WEBVTT 00:00:11.129 --> 00:00:13.269 Hello, and welcome to another episode of the 00:00:13.269 --> 00:00:15.609 Everyday Defender podcast. We're back again. 00:00:15.669 --> 00:00:17.649 It's been a month. I can't believe the time has 00:00:17.649 --> 00:00:19.670 gone by so quickly, but here we are back again. 00:00:19.789 --> 00:00:21.870 And I'm joined as always by my friend from the 00:00:21.870 --> 00:00:24.329 Netherlands. How are you doing, Koos? Hey, Chris. 00:00:24.530 --> 00:00:27.449 How are you doing? Yeah, it's August already. 00:00:28.030 --> 00:00:30.809 Summer has almost passed here in the Netherlands. 00:00:31.529 --> 00:00:34.530 I do still have my holiday in front of me. I'm 00:00:34.530 --> 00:00:37.070 leaving in October, actually. So quite a late 00:00:37.070 --> 00:00:41.630 holiday for the European listeners. But yeah, 00:00:41.750 --> 00:00:44.600 nice to see you again. Yeah, we're actually getting 00:00:44.600 --> 00:00:46.759 some better weather too. It's warming up for 00:00:46.759 --> 00:00:49.079 us. So spring is here, which also means lots 00:00:49.079 --> 00:00:51.500 of farm chores and, you know, all of that type 00:00:51.500 --> 00:00:55.000 of stuff. But hey, not complaining with the nice, 00:00:55.119 --> 00:00:56.560 well, it's raining today, but not complaining 00:00:56.560 --> 00:00:58.579 with the nice weather in general. I saw you and 00:00:58.579 --> 00:01:02.039 your wife washing the donkey the other day outside. 00:01:02.659 --> 00:01:05.799 Yes, washing a donkey is not the easiest of things 00:01:05.799 --> 00:01:10.459 to do. I saw that. That's funny. Yes, it was. 00:01:10.700 --> 00:01:13.140 Her Instagram is full of interesting things. 00:01:14.859 --> 00:01:17.500 But not talking about donkeys or llamas or anything 00:01:17.500 --> 00:01:20.519 like that today. Yeah, really interested, Kosta, 00:01:20.519 --> 00:01:23.519 to hear about Project Perception today. This 00:01:23.519 --> 00:01:25.719 is going to be one I'm super excited to hear 00:01:25.719 --> 00:01:28.180 your thoughts on that and sort of sit back and 00:01:28.180 --> 00:01:30.879 just have you go through all of that type of 00:01:30.879 --> 00:01:35.000 stuff. I wanted to talk just a little bit about 00:01:35.000 --> 00:01:38.980 some changes to authentication methods in Entra. 00:01:39.379 --> 00:01:41.719 You know, it's a very small announcement that 00:01:41.719 --> 00:01:44.060 happened, you know, last month or I think it 00:01:44.060 --> 00:01:47.840 was last month. But it is, you know, the potential 00:01:47.840 --> 00:01:50.040 to impact some folks. And I've also heard some 00:01:50.040 --> 00:01:53.299 confusion. I sort of wanted to jump into that, 00:01:53.400 --> 00:01:57.420 and then I'll have you tackle the big topic, 00:01:57.579 --> 00:01:59.620 right? And then, you know, as to close out the 00:01:59.620 --> 00:02:02.519 show again, I promised on the last episode that 00:02:02.519 --> 00:02:05.599 I'd talk about Connect 365, so I have something 00:02:05.599 --> 00:02:08.639 to share about that as well as we close out later. 00:02:08.740 --> 00:02:11.560 Well, don't call SMS MFA a small topic because 00:02:11.560 --> 00:02:14.080 I think it has some large impact. I don't know 00:02:14.080 --> 00:02:17.780 the details myself, so let us hear what you have 00:02:17.780 --> 00:02:19.439 to say on that, but I think people are still 00:02:19.439 --> 00:02:22.639 using it a lot. lot on scale and that's that's 00:02:22.639 --> 00:02:25.939 true right so so um essentially for for folks 00:02:25.939 --> 00:02:28.780 who are not aware who may have missed it um sort 00:02:28.780 --> 00:02:32.099 of somewhere mid july around the 13th of july 00:02:32.099 --> 00:02:35.759 microsoft had sort of announced that um it was 00:02:35.759 --> 00:02:39.379 going to deprecate SMS and voice authentication 00:02:39.379 --> 00:02:43.439 methods, right? So if you possibly still have 00:02:43.439 --> 00:02:45.699 MFA enabled for folks, but they're using, I guess, 00:02:45.719 --> 00:02:48.199 legacy methods and you receive maybe a text message 00:02:48.199 --> 00:02:51.699 or a voice phone call, which used to be an option 00:02:51.699 --> 00:02:55.560 or still is an option, those methods are now 00:02:55.560 --> 00:02:57.979 sort of being phased out and deprecated, right? 00:02:58.080 --> 00:03:00.400 And we've talked to it. I mean, we're going to 00:03:00.400 --> 00:03:02.180 have to go back to like season one on this show 00:03:02.180 --> 00:03:05.620 where we talked about. multi -factor authentication 00:03:05.620 --> 00:03:07.919 and conditional access. And we talked about, 00:03:07.919 --> 00:03:10.400 you know, why SMS isn't the greatest anymore 00:03:10.400 --> 00:03:13.439 anyway, right? So I think everyone understands 00:03:13.439 --> 00:03:16.240 that text messaging and voice calls and stuff 00:03:16.240 --> 00:03:19.680 are easily spoofable and just are not good mechanisms 00:03:19.680 --> 00:03:24.280 or methods for MFA. But there are still organizations 00:03:24.280 --> 00:03:28.819 that use it, right? And so... I guess my concern, 00:03:28.919 --> 00:03:31.500 and I actually just went