DISCLAIMER: this post has been written by an AirVPN co-founder (Paolo) and merges the information and the points of view elaborated by the Air founders in more than seven years. Other Air VPN staff members might add additional comments in the future.
We have been asked via Twitter to reply to the following post:
We see that the issues raised by the aforementioned article may be of general interest, so we have decided to post a detailed rebuttal here, meant to fix the remarkable amount of technical misunderstandings and errors which have led the writer to astonishingly wrong conclusions and worrying generalizations.
The rebuttal is based on AirVPN only; we can not and we do not want to write in the name of any other service, since most of the considerations you will read here may or may not (and sometimes we know that they will not) apply to other "VPN services". Anyway, it is our right to reply as if the writer were talking about us too, because he/she repeatedly claims that ALL VPN services act in the same way.
Because a VPN in this sense is just a glorified proxy.
A "VPN in this sense" is NOT a proxy. Our service encrypts and tunnels all of the client system TCP and UDP traffic to and from the VPN server. Moreover, our service, when used with our free and open source software, also makes additional steps to prevent traffic leaks outside the VPN tunnel.
A proxy tunnels (and not necessarily encrypts) only TCP traffic (proxies can not support UDP), and only the traffic of those applications which are configured to connect to a proxy. UDP traffic, system traffic and traffic of applications which may be started by the system and that you failed to configure (or that you can't even configure in Windows, in some cases) are not necessarily tunneled to the proxy. Not even your system DNS queries are necessarily tunneled over the proxy.
The VPN provider can see all your traffic, and do with it what they want - including logging.
There is no way for you to verify that, and of course this is what a malicious VPN provider would claim as well. In short: the only safe assumption is that every VPN provider logs.
If we were really interested in logging our clients traffic, we would not allow connections to and from Tor, proxies and other VPNs. We have always made very clear how to bypass the problem of "trust us" when you can't really afford to do that, and our answer has always been "partition of trust". Please see for example our post dated March 2012 (!) about it:
There's more. We work under a legal framework where the safe harbors for the mere conduits are very rigidly and clearly defined (specifically, by the 2000/31/EC, the E-Commerce Directive, articles 12, 13, 14 and 15).
The liability exemption for the mere conduit status would not exist if we were not mere conduits. If we inspected traffic and/or modified traffic (e.g. through content injection) and/or selected source and destination of the communications, we would not be mere conduits and we would lose the legal protection on liability exemptions.
We have also two decisions of the Court of Justice of the European Union which clearly define indiscriminate data retention as infringing the fundamental rights of the citizens of the EU:
- under a legal point of view, logging and/or monitoring and/or inspecting and/or modifying the content of our customers traffic without the customers explicit and written consent would be a criminal infringement, also subject to civil prosecution by the customers themselves
- under a business point of view, that would be simply suicidal (more on this later)
And remember that it is in a VPN provider's best interest to log their users - it lets them deflect blame to the customer, if they ever were to get into legal trouble. The $10/month that you're paying for your VPN service doesn't even pay for the lawyer's coffee, so expect them to hand you over.
It is enigmatic how the writer can make such claims.
We charge less than 10 USD per month for our services and we can pay a whole legal firm, 250 servers (physical, bare metal servers), the whole staff, including a tiny team of programmers. We also regularly donate money to organizations and projects whose activities are compatible with AirVPN mission.
We're not here only for the money, but if the writer wants to talk about money, so be it. He/she may rest assured that we have planned seriously a business model which remains robust if not rock solid.
It is obvious that we must keep our business model solid, because our infrastructure has become large and we have duties toward the people working with us and toward our customers. At the same time we never forget that our customers have transformed into reality the dream to build a rather big project based on and aimed to privacy protection in a time when the whole world was going to the opposite direction. By changing now direction and pointing to a business based on privacy infringements and personal data commerce would not only betray our beliefs and mission and customers, but we would become a goldfish in an ocean of sharks, we could not even think to compete.
After 7 years, we have the right and knowledge to claim that a privacy protection mission is not incompatible with the price the writer mentions and with a strictly agnostic network where no traffic inspection or monitoring is enforced.
We can also claim confidently that any business plan based on data protection and privacy infringements not declared in the terms of service would crash dramatically in the short-term in the EU: remember the legal framework we live in and feel free to do your own research on real cases and incidents in the recent past.
Last but not least, please do your own math and compute the costs to store and "hand a customer traffic data over": they imply costs of losing the mere conduit status, added to the costs of civil lawsuits from that and potentially other tens of thousands customers. Then compare them to the "costs" (in reality benefits) of no monitoring at all added to the peace of mind to strictly act in a legal/lawful way.
