Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Very Good Security

a16z.com

Post reply on HN
Print view View original

41–50 of 94 posts

Re: Very Good Security

#41

Post by drexlspivey »

fefe23 wrote: post #27
mahmoudimus wrote: post #17

Earlier quoted context omitted.

(disclaimer: I work for VGS) Hi robert204, Your question has two specific parts that I want to address: 1) Single Point of Failure 2) Larger target for malicious actors Regarding point #1: - We have invested significant amount of resources in making our product as stateless as possible and our core product can live on different cloud providers' edge networks. - We conduct failover tests every 2 weeks to ensure we hav…

You had me at "cloud providers". If you store the data on some cloud provider, then you are just as bad as what your prospective customers are doing. I don't want any of my sensitive data stored on "some cloud provider". Also, your security strategy apparently boils down to "we'll be REAL CAREFUL, pinky swear!" That strategy does not work, and has never worked before. The whole reason why you think your product is ne…

> Also, your security strategy apparently boils down to "we'll be REAL CAREFUL, pinky swear!"

Case in point: T Mobile Austria. https://nordvpn.com/blog/t-mobile-passwords-hacked-twitter/

Re: Very Good Security

#42

Post by JohnJamesRambo »

throwawaymath wrote: post #16

I can see why this is an attractive idea to fund, but in my opinion it's the wrong way to resolve the problems highlighted in the article. This is not a technical problem, it's a usability problem. We have had the cryptography necessary to technically fix this for a long time. Replace the single human-memorable token (SSN) with a unique public/private key pair. Then you provide safe authentication by signing verifica…

https://www.civic.com/

What do you think of Vinny Lingham's company that is aiming to do something similar?

Re: Very Good Security

#43

Post by drexlspivey »

fosco wrote: post #12
delinka wrote: post #10

Why don't we already have apps on our smartphones for this? - $PROVIDER wants the following data: $LIST_OF_OPTIONAL_AND_REQUIRED_ITEMS - You select which you can provide - If the data to be provided includes "billing identifier" or "credit file identifier" (and especially if the identifier is, say, SSN), then first your app obtains a new identifier from the reporting agency or your insurance carrier, and *that* numbe…

what happens if you do not trust your smartphone? I do not like storing anything of value on mine. not that it matters but I recently switched to LineageOS and my opinion has not changed.

Apple does this with apple pay and the per device identifier is stored in the secure enclave (seperate chip that can create keys and sign data). The identifier can never leave the SE, requires Touch/Face ID to unlock and can be wiped remotely in case you lose your phone.

Re: Very Good Security

#44

Post by nodesocket »

wepple wrote: post #38
fefe23 wrote: post #27

Earlier quoted context omitted.

You had me at "cloud providers". If you store the data on some cloud provider, then you are just as bad as what your prospective customers are doing. I don't want any of my sensitive data stored on "some cloud provider". Also, your security strategy apparently boils down to "we'll be REAL CAREFUL, pinky swear!" That strategy does not work, and has never worked before. The whole reason why you think your product is ne…

I’m curious, what’s wrong with storing this type of data in a cloud provider? Also, security aside wouldn’t a16z have invested because the business isn’t “do it more securely”, but “outsource PCI compliance entirely”?

I can think of a few:

  1.) Shared resources, and exploits like meltdown and spectre 
  2.) Cloud provider employees potentially have access
  3.) Cloud providers are subject to law enforcement requests
However, there are benefits. Typically Google Cloud for example is going to have much better systems and security that a home-rolled data center setup. They've been working on it for years and essentially have unlimited resources (time and capital).

Re: Very Good Security

#45

Post by mdpopescu »

lisper wrote: post #31
mdpopescu wrote: post #13

I find it interesting that Stefan Brands [1] solved the zero-knowledge authentication problem a couple of decades ago and his tools are still not widely applied. Given my bias against imaginary property, I believe that's because his patents on them are still valid -- and apparently owned by Microsoft at the moment [2]. [1] https://en.wikipedia.org/wiki/Stefan_Brands [2] http://financialcryptography.com/mt/archives/00…

> Stefan Brands [1] solved the zero-knowledge authentication problem a couple of decades ago Reference? If you want people to take you seriously when you claim that someone's solution to a problem has been overlooked you have to provide a link to the (alleged) solution, not just to the author's biography.

Er... it was on the Wikipedia page. It was published on Amazon [1] but the Wikipedia page also seems to have a link to a PDF (ref #5).

[1] https://smile.amazon.com/Rethinking-Public-Infrastructures-D...

Re: Very Good Security

#46

Post by tptacek »

This doesn't say anything. They invested in a tokenization company. That's not a new or interesting technology. What am I missing?

