The BBC is now reporting that “thousands” of documents have been removed because the DOJ improperly redacted information that can be used to identify the victims: https://www.bbc.com/news/articles/cn0k65pnxjxo

One Redditor said that they reported more than 500 nude images to the DOJ, all from Dataset 9.

We didn’t have trouble getting Datasets 10, 11, or 12. I think Dataset 9 was probably delivered fine on Friday, so the NYTimes was able to grab a complete copy. Then, NYTimes started reporting the abusive material, which prompted the DOJ to yoink the ZIP, and it’s been screwy ever since.

I saw a post from a random Redditor confirming that they found abusive material, if that’s the concern. I doubt that the reports are fabricated, but I also agree that the reports are a great excuse for the DOJ to remove legitimate files.

Hasn’t moved AFAIK, just going slowly.

Thank you! I’m not very tech savvy, so I’m very little help in this whole process. Please LMK what you find.

It’s “great” that the DOJ removed CSAM at the same time as they were removing perfectly legitimate files that are in the public interest. That’s just really smart. Puts us all in a hell of a bind.

I can’t speak for others, but I’ll plan to preserve the 87GB Set 9, the 90GB Set 9, and Set 10, until we can get an updated “complete” (current) Set 9 that can be presumed to be free of CSAM. After that, we can try to identify the legitimate files that are missing from the “complete” Set 9, and preserve those while purging the CSAM.

Amazing - Once you have the 180GB Set 9 downloaded, I’ll seed.

At this point, my working assumption is that the version you’re downloading should be presumed to be free of CSAM, but we can’t know for sure until we check it. In addition, I assume that legitimate files were also removed from the version you’re downloading, but the legitimate files are preserved in the archives we already have (along with, tragically, the CSAM.)

I think that after you download the 180GB set, we should compare it to our existing files to identify files that were removed. Then, we can identify which of the removed files were CSAM, and which of the removed files were legitimate. Going to be a hell of a task…

100% - That’s why I haven’t deleted my copy of Set 9. I have no plans to unzip it, and I’m glad that DOJ is removing the CSAM now, but I’m going to hold onto the set to preserve the valuable docs that DOJ is removing.

For those curious, here’s the NYTimes article where they report on the CSAM in the publicly-released files: https://www.nytimes.com/2026/02/01/us/nude-photos-epstein-files.html (behind paywall.)

NYTimes says that they discovered the CSAM on Friday and notified the DOJ on Saturday, and the DOJ was diligent in removing the files NYTimes had flagged.

NYTimes does not say that the material is in Dataset 9 specifically, but we observed that the DOJ was removing files from Dataset 9 on Saturday and not other datasets, so the server behavior would be consistent with CSAM in Dataset 9.