cross-posted from: https://covert.nexus/post/27235
The FTC released a staff report in 2021 analyzing the privacy practices of six major U.S. Internet Service Providers. The report found that these ISPs collect as much, if not more, data on their customers’ browsing habits than popular advertisers like Google and Facebook. Additionally, some of these ISPs either operate their own advertising businesses or sell the data to third parties, such as the NSA.
Observations
- Many ISPs in Our Study Amass Large Pools of Sensitive Consumer Data.
- Several ISPs in Our Study Gather and Use Data in Ways Consumers Do Not Expect and Could Cause Them Harm.
- Although Many ISPs in Our Study Purport to Offer Consumers Choices, These Choices are Often Illusory.
- Many ISPs in Our Study Can be At Least As Privacy-Intrusive as Large Advertising Platforms.
Oh how lovely…
And this is why you never ever use ISP DNS, run DNS over HTTPS in the browser, and always use encrypted networking.
And use VPNs appropriate to the activity, when appropriate.
Oh, and never turn on ISP-supplied WiFi, as that gives them full access to the traffic from every device on your LAN, what physical hardware you own, and even where it is located in your home (and when it leaves and comes back to your home).
never turn on ISP-supplied WiFi
maybe I’m missing something here, how do you get access to the internet for all devices (mobiles, laptops, …) without wifi then?
You can get your own modem (what plugs into the wall) or your own wifi router (you’d plug this into the isp modem). Your own modem is better, but ISPs can give you grief about “supporting” them.
ah gotcha, you meant ISP-provided devices
A wireless/ethernet router as access point, a personal proxy server, or pihole, between your devices and theirs. Or, if possible your own modem and router.
[ISP modem/router]<–LAN–>[personal wifi router]<
wifi>[cheap pc proxy @ 192.168.x.x]<wifi>[all your devices]Proxy could be ssh(socks5), tor, shadowsocks (not microsocks), dnscrypt, tinyproxy, nekobox, whatever. They’d all have the same internal address from the proxy (if set up that way) and then again one address from the router to their device. (Router and proxy order could be reversed : or just router for some basic device identity privacy - it doesn’t encrypt your data though. An encrypted proxy will. And tor or a VPN will mask your external ip) Some proxies/VPN are more secure than others.
And,RTFM. A bad configuration can be worse than no configuration.
All those “hackers” in vpn commercials are in reality your isp.
Mainly because the VPN companies want to get that same data instead.
This information, although not new, sheds light on the misconception prevalent even amongst industry professionals today that ISPs only retain customer usage data related to IP address assignment.
However VPNs are exactly the same as ISPs, especially when it comes to actions forced by the government in the jurisdiction they are in.
you’re doing blanket statements. this highly depends on the provider
If you think your VPN provider is more immune to legal authorities than your ISP you are deluding yourself.
if you think that every VPN in the world handles legal situations the same way regardless of jurisdiction then you are a total nonce
Which is why good vpns are hosted in countries with extremely high privacy laws. And some can even be bought and used without giving any personal info. And why most vpns are RAM only and literally can’t log any records.
But you knew this before you spouted off, right?
Thanks for sharing! This is highly relevant to my master thesis, appreciate it 🌻
What is your master thesis on, if you don’t mind me asking?
All good! It’s about the use of free VPNs and how they may impact user privacy and security. But I do mention that VPNs is a one of the reasons as to why some people choose to use them in the first place. And this is a good source to have as it shows exactly the reasons as to why people flee to VPNs (be it paid or free).
Spoiler, in the majority of the cases free vpn’s are not good to use, but there isn’t too many documented articles on the topic, only some. So wanted to contribute on that field :)
That is really cool, and super interesting! Thanks for sharing!
Straight from the FTC? Very nice
Isn’t it great when the US’ FTC does something other than lick corporate boot?
Almost like presidential appointment powers matter! If only Democrats would have realized that before giving Trump 3 lifelong SCOTUS appointments.
Needs to be outlawed. Ridiculous
In 2017, Trump revoked regulations put in place by the Obama administration that would have compelled ISPs to obtain user consent before selling their browsing data.
Fuck trump
Maybe I’m just not getting it, but if we’ve mostly transitioned to HTTPS and encrypted DNS… what exactly can the ISP learn other than the address they serve and MAC of your gateway? Is this report for those who use their ISP’s DNS?
With very little effort it would be possible to mitm all the customers and it would all be pointless. Look at what Facebook is recently done to steal user data. They have apparently been doing their attack for years.
I’m going to need a source on both those claims to better understand how they can happen.
For an ISP to mitm, they’d need to sign and send the website certs themselves, and that’d show up in most browsers as a big red flag.
As far as Facebook goes, I was sure that’s just javascript and tracking cookies that they’re paying websites to use. No mitm there.Facebook internal documents from their current lawsuit discovery process. Facebook call this project ghostbuster:
https://www.documentcloud.org/documents/24520332-merged-fb
https://mashable.com/article/facebook-snapchat-data-project-ghostbusters-mark-zuckerberg
Mental Outlaw briefly outlines how the mitm attack works without alerting the browser of bad certs:
https://www.youtube.com/watch?v=WkLvpxImRGw&t=30
Your ISP doing a mitm attack would be multi-step and unlikely, but not impossible. The most likely use case would probably be the involvement of the federal government or bad actors who have compromised a CA, which has happened in the past:
https://en.m.wikipedia.org/wiki/Kazakhstan_man-in-the-middle_attack
https://blog.mozilla.org/security/2015/03/23/revoking-trust-in-one-cnnic-intermediate-certificate/
https://security.googleblog.com/2015/09/improved-digital-certificate-security.html?m=1
For a malicious ISP to try to intercept traffic on it’s own, I imagine an attack like this would be used:
https://techgenix.com/understanding-man-in-the-middle-attacks-arp-part4/
Here is an alternative Piped link(s):
https://www.piped.video/watch?v=WkLvpxImRGw&t=30
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I’m open-source; check me out at GitHub.