There are interesting data security companies happening right now. For instance, Matthew Green is doing Zeutro, an ABE company. Think of ABE as Shamir's Secret Sharing on Steroids: you can encrypt data and delegate it out to different people based on boolean expressions. That at least addresses a fundamental problem in data center encryption (the fact that serverside data encryption is "all or none" with respect to applications).

This, though? I assume the announcement means VGS is doing great in the market. Congratulations, I guess?

Re: Very Good Security

#47

Post by dvdhsu »

vgs_interviewee wrote: post #18

Anonymous account, because of reasons. I interviewed and was offered a job at this company. I turned it down because they had some of the most morally bankrupt leadership I have ever seen in a startup. Frankly, it made me less likely to interview with YC companies at all. Just a quick list of giant red flags- 1. They are violating visa laws by having their employees in the Ukraine lie on their applications and say th…

I don't think they're YC: http://www.ycombinator.com/companies/

Re: Very Good Security

#48

Post by mdpopescu »

thinkingkong wrote: post #20
mdpopescu wrote: post #13

I find it interesting that Stefan Brands [1] solved the zero-knowledge authentication problem a couple of decades ago and his tools are still not widely applied. Given my bias against imaginary property, I believe that's because his patents on them are still valid -- and apparently owned by Microsoft at the moment [2]. [1] https://en.wikipedia.org/wiki/Stefan_Brands [2] http://financialcryptography.com/mt/archives/00…

I've never heard of this before. Do you know which technologies specifically address the article posted?

Chapter 1 explains the usage of the protocols. Quoting from the book, on page 33:

"The authentication mechanism is such that the receiver not only authenticates the message, but also demonstrates a property of the attributes encoded into its certified key pair. The receiver has full control over which property is demonstrated: it can be any satisfiable proposition from proposition logic, where the atomic propositions are relations that are linear in the encoded attributes. Any other information about the attributes remains unconditionally hidden." (Emphasis mine.)

I will admit that the math was not something I studied in depth (and I definitely didn't check the proofs!), since crypto is at best a hobby for me and not my main job.

Re: Very Good Security

#49

Post by CiPHPerCoder »

mahmoudimus wrote: post #21
CiPHPerCoder wrote: post #4

H(ssn) just kicks the problem downstream. - If H is a simple cryptographic hash function, it's not resistant to brute-force attacks to recover the SSN - It's not revokable What we need is something more akin to a Credit Card number. Something like an abstraction layer. It might even be implementable as a UUID. If you need to revoke it, you can do so since it's not cryptographically tied to anything. Failing that, a b…

(disclaimer: I work at VGS) We offer a variety of various format preserving aliasing algorithms. Only legacy systems tend to choose the SSNs if they have fixed-width columns in their RDBMS that are difficult to change (imagine petabytes of data). The idea behind format preserving aliases is actually based on the NIST SP 800-3G standard[1]. We use FF1 and are actively engaging with the world's leading cryptographers s…

I'd like to make but one appeal to everyone reading this thread:

Ask your cryptographer if the algorithm they're proposing you to use is IND-CCA2 secure (especially if it meets the criteria for IND-CCA3).

Litmus test: If they don't know what that means, you shouldn't be trusting them for cryptography advice.

If it isn't IND-CCA2 secure, you shouldn't be using it. Full stop.

For the curious: https://tonyarcieri.com/all-the-crypto-code-youve-ever-writt...

The IND in IND-CCA2 means "INDistinguishable"; i.e. from randomly generated line noise. For symmetric cryptography, your ciphertext shouldn't have any structure to it. (Lattices and such are a different story. If structure is permissible for your security goals, you're probably doing asymmetric cryptography anyway.)

To be clear: Format-preserving, order-preserving, order-revealing, and homomorphic encryption technology-- while an exciting research area-- fails to meet this requirement and should not be used for non-experimental purposes until their techniques have had time to mature. And even then, until they meet this requirement, only when the threat model doesn't realistically include the possibility of adaptive chosen-ciphertext attacks. (Spoiler: A real world threat model will almost certainly always include that.)

> We use FF1 and are actively engaging with the world's leading cryptographers

I've seen this "we engage with the world's leading cryptographers" genre of claim before, albeit from a much more arrogant source: https://news.ycombinator.com/item?id=6916860

Re: Very Good Security

#50

Post by spullara »

vgs_interviewee wrote: post #18

Anonymous account, because of reasons. I interviewed and was offered a job at this company. I turned it down because they had some of the most morally bankrupt leadership I have ever seen in a startup. Frankly, it made me less likely to interview with YC companies at all. Just a quick list of giant red flags- 1. They are violating visa laws by having their employees in the Ukraine lie on their applications and say th…

You should put this on Glassdoor.
Post reply on HN