through this process 00:03:31.500 --> 00:03:36.139 with a client, and actually today was the big 00:03:36.139 --> 00:03:39.360 day that they made all the changes within their 00:03:39.360 --> 00:03:41.919 environment to actually catch all of the last 00:03:41.919 --> 00:03:44.759 users that were still using SMS and stuff. So 00:03:44.759 --> 00:03:47.800 literally, this is very new for me. And incidentally, 00:03:47.800 --> 00:03:50.800 they had made the decision to move away from 00:03:50.800 --> 00:03:53.969 these methods. before Microsoft announced it. 00:03:54.009 --> 00:03:55.430 So they've been planning it now for two or three 00:03:55.430 --> 00:03:58.090 months that they've been sort of working on moving 00:03:58.090 --> 00:04:02.650 away from text messaging and voice. It just so 00:04:02.650 --> 00:04:04.469 happens that, you know, whilst this was going 00:04:04.469 --> 00:04:06.129 on, Microsoft also announced that they were going 00:04:06.129 --> 00:04:08.729 to sort of drop support. And so there's actually 00:04:08.729 --> 00:04:10.780 quite a lot. involved in this right and so i 00:04:10.780 --> 00:04:12.580 thought you know what this is really topical 00:04:12.580 --> 00:04:14.639 i originally i was going to talk a little bit 00:04:14.639 --> 00:04:16.560 more about vulnerability management but i thought 00:04:16.560 --> 00:04:18.819 let me bring this one um sort of forward and 00:04:18.819 --> 00:04:20.860 talk about this because this is happening right 00:04:20.860 --> 00:04:24.339 now so um as i said microsoft announced this 00:04:24.339 --> 00:04:28.759 in july immediately i've sort of seen um picked 00:04:28.759 --> 00:04:31.240 up on some confusion uh sort of out in the industry 00:04:31.240 --> 00:04:33.620 and i've also seen some interesting commentary 00:04:33.620 --> 00:04:37.899 right through blogs and online about this and 00:04:38.329 --> 00:04:40.029 I guess the first thing to sort of set straight 00:04:40.029 --> 00:04:44.290 is sort of what actually is happening. So SMS 00:04:44.290 --> 00:04:48.430 MFA is not being banned in Entra. Like that's 00:04:48.430 --> 00:04:50.250 one of the headlines that I saw somewhere is 00:04:50.250 --> 00:04:53.189 like SMS MFA is being banned. Like that's not 00:04:53.189 --> 00:04:54.709 actually what's happening. What Microsoft is 00:04:54.709 --> 00:04:57.709 doing is they're retiring the Microsoft provided 00:04:57.709 --> 00:05:01.089 text messaging and voice sort of capability. 00:05:01.410 --> 00:05:04.310 So, you know, up until now, those text messages 00:05:04.310 --> 00:05:06.879 that were sent. voice calls that happen, it's 00:05:06.879 --> 00:05:09.240 all Microsoft's sort of telephony stuff, right? 00:05:09.720 --> 00:05:11.600 They're essentially just phasing that out and 00:05:11.600 --> 00:05:14.160 you will no longer be able to do that in intro. 00:05:14.259 --> 00:05:17.660 There will not be options for you anymore. I 00:05:17.660 --> 00:05:19.839 know that there are still organizations that 00:05:19.839 --> 00:05:25.620 use text messaging, right? In my mind, I struggle 00:05:25.620 --> 00:05:29.000 to come up with scenarios and use cases anymore 00:05:29.000 --> 00:05:32.199 where text messaging is the only option, right? 00:05:32.240 --> 00:05:36.779 I feel like a YubiKey. can replace any you know 00:05:36.779 --> 00:05:39.399 any text messaging sort of solution right if 00:05:39.399 --> 00:05:41.639 you've got offline type scenarios or no internet 00:05:41.639 --> 00:05:44.660 type scenarios where maybe you've got a cell 00:05:44.660 --> 00:05:47.399 service but you don't have the internet those 00:05:47.399 --> 00:05:49.339 are some of the scenarios historically but i 00:05:49.339 --> 00:05:51.920 feel like ubiquis can sort of cater to that now 00:05:51.920 --> 00:05:55.160 um i know that you know in the past i've had 00:05:55.160 --> 00:06:00.139 clients where they have um minors or young people 00:06:00.139 --> 00:06:02.240 logging into the environment if you think about 00:06:02.240 --> 00:06:04.699 schools and things like that they can't always 00:06:04.699 --> 00:06:08.079 rely on those you know younger generation um 00:06:08.079 --> 00:06:11.779 or minors having smartphones right and so text 00:06:11.779 --> 00:06:14.019 messaging was a was sort of an answer or a solution 00:06:14.019 --> 00:06:16.779 to giving them mfa and i and i see that i think 00:06:16.779 --> 00:06:19.100 there are other solutions for sort of that cater 00:06:19.100 --> 00:06:21.959 to the education space and the sort of kids or 00:06:21.959 --> 00:06:24.759 children's space these days so i in my mind i 00:06:24.759 --> 00:06:28.220 struggle to to come up with like real proper 00:06:28.220 --> 00:06:31.779 scenarios that still require text messaging where 00:06:31.779 --> 00:06:35.720 i hear heard it a lot uh years ago where employees 00:06:35.720 --> 00:06:38.339 were complaining that when companies started 00:06:38.339 --> 00:06:42.060 uh embracing mfa in general and they were forcing 00:06:42.060 --> 00:06:45.839 that's how how employees sometimes felt and forcing 00:06:46.649 --> 00:06:49.589 an app on their private phones, right? So