Given all of the above, you can easily discern that the quoted assumption is false for AirVPN. The logical, unavoidable conclusion is that AirVPN best interest, even under a purely cynical, business point of view, is to NOT log (in the most extensive sense of the term) customers traffic and not commerce with their data.
But a provider would lose business if they did that!
I'll believe that when HideMyAss goes out of business. They gave up their users years ago, and this was widely publicized. The reality is that most of their customers will either not care or not even be aware of it
This is partially, only partially, true. HideMyAss was really risking to go out of serious privacy protection business soon after the incident occurred: check the massive uproar caused by the event. The AVG acquisition, with the disruptive marketing power of AVG, has probably covered the issue, but the old HideMyAss management hurried to sell the whole Privax company. Who knows, maybe just in time, maybe before the value could be hit too seriously by the incident. We can't know for sure, and the writer can't as well. Anyway, if the writer wants to claim that marketing is powerful, we agree (what a discovery!).
The logical jump from HMA incident to the assumption that every service does what HMA did is long. Do not forget that what HMA did would pose a huge amount of legal problems to us, as explained.
HideMyAss targeted the same persons who are happily using the new Facebook VPN. We respect the intelligence of our customers and we don't have the arrogance to think that we can change people mind and competence all over the world in a few years (or ever), and we don't even think that we can oppose the marketing power. More importantly, that's a problem pertaining to HideMyAss. It is not only unfair, but even defamatory to surreptitiously imply that the behavior (good or bad) of certain services is the same behavior of any other service, in the same field or not.
We have been providing AirVPN services since 2011, when we offered the service as a beta version totally free. Now we challenge the writer of the article to provide any single proof that any single user identity has been compromised by us through a betrayal of our terms of service and our mission and/or through traffic logging or inspection and/or by any infringement of the EU legal framework on privacy and personal data protection.
But I pay anonymously, using Bitcoin/PaysafeCard/Cash/drugs!
Doesn't matter. You're still connecting to their service from your own IP, and they can log that.
False. We provide our users with any tool to never make their "real" IP address appear to our servers. We have also integrated AirVPN over HTTP proxy, AirVPN over SOCKS proxy, and AirVPN over Tor usage in our free and open source software. We don't even block connections from competitor VPN servers. Finally, we accept not only Bitcoin, but Monero and ZCash as well, which are designed to provide a robust anonymity layer on the transactions.
If you really don't trust us, you can easily make your IP address never visible to our servers.
This is particularly important even if you trust us, but you can't afford (for the sensitivity of the data you need to transmit, for example) to assume that our servers are not monitored by hostile entities, an event that can happen with ANY service, not only VPN services. The fact that we have made every human effort to provide effective and easily usable protections against such occurrences is a proof of our interest in the protection of our customers privacy.
But I want more security!
VPNs don't provide security. They are just a glorified proxy.
This is ambiguous, because we would need the writer to define security scope and context exactly. Is he/she referring to integrity and security of data between your node and our servers? Or security of your system? Surely, our service is not meant as a security tool to protect against virus and spyware, and this is clearly stated at the very beginning of our Terms of Service. AirVPN can't do anything if your system is compromised.
However, the above does not imply in any way that our service is a glorified proxy. See the reasons we mentioned above and verify how a loose security mention does not change anything. Additionally, while OpenVPN is the core of our service, it is complemented by an important series of features aimed to protect privacy and data in all of those cases which OpenVPN alone has not been designed for.
Even if you don't run our free and open source software, we and our community have made any effort to provide guides and insights on how to get the most from our service to integrate it in a comprehensive environment aimed to protect your data and identity. We are very grateful to our community for the invaluable contributions throughout the years.
If we were a "malicious VPN provider", does the writer really think that we would have allowed our forums to become a golden source of information for privacy, identity and data protection? Do you really think that we would have been provided monetary support to TorProject, OpenBSD, European Digital Rights, Tor infrastructure, etc. etc.?
But I want more privacy!
VPNs don't provide privacy, with a few exceptions (detailed below). They are just a proxy. If somebody wants to tap your connection, they can still do so - they just have to do so at a different point (ie. when your traffic leaves the VPN server).
A part of this has been widely rebutted in our previous reply. Here it will be sufficient to add that even if you don't use end-to-end encryption, even if you don't use Tor on top of an AirVPN connection, a MITM who sniffs the packets in any point between the VPN server and the final destination (including the final destination itself of course) will see those packets coming from the VPN server exit-IP address, NOT from your real IP address and NOT from the entry-IP address of the VPN server you connect to. This is a paramount point which is incompetently (intentionally?) ignored by the writer. It is so important that in some extreme cases it makes the difference between imprisonment and freedom, or even between life and death.
