Q, does the meme sniffer program work the other way around as well? Rendering “safe” when less than 20% is added by anons?

If so, we could make a lot of fakenews memes, using their screenshots and pix as a trojan horse and post our text on them.