they 00:06:49.589 --> 00:06:51.730 were like, this is my private phone. If the company 00:06:51.730 --> 00:06:53.769 wants me to do something, then they hand me out 00:06:53.769 --> 00:06:57.550 a business phone and I'll use that for it. But 00:06:57.550 --> 00:07:00.850 like you said, with physical YubiKeys, MFA pass 00:07:00.850 --> 00:07:03.610 keys, physical pass keys, I don't think that 00:07:03.610 --> 00:07:06.850 argument really stands anymore. And I think that's 00:07:06.850 --> 00:07:08.750 a very good one because I think that is a very 00:07:08.750 --> 00:07:11.550 valid argument that still happens possibly in 00:07:11.550 --> 00:07:15.930 Europe. I think your laws around ability to disconnect, 00:07:16.009 --> 00:07:20.290 and using personal devices and things for work 00:07:20.290 --> 00:07:22.670 functions is a lot more strict, I think, than 00:07:22.670 --> 00:07:25.029 other parts of the world. I think California 00:07:25.029 --> 00:07:27.310 also has some very interesting laws that may 00:07:27.310 --> 00:07:31.490 sort of, you know, sort of come into play here. 00:07:31.670 --> 00:07:33.750 But you're right, like the YubiKey answer just 00:07:33.750 --> 00:07:36.050 solves that problem, right? And it's a lot cheaper 00:07:36.050 --> 00:07:40.310 to give someone a YubiKey that costs $100 than 00:07:40.310 --> 00:07:42.769 to give them an iPhone that costs $1 ,000, right? 00:07:43.160 --> 00:07:46.199 It's not even $100 if you buy them in bulk and 00:07:46.199 --> 00:07:49.620 you go for the less expensive non -biometric 00:07:49.620 --> 00:07:51.980 versions. It's only a couple of, I don't know, 00:07:52.060 --> 00:07:55.439 $20 or something. Yeah, no, absolutely. So look, 00:07:55.620 --> 00:07:59.519 I think we sort of understand and agree that's 00:07:59.519 --> 00:08:02.660 probably very few sort of real scenarios where 00:08:02.660 --> 00:08:06.379 text messaging is still really needed. But Microsoft 00:08:06.379 --> 00:08:10.399 are not removing... you know, saying you can't 00:08:10.399 --> 00:08:11.920 use it. They're just saying they're no longer 00:08:11.920 --> 00:08:14.959 going to provide that service, right? So essentially, 00:08:15.139 --> 00:08:18.680 if you do need that capability, Microsoft are 00:08:18.680 --> 00:08:21.819 going to have partners, sort of telco partners 00:08:21.819 --> 00:08:25.920 that will provide that, right? Now, I went and 00:08:25.920 --> 00:08:27.660 had a look today to see if I could track down. 00:08:28.170 --> 00:08:30.990 a list of these partners and i cannot so that 00:08:30.990 --> 00:08:34.470 list is not available online yet um i believe 00:08:34.470 --> 00:08:36.610 it's only being really the list itself is only 00:08:36.610 --> 00:08:39.909 being released on the 18th of september i would 00:08:39.909 --> 00:08:43.110 expect we would see folks like twilio and and 00:08:43.110 --> 00:08:45.549 you know those types of telco as a service you 00:08:45.549 --> 00:08:48.850 know api as a service type vendors um on that 00:08:48.850 --> 00:08:52.940 list so the takeaway is Microsoft's going to 00:08:52.940 --> 00:08:55.820 stop providing the service, SMS and voice services, 00:08:55.879 --> 00:08:58.500 but you could, if you still desperately need 00:08:58.500 --> 00:09:04.000 it, you can. come up with ways to engage third 00:09:04.000 --> 00:09:06.100 parties. And ultimately, you're going to be paying 00:09:06.100 --> 00:09:08.100 for those text messages yourself now, right? 00:09:08.200 --> 00:09:10.500 So it's not going to be Microsoft funding or 00:09:10.500 --> 00:09:14.039 footing the bill for the messages and voice calls 00:09:14.039 --> 00:09:15.659 and stuff being sent. If you're going to sign 00:09:15.659 --> 00:09:17.080 up to something like Twilio, you're going to 00:09:17.080 --> 00:09:20.360 end up paying for it. So that's the first piece 00:09:20.360 --> 00:09:22.080 I wanted to kind of clarify on this because I 00:09:22.080 --> 00:09:24.039 feel like that got lost a little bit in translation 00:09:24.039 --> 00:09:26.840 with this announcement. But the second piece 00:09:26.840 --> 00:09:28.700 and probably the more important piece is that 00:09:29.529 --> 00:09:31.450 Passkeys are going to become the default experience 00:09:31.450 --> 00:09:35.789 now, right? And I feel like this is probably 00:09:35.789 --> 00:09:40.549 more likely to cause folks issues and have a 00:09:40.549 --> 00:09:43.070 bigger user impact in organizations if you don't 00:09:43.070 --> 00:09:45.750 plan for it, right? So we all know Passkeys, 00:09:45.850 --> 00:09:49.309 it's a great solution. Some organizations, I 00:09:49.309 --> 00:09:51.549 think, are slower to adopt Passkeys than others. 