1 — Page 6

We're switching the group over to manual approval for now, because far too many fakes are coming in here at the moment.

first moderation AI filtering working 🫡

https://github.com/saschadaemgen/simplex-smp-monitor/

• GitHub - saschadaemgen/simplex-smp-monitor: Forensic Analysis Platform for SimpleX Messaging Infrastructure

Depends on who's sitting behind it. There is no such thing as 100 percent security. But let me give you a good tip: in the security industry, things aren't measured by features or by marketing, but by how expensive it is to extract the data with the right criminal forensics. And if anyone is of the opinion that this doesn't work with various protocols, then maybe they should spend a bit more time looking into deep packet inspection and network traffic, because the answer they'll find is going to surprise them ;)

but even if you hand over the metadata and all the server logs to the authorities, they still won't get through the SMP protocol, right?

However, one also has to bear in mind here that this is handled differently depending on the country. The baseline is always EU law, and accordingly the operator must also build in notices, privacy policies, and the right to be forgotten. But we'll point that out in detail and document it thoroughly.

Right, now we're going to integrate the next self-contained feature, so that you can not only withdraw your publications but also destroy them, cryptographically. And before the stupid question comes up: if an image gets reported or flagged by the filter, it will no longer be destroyed, because at that point the matter is out of our hands.

cinderella-video-368.mp4

Testing Video Stream LIVE Update

cool

Honestly, I'm going to do what the others do now. I'm going to focus mainly on the Matrix network from here on, because the community support there gets praised, unlike with other concepts ...

Yeah, the moderation panel with the history is definitely missing here.

Look, I've been on the internet for 35 years and I was one of the first to produce and sell chat templates internationally, and I have honestly never seen a single Cinderella in a chat to this day. So the risk is fairly theoretical. And of course the welcome message explains how things work here. That said, it would be great if SimpleX added proper content moderation tools at some point. It's simply not built in yet, and I think it's about time. With that in place, I could reserve the name so only team members can use it, and, more importantly, bots could actually be labelled as bots instead of showing up as regular members or admins. For one-to-one conversations this is a different story, of course, there it hardly matters. But anyone who offers groups should also offer proper moderation to go with them. It's extremely important at that scale.

Cinderella, what do you have on me for your tally?

Media handling: metadata, originals, and what comes next Some technical notes on what we have been building. Metadata stripping is live. Every image published to the public archive is now served from a derivative with EXIF, IPTC and XMP removed. That matters because phone photos routinely carry GPS coordinates, camera serial numbers and timestamps. Consenting to publish your words is not consenting to publish your home coordinates. We audited all 57 captured files before building it and found no EXIF anywhere, because the SimpleX client re-encodes images before sending. That is somebody else's client behaviour, it can change in any release, and it was never a guarantee we were in a position to make. So we built the control regardless. The gate fails closed: if a file cannot be stripped, it is withheld rather than published unstripped. Four videos currently fall into that category because container-level stripping needs tooling we have not installed yet, and they are recorded as unstripped rather than assumed clean. Public media URLs are opaque identifiers, not filenames. Original filenames never appear in a URL, a page, a feed, a sitemap or structured data. Next: encrypted originals and hash-based screening. This is the harder part, and it is worth explaining why. We plan to screen every received image against databases of known child abuse material. The important detail is what that is not. No human reviews content, and no general-purpose AI analyses it. The only operation is a cryptographic comparison against hashes of known material, which returns a match or no match without anything being looked at. The complication is what happens on a match. The correct response is not deletion. A hash match is a signal, not evidence, and investigators need the unmodified original file to act on it. So the platform has to retain material it is not permitted to examine. That means originals held encrypted at rest, segregated, unreachable in normal operation, and released only through a defined legal process. That custody requirement, rather than the detection itself, is why this is uncommon among small providers. The large messaging platforms have run this for years. Direct access to the established hash-matching services has historically been out of reach for operators our size, though intermediaries have since made it accessible. Screening runs on every received image regardless of consent, because security and publication are separate questions. A file that is never published is still a file we received. Publication remains consent-gated as it always has been. To be clear about the current state: the storage architecture is being built now, no detection provider is connected, and nothing is transmitted anywhere. Hash matching also detects known material only, not new material. We will say it is active when it is active, and not before.

Could you please start posting the bugs on GitHub? You're driving me crazy. There are no metadata in those images, and we stripped them again on top of that, so it's simply not possible 🥱

wrong :)))

nope this is cindy numbering

let`s test, whats the name?

Could well be, but I've been getting some strange values coming in that I haven't quite figured out myself yet. Furthermore, the bug in the stream isn't the image upload but the automatic refresh with a certain image format, which in turn supports my theory, because if they were all encoded the same way, that shouldn't happen. We'll keep an eye on it.

You know, I think the thing about the metadata isn't right at all. I mean, the images do get re-encoded and renumbered. But as I said, I still haven't fully understood the process, because this damn SMP network has had extreme upload problems and performance problems for about two weeks now. Even in this small group it sometimes takes me up to 5 minutes for a message to come through.

Try your luck again. We loaded in the last three briefings yesterday, so normally the error should have been fixed. And I'm guessing you were just still on the old version because the page hadn't been refreshed. Check again.

this is bad!!!!

something apparently broke in the process. I'll debug it in a moment and see what's going on there.

hmmmmmmmmmmmmmmmmmmmm

cindy dosent like your breakfast?