text stringlengths 0 1.36k |
|---|
[50.74 --> 51.58] I'm Alex Sexton. |
[52.14 --> 53.34] And I'm Rachel White. |
[53.72 --> 56.18] All right, everybody. Let's get this party started. |
[56.18 --> 57.76] Let's just dive right into the first topic. |
[58.22 --> 59.42] So Google broke the Internet. |
[60.62 --> 67.48] I don't know why they keep pointing out flaws in the Internet security, but they broke the Internet again. |
[69.56 --> 75.68] I thought they helped disclose it, but wasn't it like some German W, some acronym? |
[76.32 --> 77.22] I think it was the Germans. |
[77.70 --> 78.08] That's all I'm saying. |
[78.88 --> 79.80] It was the Germans. |
[80.24 --> 80.40] Yeah. |
[81.58 --> 82.32] Likely story. |
[82.32 --> 83.78] Okay. No, anyway. |
[83.92 --> 87.04] So SHA-1 hashing algorithm has been cracked. |
[88.30 --> 94.22] I guess like in 2005, there was a paper written that said theoretically it could be cracked, but nobody had done it yet. |
[95.42 --> 101.66] And apparently as of like 2010, the federal government said no government encryption can use any SHA-1 algorithms, |
[101.76 --> 106.26] which is a pretty good indication that foreign governments have been able to crack this for a while. |
[106.26 --> 107.06] Yeah. |
[107.06 --> 107.12] Yeah. |
[107.52 --> 118.64] The only person I've seen strongly support SHA-1 for the last six or seven years is Linus Torvalds and Git. |
[119.30 --> 120.16] It's so annoying. |
[120.60 --> 122.24] And not just kind of. |
[122.34 --> 126.32] He really was like, you guys are all super dumb for caring about this. |
[126.84 --> 127.40] I know. |
[127.70 --> 128.58] It's really crazy. |
[128.84 --> 130.28] He's still downplaying it, actually. |
[130.52 --> 130.72] Yeah. |
[130.72 --> 136.72] So backing up a little bit, let's just kind of get into like what is SHA-1 and what does it do? |
[137.22 --> 139.70] So does anybody else want to take a crack at this or do you want me to explain it? |
[140.26 --> 143.94] I only know it from like Git related stuff. |
[144.44 --> 145.82] So that's all. |
[145.82 --> 146.22] Right, right. |
[146.42 --> 149.00] Well, that's actually a really good way to explain it though, right? |
[149.14 --> 153.60] So the way that Git uses SHA-1 is kind of indicative of how everybody uses it, |
[153.60 --> 158.44] which is that you take a bunch of data and you say, oh, I want a unique identifier for this data. |
[158.44 --> 159.48] So you hash it, right? |
[159.98 --> 163.86] And that's what Git does to every change that comes into the Git tree. |
[163.96 --> 166.98] It gets this hash of the data and it uses that as the identifier. |
[167.32 --> 171.58] So if you like go to GitHub and you go to a project and then you click on commits |
[171.58 --> 174.02] and then you click on one of those commit links, in the URL bar, |
[174.10 --> 176.78] you'll see like this randomly kind of generated identifier. |
[177.12 --> 180.06] And that is a unique identifier for that hash. |
[181.08 --> 184.84] The problem is that if you could forge these, if you could, you know, |
[184.84 --> 188.38] like that's a very small amount of data representing a large amount of data. |
[188.56 --> 190.92] So theoretically, if you can reverse engineer the algorithm, |
[191.00 --> 195.66] you could come up with a different set of data that would also hash to that same thing. |
[196.64 --> 200.58] And people have been theoretically able to do this for a while and now they really can. |
[201.96 --> 204.28] It still costs like $100,000. |
[206.82 --> 207.22] Yeah. |
[207.22 --> 212.64] It'll get cheaper, but right now, like with the current algorithm, the current break, |
[212.68 --> 216.26] it's insane how much faster they can do it. |
[216.34 --> 225.26] But still, like with AWS, like spot instances, it costs around $100,000 to just like break a random thing. |