00:09:51.830 --> 00:09:55.200 And with this change, there is definitely the 00:09:55.200 --> 00:09:57.340 potential where you're going to get users who 00:09:57.340 --> 00:09:58.899 are going to get sort of forced down this path 00:09:58.899 --> 00:10:01.820 key route. And if you don't get ahead of it and 00:10:01.820 --> 00:10:03.840 plan it with, you know, comms and all the rest 00:10:03.840 --> 00:10:06.259 of the things, there's going to be a challenge, 00:10:06.360 --> 00:10:09.259 right? So if we look at the timeline here on 00:10:09.259 --> 00:10:12.019 how this is supposedly going to play out, so 00:10:12.019 --> 00:10:14.460 the 1st of September, which, you know, as we 00:10:14.460 --> 00:10:17.700 sit here recording, it's going to be in, you 00:10:17.700 --> 00:10:19.379 know, just over a week's time. It's the 1st of 00:10:19.379 --> 00:10:23.230 September. Actually, it's exactly a week, is 00:10:23.230 --> 00:10:27.309 the 1st of September. So as you listen to this, 00:10:27.429 --> 00:10:29.549 it's probably yesterday, right, that this happened 00:10:29.549 --> 00:10:33.269 or a couple of days ago. But 1st of September 00:10:33.269 --> 00:10:36.350 is when essentially what's going to happen is 00:10:36.350 --> 00:10:39.190 users that are enabled for SMS or voice authentication 00:10:39.190 --> 00:10:42.330 methods are going to be auto -enabled for passkeys. 00:10:42.809 --> 00:10:45.389 So you're probably going to find if you don't 00:10:45.389 --> 00:10:47.450 get ahead of this and you have users that are 00:10:47.450 --> 00:10:50.029 using... you know, text messaging as a primary 00:10:50.029 --> 00:10:52.169 authentication method, they're going to start 00:10:52.169 --> 00:10:55.129 getting nudged for passkey registration, right? 00:10:55.350 --> 00:10:57.830 Now, initially, they're going to be able to skip 00:10:57.830 --> 00:10:59.950 that passkey registration and they can skip the 00:10:59.950 --> 00:11:02.629 prompt. But, you know, as time goes by, that... 00:11:02.799 --> 00:11:05.080 the ability to do that is going to go away. So 00:11:05.080 --> 00:11:07.940 that happens first of September. So you don't 00:11:07.940 --> 00:11:10.799 have much time. On the 18th, apparently, is when 00:11:10.799 --> 00:11:12.960 we'll start seeing the supported list of telco 00:11:12.960 --> 00:11:15.519 providers and what the deployment guidance looks 00:11:15.519 --> 00:11:19.720 like, you know, for all of this pricing, all 00:11:19.720 --> 00:11:21.980 of that type of stuff in the Microsoft Security 00:11:21.980 --> 00:11:24.539 Store. That's where we'll see that. And then 00:11:24.539 --> 00:11:27.299 I think 30th of October is when you can actually 00:11:27.299 --> 00:11:29.639 start configuring these sort of third -party 00:11:29.639 --> 00:11:33.509 providers. Now, you know... I don't know how 00:11:33.509 --> 00:11:37.549 concrete these dates are. This is stuff that 00:11:37.549 --> 00:11:39.350 I've sort of pieced together, you know, reading 00:11:39.350 --> 00:11:42.149 various articles and looking at various sources. 00:11:42.269 --> 00:11:44.850 So some of the stuff isn't from Microsoft officially. 00:11:45.070 --> 00:11:47.049 So again, you know, take it with a grain of salt 00:11:47.049 --> 00:11:50.029 in terms of when telcos are going to be providing 00:11:50.029 --> 00:11:52.049 stuff, when the partners will be released, all 00:11:52.049 --> 00:11:53.590 of that stuff. But hopefully this gives you a 00:11:53.590 --> 00:11:55.110 little bit of an indication as like the change 00:11:55.110 --> 00:11:57.409 is happening soon. You need to actually start 00:11:57.409 --> 00:12:00.360 looking at it. And then the 1st of February is 00:12:00.360 --> 00:12:02.320 sort of the next really big milestone date on 00:12:02.320 --> 00:12:03.639 the Microsoft side. That's when they're officially 00:12:03.639 --> 00:12:06.759 retiring that service in intro, right? So after 00:12:06.759 --> 00:12:12.220 the 1st of February, you know... Users are going 00:12:12.220 --> 00:12:15.639 to be forced to register their pass keys. There's 00:12:15.639 --> 00:12:19.360 not going to be a skip option anymore or a snooze 00:12:19.360 --> 00:12:22.080 option on that registration. So that prompt will 00:12:22.080 --> 00:12:24.399 essentially be forced for them to be logging 00:12:24.399 --> 00:12:26.120 in. So you've got to really get ahead of your 00:12:26.120 --> 00:12:30.139 registration campaigns and comms processes and 00:12:30.139 --> 00:12:33.779 all of that stuff to make sure that you're informing 00:12:33.779 --> 00:12:36.460 your user base of what's happening. And I think 00:12:36.460 --> 00:12:40.200 this is really the key message that... I wanted 00:12:40.200 --> 00:12:44.379 to get out for this today. If you have a large 00:12:44.379 --> 00:12:46.179 environment and every single one of your users 00:12:46.179 --> 00:12:49.779 is going to be affected by this, well, you've 00:12:49.779 --> 00:12:51.480 got a week, so that's not really a lot of time. 00:12:51.539 --> 00:12:57.279 There is an opt -out capability sort of to opt 00:12:57.279 --> 00:12:59.080 out of this between that 1st of September and 00:12:59.080 --> 00:13:02.039 1st of February window. It does delay that auto 00:13:02.039 --> 00:13:04.360 -enrollment, and it's sort of a one -time only 00:13:04.360 --> 00:13:07.779 thing. Once February hits, though, there's no 00:13:07.779 --> 00:13:10.240 more opt -out, right? So very important to know 00:13:10.240 --> 00:13:12.019 that you've got, you know, potentially between 00:13:12.019 --> 00:13:13.919 now and February to try and solve this problem. 