Rachel Maddow Deepfaked on YouTube with Millions of Views of Slop Videos
Update on Jan 7: the channel has been scrubbed from YouTube. I suppose they finally got around to evaluating all of the many reports submitted!
As of right now (Jan 6), there’s a deepfake AI clone of Rachel Maddow on YouTube called “Maddow’s Brief”. It’s apparently fooling lots of people, because overnight it’s grown from 20K subs (when I first found it) to 38K. Just one of the slop videos, posted less than 24 hours ago, already has a million views.
I reported two videos to Google last night. So far, nothing has yet been taken down. Many people in the comments are saying things like “Right on, Rachel Maddow! Telling it like it is! Screw Trump!”
This is all generated slop. It is clearly not actually Rachel Maddow, and even if what “she” is saying happens to be true when evaluated, it is still nonsensical. It is not grounded in any real truth claims by any verifiable human sources. It is a con job. It is spam.
Why is YouTube allowing this sort of thing to take place at all? From an ML standpoint, it should be relatively trivial to flag such deepfakes in an automated fashion and immediately escalate to human review—especially for a new channel like this (it only started on Jan 1, 2026!) sporting a deluge of videos in a brief period of time.
This is not some sort of “oh gosh, now what do we do?” problem. This is intentional. YouTube & Google have intentionally decided they don’t care about deepfakes and slop. Because if they really wanted to do something about it, you wouldn’t see this happening. Period.
Without a significant change to the policies of platforms like YouTube around this stuff, it is the beginning of the end of the Internet as we have known it. We cannot allow this madness to continue. #GenerativeAI is an existential threat to knowledge and truth.
