False-positive PGP spoofing

Talk of the Cicada PGP message still continues today.

But this post isn't specifically about the Cicada PGP message or anything to do with Cicada 3301.

It's only related because the idea for this post come from discussion on Twitter.

Before you continue reading, I must point out that what I am publishing here is nothing new and has been talked about before, probably extensively.

I am simply presenting it in simple terms so it can be understood by anyone and so that anyone can try for themselves by following the same steps.

Admittedly, PGP is a real weak spot for me so I did get some help with this from Twitter user @dean_emmet.

I am using Kleopatra.

The first thing you need to do is import the Cicada 3301 PGP public key.

I have dumped it here https://pastebin.com/a1ZcAGmE

Import the key into Kleopatra.

Once imported you should see it in your list.

Now create an empty file and dump the following message into it:

Now verify the message.

As you can see, the message has been verified with the Cicada 3301 key we imported – April 5, 2017.

Note, the original Cicada PGP message was posted April 4, 2017, however the date presented for verification will vary depending on your timezone.

Moving on.

To show how a message that has been tampered with may fail verification, we will add some random text into the message.

Now try to verify the message.

You can see the verification now fails because the original signed message has been tampered with.

So how do you spoof the verification?

If you edit only the third line of the message it will create a false-positive spoof.

Now try to verify the message.

The key to understanding what is really happening is found inside the verbose output.

The verbose output suggests that the message may actually be authentic and has passed verification.

But look closer at the line that says:

gpg: invalid armor header: THIS TEXT SHOULD NOT BE HERE SIGNED Cicada 3301\n

This single line could easily go unnoticed.

However, it is significant because it’s actually the key indicator which is telling you that the original signed message has been tampered with, despite the verbose output that follows the warning which falsely indicates that it has passed verification.

Most people do not know or understand this and will only see the output which says it has passed verification.

This, generally speaking, will be being enough for most people to be satisfied the message is authentic.

This is false-positive spoofing.

One might ask, how is this even important?

It's important, in my view, because if any PGP message was intercepted prior to distribution it could theoretically be spoofed as a false-positive and then the spoofed message distributed in its place.

To most people who really don't know any different, it would simply be interpreted as authentic.