00:13:15.080 --> 00:13:17.779 There are some great resources available. So 00:13:17.779 --> 00:13:20.919 I've put some links in the show notes to a PowerShell 00:13:20.919 --> 00:13:22.799 script that actually can help you identify this. 00:13:22.879 --> 00:13:26.299 So if you don't actually know, and I would encourage 00:13:26.299 --> 00:13:28.279 you to run this, even if you think everyone in 00:13:28.279 --> 00:13:30.899 your organization is running, you know, Authenticator 00:13:30.899 --> 00:13:34.889 or Passkeys or something. I still run this just 00:13:34.889 --> 00:13:38.590 to make sure, right? With my client, we have 00:13:38.590 --> 00:13:40.950 been running this sort of script over the last 00:13:40.950 --> 00:13:44.029 little while to identify people. And we actually 00:13:44.029 --> 00:13:46.730 noticed in the last month, some people who had 00:13:46.730 --> 00:13:49.769 previously been using Authenticator had all of 00:13:49.769 --> 00:13:52.309 a sudden switched to text messaging again. So 00:13:52.309 --> 00:13:54.610 just because you think that you had it under 00:13:54.610 --> 00:13:57.870 control last week or last month. may not be the 00:13:57.870 --> 00:13:59.970 case right now. So there is a link to a script 00:13:59.970 --> 00:14:02.470 on GitHub that Microsoft has published for this. 00:14:02.529 --> 00:14:04.590 We'll help you identify the scale and the scope 00:14:04.590 --> 00:14:07.889 of the change in your environment. There's some 00:14:07.889 --> 00:14:11.129 interesting information about comms and the process 00:14:11.129 --> 00:14:13.610 and all of that type of stuff on Microsoft Learn. 00:14:13.690 --> 00:14:16.070 So I've linked to that as well. And I'm hoping 00:14:16.070 --> 00:14:18.889 that this will kind of help folks, you know, 00:14:18.889 --> 00:14:21.549 just try and get ahead of this a little bit if 00:14:21.549 --> 00:14:26.000 you can. Not a lot of time here. So, yeah. I 00:14:26.000 --> 00:14:28.860 was surprised about the timelines, actually, 00:14:28.980 --> 00:14:31.559 Chris. I'm not sure if we have seen Microsoft 00:14:31.559 --> 00:14:36.860 announce something which was forced on everybody 00:14:36.860 --> 00:14:39.779 in such short notice. That's right, yeah. Usually 00:14:39.779 --> 00:14:41.720 you would think there's at least nine months 00:14:41.720 --> 00:14:43.799 or a year or whatever it is, right? But this 00:14:43.799 --> 00:14:45.419 is very, very sudden, and I think that's the 00:14:45.419 --> 00:14:48.139 thing. It was only announced in July, and it's 00:14:48.139 --> 00:14:51.480 taken that long. just for people to become aware 00:14:51.480 --> 00:14:54.799 of it right um so if you know um yeah definitely 00:14:54.799 --> 00:14:58.230 is is a very very short runway on this um and 00:14:58.230 --> 00:15:00.769 of course most people in the community the mvps 00:15:00.769 --> 00:15:03.230 everybody who's working with microsoft identity 00:15:03.230 --> 00:15:07.289 solutions very uh intensively they already saw 00:15:07.289 --> 00:15:09.450 this coming of course one day and they probably 00:15:09.450 --> 00:15:12.429 already moved away from sms authentication but 00:15:12.429 --> 00:15:14.610 i can imagine especially a lot of larger customers 00:15:14.610 --> 00:15:17.309 who uses its scale when they announce something 00:15:17.309 --> 00:15:20.429 in july and only six weeks later they are forced 00:15:20.429 --> 00:15:23.929 uh with prompts to the user which they can opt 00:15:23.929 --> 00:15:26.720 out of but six weeks is short notice i would 00:15:26.720 --> 00:15:30.240 say yeah yeah i mean a lot of organizations you 00:15:30.240 --> 00:15:33.539 can't get comms out to your entire business in 00:15:33.539 --> 00:15:36.500 six weeks right if you think about the the process 00:15:36.500 --> 00:15:38.580 and i mean for right for you know whether it's 00:15:38.580 --> 00:15:41.299 right or wrong or whatever it's just the reality 00:15:41.299 --> 00:15:44.399 is that it takes a while for the the information 00:15:44.399 --> 00:15:46.919 to filter through you know you have to engage 00:15:47.629 --> 00:15:49.690 you know, change management people, comms people, 00:15:49.850 --> 00:15:52.190 draft up the comms, review the comms, make sure 00:15:52.190 --> 00:15:54.289 the comms go out to all the right people, make 00:15:54.289 --> 00:15:55.929 sure that all the distribution lists are up to 00:15:55.929 --> 00:15:57.870 date so that everyone actually receives the comms. 00:15:57.909 --> 00:16:00.669 Like it's a, you know, it's not so easy when 00:16:00.669 --> 00:16:03.370 you're trying to do this at large scale. So hopefully 00:16:03.370 --> 00:16:05.629 this helps some folks. Yeah, and on the other 00:16:05.629 --> 00:16:07.990 hand, we're talking about cybersecurity here, 00:16:08.110 --> 00:16:12.990 and maybe companies should realize that they 00:16:12.990 --> 00:16:17.330 should adapt to have a faster way to implement 00:16:17.330 --> 00:16:20.669 changes. Because, well, nowadays with everything 00:16:20.669 --> 00:16:23.190 that's going on, with all the cyber attacks going 00:16:23.190 --> 00:16:26.450 on everywhere, you have to adapt quickly. Same 00:16:26.450 --> 00:16:28.409 as with patching, vulnerability management we 00:16:28.409 --> 00:16:31.429 talked about earlier. You can't have a server 00:16:31.429 --> 00:16:34.730 patch four times a year anymore. Because you 00:16:34.730 --> 00:16:37.149 don't get the comms out quicker to employees 00:16:37.149 --> 00:16:41.190 or whatever. It's something you have to adapt 00:16:41.190 --> 00:16:44.169 to, I guess, in this modern world. No, you're 00:16:44.169 --> 00:16:45.870 not wrong. That's a very good point. That is 00:16:45.870 --> 00:16:48.970 a very, very good point. Yeah. Yeah, and that 00:16:48.970 --> 00:16:52.330 maybe also allows me to bridge this a little 00:16:52.330 --> 00:16:55.129 bit into the next topic, project perception. 00:16:56.299 --> 00:17:01.419 Also an announcement in July, Monday 27th, Microsoft 00:17:01.419 --> 00:17:04.680 made a big announcement where they launched Project 00:17:04.680 --> 00:17:08.519 Perception to the world. I saw it all over my 00:17:08.519 --> 00:17:10.859 LinkedIn feed. Everybody was talking about it. 00:17:11.359 --> 00:17:13.700 Only a couple of people were actually at the 00:17:13.700 --> 00:17:15.779 launch event, but it was a big launch event. 00:17:15.819 --> 00:17:18.200 There were also some international press there, 00:17:18.339 --> 00:17:22.839 and they made quite some headlines there. What 00:17:22.839 --> 00:17:25.059 is Project Perception? Well, in short, it's marketing. 00:17:25.099 --> 00:17:28.160 Microsoft's agentic AI cybersecurity platform. 00:17:28.460 --> 00:17:30.420 And I know what you're thinking. You're hearing 00:17:30.420 --> 00:17:33.079 AI and you're probably already reaching out to 00:17:33.079 --> 00:17:35.220 the knob of your radio to turn off this podcast. 00:17:35.339 --> 00:17:39.440 Don't do it. It's no AI slop or AI marketing 00:17:39.440 --> 00:17:43.740 speak going on here. So Microsoft hasn't been 00:17:43.740 --> 00:17:45.940 very subtle about the framing either. So attackers 00:17:45.940 --> 00:17:50.630 now also. use ai and ai agents to attack companies 00:17:50.630 --> 00:17:55.190 and we mentioned being faster with your changes 00:17:55.190 --> 00:17:57.670 and that is actually something that you have 00:17:57.670 --> 00:18:01.789 to apply on defending for cyber attacks as well 00:18:03.309 --> 00:18:06.170 According to Microsoft, reactive security is 00:18:06.170 --> 00:18:10.789 over. Their answer boils down to it takes agents 00:18:10.789 --> 00:18:14.250 to fight agents. Well, you might think this is 00:18:14.250 --> 00:18:16.769 a lot of marketing speak by Microsoft, and it 00:18:16.769 --> 00:18:19.390 probably is a little bit. But if you look at 00:18:19.390 --> 00:18:22.130 the research, there are quite some figures to 00:18:22.130 --> 00:18:26.329 back this up. If you look at the results from 00:18:26.329 --> 00:18:30.029 a threat report from CrowdStrike, they measured 00:18:30.029 --> 00:18:33.680 an 89 % year -over -year increase. in AI -enabled 00:18:33.680 --> 00:18:38.900 adversary operations. And there's also an international 00:18:38.900 --> 00:18:42.759 AI safety report, which is a report written up 00:18:42.759 --> 00:18:45.480 by more than 90 authors. And they report that 00:18:45.480 --> 00:18:49.420 AI agents are already finding 77 % of all software 00:18:49.420 --> 00:18:52.799 vulnerabilities. The number that even surprised 00:18:52.799 --> 00:18:57.140 me more is that that capability is doubling every 00:18:57.140 --> 00:19:00.519 seven months. So if you look at how AI is being 00:19:00.519 --> 00:19:02.599 capable of finding software vulnerabilities, 00:19:03.079 --> 00:19:05.859 you can imagine that if you just let people loose 00:19:05.859 --> 00:19:08.019 with their AI agents are probably attacking you 00:19:08.019 --> 00:19:10.880 and abusing those, exploiting those vulnerabilities 00:19:10.880 --> 00:19:14.180 faster than you can patch them or defend against 00:19:14.180 --> 00:19:17.240 them. Like I said, Microsoft really wants to 00:19:17.240 --> 00:19:19.859 have something in place to coordinate the defense 00:19:19.859 --> 00:19:23.319 around such. attacks. What they kind of do, they 00:19:23.319 --> 00:19:25.859 use signals and perception, which they already 00:19:25.859 --> 00:19:29.059 have from your endpoints, your servers, your 00:19:29.059 --> 00:19:32.380 identity. That's Defender XDR as we know it, 00:19:32.380 --> 00:19:35.559 right? A lot of signals going in there. A lot 00:19:35.559 --> 00:19:38.859 of those signals are stored into tables and Microsoft 00:19:38.859 --> 00:19:43.420 is trying to have with AI agents, they want to 00:19:44.319 --> 00:19:46.660 Apply security context on top of that and have 00:19:46.660 --> 00:19:50.099 a continuously monitoring of your state, so to 00:19:50.099 --> 00:19:52.819 speak. So like I said, they work with agents. 