everything. Launch again. Still broken. Third attempt - same shit. Then I add logs (yeah, I'm a coding newbie, should've done this from the start😅), and realize step 2 doesn't even exist. AI just hallucinated it finished and confidently reported mission accomplished. That's when it hit me: don't want to waste my energy on it, what if AI could check AI? What if something could catch these lies in real-time? Two sleepless nights later, I bring Hyde v1 to Vitaly's birthday. At the party, we show our whole family what vibeсoding is and how the world is changing. Vitaly and I are on our family's frontline of change - they're always very curious what we bring. This time: computer vision algorithms and Hyde. Everyone who uses AI (spoiler: less than half our huge family) instantly recognized their own pain. "Send us this extension NOW." We decided to polish a new version and share it with everyone who might need it. Hope that's you. Use it, improve it (open source), and give us feedback. Two brothers. One painful lesson. Your protection from AI bullshit.