API-first solution combining NLP/visual/audio authentication. Detect synthetic media and malicious intent patterns.
This is the 2nd launch from Scam AI. View more
Halo by Scam AI
Launching today
The person on your next video call might not be real. With Halo you don't have to guess. Halo secures your Zoom, Teams, or Google Meet call live and flags synthetic faces the moment it detects one, entirely on your device. Deepfake video calls are already being used to scam people and businesses around the world, it's just that most people have no way to tell. From confirming who you're hiring to confirming who you're wiring money to, Halo catches it before it costs you.
Payment Required
Launch Team / Built With
Scam AI
Hi Product Hunt 👋
I'm Neo, co-founder of Scam AI, and I'm excited to introduce Halo a real-time, on-device deepfake AI model for video meetings. It secures your Zoom, Teams, or Google Meet call live and flags synthetic faces the moment it detects one.
In 2024, a finance employee at Arup joined a video call with what looked like his CFO and several colleagues. Every face on that call was a deepfake. He wired $25 million before anyone realized.
We kept coming back to that story because almost everything built to catch deepfakes works after the fact: you upload a recording, wait, and get a report once the decision has already been made. Nothing was built to catch it while the call is still happening, when it actually matters.
What Halo does:
Secures Zoom, Teams, and Google Meet calls live
Flags synthetic faces in real time, in real time
Runs entirely on-device, no recordings, no cloud inference
Built for the moments trust actually gets tested on video: confirming who you're interviewing, confirming who you're wiring money to
Halo is live now at scam.ai/halo. I'll be here all day, happy to answer anything about how it works, and to help however I can!
AgentX - Multi-agent and eval framework
@neo_tiangratanakul great stuff! Incredible and timely tool for stopping deepfakes and scams before they strike!!
Scam AI
Thanks so much! Really appreciate the support 🙏
The Arup case is exactly the story that makes the case for this on its own. Curious about the failure mode in the other direction though, what happens when it flags a real person as synthetic because of bad lighting or a compressed webcam feed mid call. Does it surface a warning you can dismiss, or does it interrupt the call, since getting that wrong costs you a real relationship, not just a false sense of security.
Scam AI
Really good question! Honestly one we thought about a lot. We made it so that Halo checks the quality of what it's seeing: bad lighting, shaky video, heavy compression, or if a face is too far away from the camera. If the conditions are bad it gather more data or waits for the lighting or image size to get better later in the call, and only makes a decision once it actually has good enough data to be confident. So a bad frame on its own isn't going to set off a false alarm. Since, just like you said, getting that wrong costs a real relationship.
@neo_tiangratanakul Gathering more data before deciding rather than reacting to a single bad frame is the right shape for it. Good to hear that was already thought through rather than something you're bolting on after launch.
Scam AI
Haha yeah we're glad you're thinking about the same issues we faced as well. Thanks for all the support Raffay!
Scarlett.
Congrats on the launch, Neo!
I have 2 questions:
How does Halo detect new deepfake models it hasn’t seen before? And my second question: how do you avoid false alarms during calls?
Scam AI
Thanks Aleksandar!
Halo is generalizable to do alright against new deepfake methods but to make sure it keeps up to date we have an agentic system that constantly scrapes the web for new deepfake methods, collects data, and fine tunes our model towards these new deepfakes.
For your second question: Halo runs a quality check on the input itself before it ever makes a call. Bad lighting, shaky footage, heavy compression, or a face too far from the camera can introduce noise, so we filter that out first. On top of that, if the model's ever on the fence and not fully confident, it pulls in more signals before deciding, so we don't throw false alarms during calls!
Does it hook into the system audio/video pipeline via virtual camera, or is it reading the display buffer directly? congrats for shipping 🙌
Scam AI
Thanks Vikram 🙌 It's neither, actually we use WGC to grab the rendered meeting window directly. That way our model sees and takes in the same screen the user is seeing!
Congrats on the launch, Neo! The "catch it during the call" angle is what really sets Halo apart — most deepfake tools I've seen assume you already have a recording to analyze after the fact, which is basically useless mid wire-transfer conversation.
One question: since Halo runs entirely on-device, how does it actually hook into the meeting client? Virtual camera, native SDK integration, or a browser extension for Meet? Curious how the coverage differs across Zoom desktop vs. Meet in the browser.
Scam AI
Thanks Rohit and I totally agree, after-the-fact detection doesn't help much for situations like a wire transfer. Hence why we're so proud to have made a product like Halo!
Halo actually uses WGC to see the rendered meeting window directly, the same way whether it's Zoom desktop or Meet in a browser tab. Which allows us to cover every video conferencing app and browser!
minimalist phone: reduce your screentime
Remember how some guys created a tool that creates AI live video chats. It seems that when you create one solution, you need to create another "anti-solution".
Do you already have your user base?
Scam AI
Thanks so much! Honestly, this "anti-solution" space has always been super inspiring to us, there's something exciting about getting to keep building things that protect people as fast as this space moves. Though we see some legitimate use for AI video generation, we also see how much harm it can cause when it's used maliciously. But honestly, stopping bad actors from abusing AI is a big reason why we started this journey in the first place!
As for user base: we've actually already got some enterprise customers using our detection tech, and Halo is really us expanding that out to the consumer side now 🙂
Scam AI
Around 1 in 7 digitial onboarding fraud attempts involve deepfakes and there has been a 16x increase in deepfake attacks in the last 2 years. From our enterprise customers we're seeing these same statistics parroted. I think we're lucky that right now it isn't too wide spread but we believe this to be more common place in the next year. Which is why we built Halo to get ahead of the game!