00:19:52.920 --> 00:19:55.839 Well, not just separate agents, but actually 00:19:55.839 --> 00:19:58.900 entire teams of agents. And they divided them 00:19:58.900 --> 00:20:00.779 into separate teams. And I think this is quite 00:20:00.779 --> 00:20:03.299 a unique approach. They have red, blue, and green 00:20:03.299 --> 00:20:08.000 teams. The red teams sound familiar. They uncover 00:20:08.000 --> 00:20:11.069 attack parts and they should... be able to expose 00:20:11.069 --> 00:20:16.430 some attack parts in your environment before 00:20:16.430 --> 00:20:18.910 the attackers do, hopefully. The blue agents 00:20:18.910 --> 00:20:21.589 are there to do the triage on the incidents, 00:20:21.869 --> 00:20:24.950 but they can also correlate signals and come 00:20:24.950 --> 00:20:28.750 up with new detections to help you find things 00:20:28.750 --> 00:20:31.430 faster in the future. And then you have the green 00:20:31.430 --> 00:20:34.789 agents, and they are there to remediate findings 00:20:34.789 --> 00:20:40.059 and harden your posture. That concept of green 00:20:40.059 --> 00:20:44.470 agents and green teaming, it's not the... First 00:20:44.470 --> 00:20:46.549 time I've heard it, but it certainly is a new 00:20:46.549 --> 00:20:48.089 concept, isn't it? It's something we're starting 00:20:48.089 --> 00:20:50.690 to hear more and more now is that third line 00:20:50.690 --> 00:20:53.210 of defense, the green agents and the green team 00:20:53.210 --> 00:20:55.589 stuff. Yeah. Yeah. And it's funny that you mentioned 00:20:55.589 --> 00:20:57.869 that because the technology behind these green 00:20:57.869 --> 00:21:01.529 agents is not that new. I haven't covered this 00:21:01.529 --> 00:21:04.009 on this podcast yet. It was on my to -do list. 00:21:04.410 --> 00:21:07.650 It was called Project M -. Probably maybe you 00:21:07.650 --> 00:21:10.630 have heard of this. I have. It is an M and then 00:21:10.630 --> 00:21:14.180 dash, but it pronounces M -. It's an acronym 00:21:14.180 --> 00:21:16.900 apparently. It stands for multi -model agentic 00:21:16.900 --> 00:21:19.220 scanning harness. But sometimes I think these 00:21:19.220 --> 00:21:21.980 acronyms are just made up after the fact. I don't 00:21:21.980 --> 00:21:25.200 know. Well, it began as an internal project at 00:21:25.200 --> 00:21:27.519 Microsoft to hunt down vulnerabilities in software. 00:21:27.779 --> 00:21:30.619 And it even uncovered a batch of previously unknown 00:21:30.619 --> 00:21:34.960 bugs in Windows itself. And in July, MDash was 00:21:34.960 --> 00:21:37.359 actually released in private preview and integrated 00:21:37.359 --> 00:21:41.619 into the Defender exposure management pane. So 00:21:41.619 --> 00:21:43.869 as part of Defender... of XDR exposure management, 00:21:44.069 --> 00:21:48.289 they were already applying AI to help you finding 00:21:48.289 --> 00:21:50.750 some hardening improvements in your environment 00:21:50.750 --> 00:21:53.630 and through exposure management. And now they 00:21:53.630 --> 00:21:56.490 took MDash and they equipped the harness with 00:21:56.490 --> 00:21:59.490 an in -house cybersecurity model. And this was 00:21:59.490 --> 00:22:01.869 the second time. This was also a part of the 00:22:01.869 --> 00:22:04.410 big announcement, to be honest. So Microsoft 00:22:04.410 --> 00:22:08.930 has an in -house Microsoft AI division and they... 00:22:09.720 --> 00:22:12.240 are working and developing all kinds of specific 00:22:12.240 --> 00:22:15.420 LLMs for specific tasks. And for cybersecurity, 00:22:15.559 --> 00:22:19.480 they created a new model called MAI Cyber One 00:22:19.480 --> 00:22:23.099 Flash. Try to say that 10 times faster, Chris. 00:22:23.980 --> 00:22:27.779 Well, the MAI stands for Microsoft AI. So that 00:22:27.779 --> 00:22:32.349 is the... depicting that it is an in -house model 00:22:32.349 --> 00:22:37.309 from the AI division inside Microsoft. They also 00:22:37.309 --> 00:22:40.970 released a transcribed voice image and now Cyber 00:22:40.970 --> 00:22:48.690 1 Flash. I think Flash is added to... to let 00:22:48.690 --> 00:22:51.509 you know that this is a fast model. There will 00:22:51.509 --> 00:22:54.569 probably be larger cyber models that are working 00:22:54.569 --> 00:22:58.910 on as well. So again, with the combination with 00:22:58.910 --> 00:23:01.710 this new cyber model and the MDash capabilities 00:23:01.710 --> 00:23:05.190 they already created, that has now become the 00:23:05.190 --> 00:23:09.349 green agent, the green teams in Project Perception. 00:23:10.840 --> 00:23:13.640 So I saw quite some impressive demos, and I'm 00:23:13.640 --> 00:23:17.019 linking in the show notes also the original release 00:23:17.019 --> 00:23:19.319 with, I don't know, a one -hour video with a 00:23:19.319 --> 00:23:21.420 lot of demos in there. It's quite impressive 00:23:21.420 --> 00:23:23.779 how you can see that the red agents, for example, 00:23:23.779 --> 00:23:26.640 are hunting down vulnerabilities in some kind 00:23:26.640 --> 00:23:31.839 of web application or databases using legacy 00:23:31.839 --> 00:23:35.599 authentication, for example. The red teams will 00:23:35.599 --> 00:23:39.059 actually actively try and take advantage of these 00:23:39.059 --> 00:23:42.259 bugs and prove. that they are exploitable. And 00:23:42.259 --> 00:23:44.579 then the blue agents will propose new detections 00:23:44.579 --> 00:23:48.160 