54

My dad sent me this video asking if he should be worried about this?

The video shows:

  • a wifi AP broadcasting an airport's wifi name
  • security researcher seeing the sites the victim browses
  • security researcher viewing files accessed by victim on cloud storage
  • victim installing attacker's "free wifi" app
  • the app giving the security researcher full control over the device

Obviously most folks should be wary of untrusted WiFi networks, but there’s a couple of strange things occurring that makes me wonder if this is just an over-hyped hacker story.

First, the Google search he performs seems to be protected by TLS, how is that possible with just MitM?

Then he does some truly mind-blowing stuff like being able to access the microphone, record audio and send it to himself. No way that’s done via just an MitM over WiFi.

Am I missing something, or does this community concur that this video is either over-simplified or just plain deceitful?

Connecting to a strange WiFi might get you into trouble, but it alone cannot cause this level of compromise... can it?

3
  • 17
    I am not sure a user connecting to an open AP could be considered a Man-in-the-Middle attack. I'd qualify it more as a social engineering attack: you present yourself as some "safe" entity (be it "hi this is your faculty's IT department" or "hi I'm the electrician from the insurance company" or "hi this is your airport's WiFi AP") and convince them to give you sensitive information / access to their device (be it "I need your user/pass to fix your computer" or "I need to enter the server room to check the wall sockets" or "I need you to install this app for you to have free WiFi").
    – walen
    Commented Jan 20, 2020 at 8:48
  • 9
    Also, the purpose of such videos is not to be educational or useful, but to urge the viewer to share it, and to gather ad clicks on the website they are shared on. Therefore they are often over-sensational and inaccurate, even in cases they do contain a grain of truth.
    – vsz
    Commented Jan 21, 2020 at 5:13
  • @walen The two aren't exclusive. Most MITM attacks exploit social engineering. Heck, even the stupidest viruses like ILOVEYOU were really social engineering.
    – Luaan
    Commented Jan 22, 2020 at 12:09

5 Answers 5

72

Am I missing something, or does this community concur that this video is either over-simplified or just plain deceitful.

I wouldn't say it's deceitful, but it's definitely overhyped/oversimplified.

First, the Google search he performs seems to be protected by TLS, how is that possible with just MiTM?

Yes. In order to do that, he would have to either strip SSL or install a root CA certificate on the mobile device. So you can't simply MITM https websites (the video over-simplifies it).

Then he does some truly mind-blowing stuff like being able to access the microphone, record audio and send it to himself. No way that’s done via just an MiTM over WiFi.

Of course not. You cant just use a phone's microphone via MITM over wifi. As you can see in the video itself, he says that you make a victim install an application and then you can record microphone or access data on the phone. He obviously oversimplifies it. Not only will the victim have to install the application, but also have to give all the required permissions to the app (if you are dumb enough to do that, I guess you could make someone install root CA as well).

Connecting to a strange WiFi might get you into trouble, but it alone cannot cause this level of compromise .... isn’t it?

At the end of the day using public WIFI is similar to being in the same network as the attacker but that's about it. Don't be stupid, keep software updated and be informed about security. The story is overhyped. Same as the ads from VPN companies.

1
  • Comments are not for extended discussion; this conversation has been moved to chat.
    – Rory Alsop
    Commented Jan 26, 2020 at 14:48
40

All the attacks are possible and not over-hyped. In fact, these attacks are found in the wild. But you missed a detail that makes your conclusions incorrect.

There are some steps that the attacker and the victim take that are skipped in the video, but those skipped steps are specific to devices and those specific attacks. The premise and the underlying threats and possibilities are not affected by the missed steps. The video can remain valid for a much longer time by not delving into those missed technical steps.

Because these attacks actually happen in the wild, it's not fair to say that the missing parts make this an "over-simplified" video, just edited to be accessible by a wide audience (and not a technical audience).

The skipped steps for these specific attacks on the victim side:

  • ignore security warnings
  • install certificates (which the victim can be easily deceived into doing)
  • install the app supplied by the attacker and give it the permissions it asks for (the "free wifi" app shown in the video)

Installing the certificates allows the attacker to view TLS-protected sites (browsing, Google searches, accessing cloud storage)

The video clearly states that it's the installing of the app that gives the attacker full control over the device (microphone, etc.) The app could also be used to install the certificates. So, no, it's not just being a man-in-the-middle that does it, but by being in the middle, it is possible to serve the malicious app. This is the detail you appear to miss.

I suspect that the video is actually edited to be out of chronological order to show the simple attacks first, then get more serious. The situation on the attacker side could simply be:

  1. broadcast a free wifi network that the victim is likely to trust (airport wifi)
  2. show a webpage asking the victim to install an "authentication" app in order to gain access to the free wifi (for security reasons)
  3. ask for permissions which give the app full control, install root certificates
  4. grant the victim access to the Internet
  5. profit
15
  • 7
    "The video clearly states that it's the installing of the app that gives the attacker full control over the device (microphone, etc.) The app could also be used to install the certificates. So, no, it's not just being a man-in-the-middle that does it, but by being in the middle, it is possible to serve the malicious app. This is the detail you appear to miss". You're right. Thanks!. Commented Jan 19, 2020 at 13:58
  • 77
    The video absolutely is deceitful clickbait. It says in big scary capital letters: "when you connect to a hacker's rogue Wi-Fi, they can see every password you enter [and] email you send [and] document you access." That's BS. Connecting to their Wi-Fi is not the problem. Installing their app or their certificate is the problem. You act like the video is improved by the fact that it ignores or glosses over the things you actually need to avoid doing to avoid being hacked - when a responsible video would have talked only about those things.
    – benrg
    Commented Jan 20, 2020 at 7:37
  • 41
    @schroeder How does this video help such people in any way? Should they just... never connect to free wifi? The video tells them at the end to install "security solutions" to protect themselves (no examples are provided). How do you think that's going to go down? They'll probably search the Play Store for "security solution" and install whatever turns up, and give it all the permissions it requests. Installing dodgy apps is the real problem. Free wifi is a vector simply because free wifi is a thing that a lot of people want. "Security" apps are another vector. This video is not helping.
    – benrg
    Commented Jan 20, 2020 at 8:04
  • 9
    @benrg you act like "don't use free wifi" is some kind of new, controversial, or avant-garde advice. Are you unaware that that is the primary and most common advice by every authority to users and has been for years? And what's the alternative message for the video for non-technical audiences? "Here's an intentionally malicious network where the attacker will use every technical and psychological trick in the book (and some not in any book). But that's ok. You can use it safely as long as you do these steps ...."
    – schroeder
    Commented Jan 20, 2020 at 8:58
  • 7
    The thing one needs to be aware of is that you don't install some random app just to get access to a public WiFi, even if it's a "genuine" WiFi AP. Especially if the app requires a bunch of fishy unrelated permissions.
    – Jasper-M
    Commented Jan 21, 2020 at 10:37
18

Everything derives from installing the App. Normally phone Apps are signed and limited to designated App Stores. You would have to disable unknow source protection, and manually allow installation. After that, the phone is owned and MiTM isn't even a factor.

So yes I think it's overhyped.

3
  • 1
    it's entirely possible to get your "WiFi authentication" malware onto the app stores to give it an added layer of realism. some of the wifi authentication processes for popular free internet provides ( used in pubs, hotels etc) require using an application to connect. The university internet provider Eduroam also uses a similar app that takes control of your wifi hardware and certificates in order to set you up on campus wifi. Those app's are available on the app store so i'm sure a functionally identical application by a malicious developer could be published too.
    – J.Doe
    Commented Jan 20, 2020 at 15:17
  • 9
    @J.Doe - All that is true but the logic is circular. Fundamentally what you're showing is that if an App compromises your phone, your phone can be compromised. None of which has anything to do with MiTM. Commented Jan 20, 2020 at 15:42
  • 3
    @J.Doe: if I can convince you to install my malicious "free wifi" app or "wifi security" app, you don't need to be on my network for me to steal your traffic. The MITM part here is superfluous.
    – Lie Ryan
    Commented Jan 21, 2020 at 13:02
0

Just to complement the answers above.

A man in the middle attack is just forwarding the data it receives after looking at or modifying it. You can do some stuff with it, like changeing some values or denying communications. Online game hacking is basically this same stuff. For example, the device sends a jump command, you can intercept it and change it for a crouch command. Notice that we did not do anything to the device itself, we just modified the data the device sent.

Now the good stuff, to get access to the device you need to locally install a program to control it or you need to find one or multiple vulnerabilities that will let you run commands remotely. These vulnerabilities vary on device to device and are very hard to find in updated devices (zero days). Metasploit has a list of known vulnerabilities ready to use, but its more difficult than just spoofing the Wi-Fi traffic. So a man in the middle can mess with unencrypted data, but to mess with the device you need to exploit it. If you do exploit it and get root, then you pretty much own the device and can do anything!

-3

This is definitely not overhyped. It can be done. I did the same thing recently to my test phones.

The attacks are indeed possible with the right tools, and the victim wouldn't even necessarily know.

One of the phones, I was able to dump call logs, sms messages, and contacts. I also accessed the mic for recording and both front and rear cameras for snapshots. I also sent an MP3 file to her speakers of a dog growling. I'm sharing 2 screenshots of the metasploit session helpscreen showing a few commands available to me.

some of the available commands[![][1]]2

5
  • 10
    "and the victim wouldn't even necessarily know" - what did you install and what user interaction did it require? That seems to be the sticking point for people above .
    – schroeder
    Commented Jan 20, 2020 at 16:40
  • 1
    Ok- sorry. Yeah, on phone 1 I used Msfvenom to create an Android APK payload. I then copied that file to the phone's Download folder via USB. I clicked on the Newly minted malicious APK file and plowed through all the virus warnings. Once installed, I had to open a reverse TCP session with meterpreter to get the access I needed. That's where I got the 2 screenshots. So, yeah, there's a lot of interaction from the victim to the attacker with that method. Phone 2 was less interactve. I put a payload in a link and disguised it as a tinyurl. All I did was click the link and noticed nothing else. Commented Jan 21, 2020 at 3:53
  • 5
    Your phone 1 is nothing like the scenario in the question. Your phone 2 is, but you didn't go into details on that one. Can you expand on how your phone 2 test is similar to the video?
    – schroeder
    Commented Jan 21, 2020 at 7:10
  • 3
    Please provide details about this phone 2 and the link in question, because that sounds like nonsense to me.
    – Will
    Commented Jan 21, 2020 at 11:37
  • 1
    payload in a link - I assume this payload exploited a known or 0-day vulnerability in the phone's browser or OS to effectively do something similar and get a malicious app installed silently? Commented Jan 22, 2020 at 7:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .