[kbuck]

On most (all?) carriers, you can forward spam SMS messages to 7726 ("spam" on the keypad) to report messages as spam.

That said, I've got no idea if they actually do anything actionable with this data. It certainly doesn't seem to have reduced my spam volume. Now I just let Android Messages filter the spam out.

[reaperducer]

On most (all?) carriers, you can forward spam SMS messages to 7726 ("spam" on the keypad) to report messages as spam.

I wonder if that's what the iPhone's "Report Junk" button does with text messages.

For some reasons, my iPhones on AT&T always offer the option to report a text message as spam, but my iPhones on Verizon do not. Another curiosity.

[aendruk]

Tell me more about this report button! I’ve long wished for such a feature but thought it didn’t exist. This is something built in to iOS?

[troydavis]

I’ve wondered the same thing. I’ve used 7726 to report large, long-lived campaigns (to AT&T Mobile’s 7726) and as far as I could tell, nothing happened. The senders rotate TNs so often that AT&T would either need to track it back to the point of ingress or do content-aware blocking.

[aendruk]

My most recent experience was enduring several weeks of daily spam, all diligently forwarded to 7726 (“Thank you for reporting SPAM. We'll take it from here.”), only to finally get fed up enough to send a complaint to the FCC after which the spam stopped immediately.

[aendruk]

In iOS when you forward a message, bafflingly, it does not copy the original source address, rather just the body so depending on the message you’re likely to be either misleading the recipient, plagiarizing without attribution, or sending spam content firsthand. Contrast with email in which convention is to copy a few headers to preserve context through forwarding.

In the case of 7726, I’m further confused that there seems to be no acknowledgement of this source of ambiguity. Do they want to know the source of the spam, so I should manually add it to the message? Or are they just training a content recognition model and by sending anything other than the original text verbatim I’m throwing it off?

Also, when the forwarded spam contains a URL, iOS often automaticity chops off that part of the message and shows an unhelpfully truncated version of it below the message in a separate bubble. Is iOS treating the forwarded spam as trusted data and probing the spammer’s URL, tracking parameters and all?

[badcppdev]

Sending it to 7726 prompts your service provider to identify the matching incoming message (which it has in its logs) as spam and investigate, etc.

[hollosi]

SMS forwarding sends the content, not the sender (phone number) info. In principle, they can search your previous messages and find the sender, but it's unclear if that would raise some privacy issues.

Most likely they just use the reported message to train their spam filter, not to block the particular sender number of that message.

[ericbarrett]

You're probably right, but mining the results of that trained filter should provide quite a bit of data to a telco even so. It's not like it's a new bad actor for every message.