Imagine the case of a whistleblower giving out relevant information via VoIP or other applications relying on UDP to a self proclaimed journalist who then betrays the confidentiality of the source, or even to a serious journalist who is unaware of the fact that his/her computer is compromised, or that his/her line is wiretapped. The whistleblower can't use a proxy reliably. The journalist, or the wiretapping entity, can trace the source IP address and the identity of the whistleblower can be disclosed (just to make a trivial example which does not require any wiretapping or compromised system, think of Skype exploit, for which any party could discover the IP address of the other party). In most of these cases, end-to-end encryption would have been irrelevant for the whistleblower.
Whenever the source can't trust the destination integrity, whether the recipient is in good faith or not, our service makes a vital difference.
But I want to confuse trackers by sharing an IP address!
Your IP address is a largely irrelevant metric in modern tracking systems. Marketers have gotten wise to these kind of tactics, and combined with increased adoption of CGNAT and an ever-increasing amount of devices per household, it just isn't a reliable data point anymore.
Marketers will almost always use some kind of other metric to identify and distinguish you. That can be anything from a useragent to a fingerprinting profile. A VPN cannot prevent this.
True. We have never said or written the contrary. In addition to changing IP address, which is anyway important in spite of the writer claims, further steps are strictly necessary to prevent profiling, from "separation of identities" to script blocking, from browser fingerprint changes to system settings obfuscation. Our community has widely covered this issue and provided precious suggestions.
Here the writer makes a totally irrational shift: first he/she wants to make you think that our service is just a "glorified proxy", then he/she wants to insinuate that our service is useless because it is not some sort of supernatural system capable to protect users from their own behavior and from every possible tracking system which exploits the user system, not the service.
So when should I use a VPN?
There are roughly two usecases where you might want to use a VPN:
You are on a known-hostile network (eg. a public airport WiFi access point, or an ISP that is known to use MITM), and you want to work around that.
You want to hide your IP from a very specific set of non-government-sanctioned adversaries - for example, circumventing a ban in a chatroom or preventing anti-piracy scareletters.
In the second case, you'd probably just want a regular proxy specifically for that traffic - sending all of your traffic over a VPN provider (like is the default with almost every VPN client) will still result in the provider being able to snoop on and mess with your traffic.
However, in practice, just don't use a VPN provider at all, even for these cases.
The first case is true, and it is very important.
However, it is totally false that you can safely rely on a proxy for the second case purpose. Many applications, including torrent software, can:
- bind to the physical network interface, or do some dangerous UPnP
- use UDP (not supported by a proxy)
- send DNS queries out of the proxy
- include the assigned "real" IP address inside their layer of communications, example: https://blog.torproject.org/bittorrent-over-tor-isnt-good-idea
In the aforementioned cases, correct usage of our service will fulfill the purpose to never disclose your real IP address and/or the UDP traffic and/or the DNS queries. A proxy will not and you can be potentially tracked back, either by copyright trolls or any hostile entity.
Additionally, our service has many more use cases:
- tunneling UDP traffic (not available with a proxy or Tor)
- circumventing censorship based on IP addresses block
- circumventing censorship based on DNS poisoning
- preventing injection of forged packets (not necessarily available with a proxy even in TCP, and surely not when you need UDP flow integrity)
- using Tor anyway when Tor usage is blocked or triggers interest of ISP or any hostile entity about you
- protecting your identity when the final recipient of your communications is compromised (not available with end-to-end encryption alone, and not available with Tor when you need UDP, imagine if you need to stream a video in real time which requires source identity protection)
- making your services (web sites, torrent clients, FTP servers for example) reachable from the Internet when your ISP does not allow port forwarding (not available with a proxy), without exposing your IP address
- having a static exit-IP address
- bypassing various types of traffic shaping
- tunneling simultaneously the traffic of all the devices in your local network, even with remote port forwarding, and even those which can't run OpenVPN provided that you have a device acting as a gateway to the VPN (typical examples a pfSense box or a DD-WRT / AsusWRT / Merlin / Tomato etc. router or any computer configured to work as a router)
and maybe you can see more use cases which we have missed here.
The fact that the writer omitted all of the above says a lot about his/her competence and/or good faith.
So, then... what?
If you absolutely need a VPN, and you understand what its limitations are, purchase a VPS and set up your own. I will not recommend any specific providers (diversity is good!), but there are plenty of cheap ones to be found on LowEndBox.
This is hilarious, and not only because the whole point of the writer's post ends up into advertising LowEndBox.
We will not insult our readers' intelligence with an explanation of why that is a terrible idea when you seek more privacy and some anonymity layer in your interactions with the Internet.
Draw your own conclusions.
Kind regards and datalove