Our mission
A safety layer for the people most targeted.
Someone close to us was targeted by a deepfake. A friend's younger sister was harassed with images online. Each time, the pattern repeated: by the time anyone could react, the harmful content had already been seen, screenshotted, forwarded, and amplified.
The tools meant to stop this all lived in the cloud. To ask "is this image safe?", a victim first has to upload their most private moment to someone else's server. The safeguard was its own kind of exposure.
So we made a different bet: put the safety check at the edge. On the device, in real time, private by construction.
SafeScreen started as a focused engineering build. We proved two compact AI models could run fast enough, on a phone, to blur harmful content before you see it, without a single byte leaving the device. It worked. Now we're turning it into something everyone can put on their phone.
of people targeted by deepfake abuse are women and girls. The technology that creates this harm is getting cheaper and faster. The defense has to live where the harm lands: on the screen, in the moment.
What we stand for
Three commitments we won't trade away.
Protect the most targeted
Image-based abuse falls hardest on kids, teens, and women. We build for them first.
Privacy is non-negotiable
A safety tool that harvests your data is not a safety tool. Everything runs on your device, with nothing to upload and nothing to leak.
Honesty over hype
We tell you what works, what doesn’t yet, and where the edges are. Confidence you can verify beats claims you have to believe.
A private, real-time safety layer for everyone, starting with the people who need it most.
Next on the roadmap: faster on-device acceleration, region-precise blur, and text-based protection.