[226.10 --> 229.28] But how much do I have to pay like a Russian hackers that have a botnet? |
[229.86 --> 230.44] Oh, sure. |
[230.44 --> 235.52] Yeah, I mean, it's just like one Bitcoin, which is roughly $20,000. |
[235.86 --> 236.18] I wouldn't necessarily be worried about this. |
[237.48 --> 238.56] Sorry, I didn't hear that. |
[238.94 --> 244.68] I was saying I probably need to pay one Bitcoin, which is roughly $20,000 to get Russian hackers to break it. |
[244.72 --> 245.44] Oh, yeah, for sure. |
[246.58 --> 253.96] Yeah, the cost is still prohibitive to the point where no one's going to like troll you with this. |
[253.96 --> 261.54] Someone really needs to want, like there has to be a reason someone's doing this at this point. |
[261.66 --> 264.40] But that will only be true for like two months or something, right? |
[264.44 --> 270.20] Like people will make this better instantly and then exploit everybody across the board. |
[271.10 --> 271.40] Right. |
[271.54 --> 274.76] And it's pretty much a given now that governments can do this kind of at will. |
[275.30 --> 275.62] Oh, yeah. |
[275.62 --> 288.90] So what that means is that if your integrity checks involve you hashing with this algorithm, then now if you're just using those checks, people can just inject malware, just, you know, whatever they want. |
[289.66 --> 289.74] Right. |
[289.84 --> 291.16] So I have a question. |
[291.16 --> 305.26] If this has been relatively like not super secure for a while, what was the catalyst for people to be like, OK, it's finally time to stop using this thing? |
[305.26 --> 307.52] Was it was it something that Google did that you said? |
[308.28 --> 308.52] Oh, yeah. |
[308.68 --> 309.02] Yesterday. |
[309.90 --> 310.22] Yeah. |
[310.46 --> 310.68] Yeah. |
[310.82 --> 317.82] Well, honestly, I think most people in the security community have felt like since 2005 that you should stop using this. |
[317.90 --> 320.86] There are other algorithms that are just as good that don't have this problem. |
[321.08 --> 325.26] And in 2010, I think most reasonable companies said, hey, we should stop using this. |
[325.26 --> 328.42] Like I said. |
[329.42 --> 336.08] Browsers already like don't allow like you'll get a very big red X instead of a green lock. |
[336.88 --> 341.10] If, you know, the shot one is used for for web security stuff. |
[341.50 --> 348.08] It's been well known to be very crackable by someone with a ton of money for a long time. |
[348.08 --> 348.84] Yeah. |
[349.02 --> 353.68] But but like Alex said, Linus Torvald has just remained unimpressed by evidence. |
[354.16 --> 358.72] And so it is still in heavy use in Git and GitHub and a bunch. |
[358.78 --> 363.36] I mean, that's that's fine because I'm wholly unimpressed by him. |
[363.64 --> 363.82] So. |
[366.50 --> 366.98] Yeah. |
[367.46 --> 369.30] It's you're only going to make him stronger. |
[369.30 --> 369.38] Sure. |
[371.08 --> 371.66] But yeah. |
[371.76 --> 382.22] So to answer your question, the thing that happened yesterday was that some people from Google and the Germans came out and just said, hey, look, we cracked it. |
[382.30 --> 383.68] Like, here's exactly how we cracked it. |
[383.68 --> 389.12] So it went from theoretical to, you know, here is, you know, an open version of this. |
[389.66 --> 395.42] To be to be totally clear, though, it's it's still like they have to try a ton of things. |
[395.42 --> 403.54] It's it's like they were able to reduce the subset that you had to brute force to a small enough amount to be significant. |
[403.54 --> 409.84] But it still takes like one hundred and ten years of computing time or something like that. |
[409.84 --> 412.44] Like you had to you had to put a lot of machines into it. |
[412.68 --> 417.46] But that number will slowly turn down to, you know, seconds or whatever, I'm sure. |
[418.56 --> 418.76] Yeah. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.