and hunting queries to make sure you're alerted 00:23:48.160 --> 00:23:50.900 when abuse happens. And the green agents for 00:23:50.900 --> 00:23:54.220 the whole MDash stuff is applying fixes. So they're 00:23:54.220 --> 00:23:57.880 proposing even literally GitHub or Azure DevOps 00:23:57.880 --> 00:24:02.519 pull requests to improve settings in your environment 00:24:02.519 --> 00:24:09.380 to fix those posture improvements. Okay, wow. 00:24:10.120 --> 00:24:13.599 With the whole new CyberOne Flash model they 00:24:13.599 --> 00:24:15.640 announced, they were quite proud, and I think 00:24:15.640 --> 00:24:19.279 rightfully so, that they were able to pair this 00:24:19.279 --> 00:24:24.200 with ChatGPT 5 .4 and achieve a score of 96 % 00:24:24.200 --> 00:24:28.609 on CyberGym. So Cybergym is a universal benchmark 00:24:28.609 --> 00:24:32.369 used by a lot of companies nowadays to prove 00:24:32.369 --> 00:24:37.150 how good a LLM model is into detecting software 00:24:37.150 --> 00:24:40.549 vulnerabilities. And I believe this was the highest 00:24:40.549 --> 00:24:45.269 score yet. They scored even 12 points above Anthropix 00:24:45.269 --> 00:24:51.480 Mythos. So Mythos was obviously... became quite 00:24:51.480 --> 00:24:56.400 famous lately. But they scored above that in 00:24:56.400 --> 00:24:59.420 the cyber gym and even with less power consumption. 00:24:59.799 --> 00:25:04.160 So this was... this was also a big thing and 00:25:04.160 --> 00:25:06.680 great to hear Microsoft thinking about this because 00:25:06.680 --> 00:25:08.779 we don't know what project perception is actually 00:25:08.779 --> 00:25:11.000 going to cost just to skip to the conclusion 00:25:11.000 --> 00:25:14.180 on this right now. It's a limited private preview 00:25:14.180 --> 00:25:16.660 right now. They're figuring out how they're going 00:25:16.660 --> 00:25:21.759 to make this happen cost -wise. But they did 00:25:21.759 --> 00:25:24.740 mention that they are aware that this whole security 00:25:24.740 --> 00:25:28.079 and AI defense isn't always on mission, right? 00:25:29.099 --> 00:25:32.519 So you have, to constantly be looking out for 00:25:32.519 --> 00:25:35.119 vulnerabilities and attacks and you're processing 00:25:35.119 --> 00:25:39.589 lots of tokens if you do that 24 -7. And these 00:25:39.589 --> 00:25:42.069 costs are a real constraint for the defenders. 00:25:42.210 --> 00:25:44.910 Because if you think of it, an adversary has 00:25:44.910 --> 00:25:48.009 the benefit here. They are only targeting a company 00:25:48.009 --> 00:25:50.970 or targeting a set of applications for a limited 00:25:50.970 --> 00:25:54.009 time. So they're spending tokens only for a limited 00:25:54.009 --> 00:25:56.269 time when they're using AI for that. And the 00:25:56.269 --> 00:25:59.349 defenders have to defend 24 -7. So Microsoft 00:25:59.349 --> 00:26:02.890 really recognizes this. And that's also why they 00:26:02.890 --> 00:26:06.730 shared some details on how they are actually 00:26:06.730 --> 00:26:09.990 routing different... models together so they 00:26:09.990 --> 00:26:13.089 use small models like the cyber one flash model 00:26:13.089 --> 00:26:18.359 for the day -to -day more more uh bulk of the 00:26:18.359 --> 00:26:21.279 ai operations and when it comes to harder tasks 00:26:21.279 --> 00:26:23.799 they're able to route it to a more expensive 00:26:23.799 --> 00:26:28.299 more elaborate bigger model um for for specific 00:26:28.299 --> 00:26:30.400 tasks and that would be the harness that's doing 00:26:30.400 --> 00:26:32.599 that right that's where you would use m dashes 00:26:32.599 --> 00:26:34.539 to make you know route between the models and 00:26:34.539 --> 00:26:36.619 stuff and yeah i'm i was going to ask you do 00:26:36.619 --> 00:26:39.299 you know if if the mai models are actually available 00:26:39.299 --> 00:26:43.440 for for use um like by themselves or do you have 00:26:43.440 --> 00:26:45.859 to use because i've i've heard of them obviously 00:26:45.859 --> 00:26:51.940 microsoft strategically they will be bit you 00:26:51.940 --> 00:26:54.579 know well served by developing their own models 00:26:54.579 --> 00:26:57.960 so they can be less reliant on open ai and and 00:26:57.960 --> 00:27:00.259 others right and so obviously for a while now 00:27:00.259 --> 00:27:02.200 they've been working on these mai models but 00:27:02.200 --> 00:27:04.640 i'm curious as to whether you could just use 00:27:04.640 --> 00:27:08.500 um one of the like the mai the cyber flash model 00:27:09.600 --> 00:27:12.500 itself in in stuff that you're doing you know 00:27:12.920 --> 00:27:17.039 Well, I can put some more research in that and 00:27:17.039 --> 00:27:21.920 maybe we can have this topic on a future episode 00:27:21.920 --> 00:27:24.440 because I think the whole discussion about open 00:27:24.440 --> 00:27:27.460 versus closed LLM models and running the models 00:27:27.460 --> 00:27:30.579 locally even without being depending on the big 00:27:30.579 --> 00:27:33.000 clouds and everything is a very interesting topic. 00:27:33.220 --> 00:27:35.880 I did a quick Google search and apparently the 00:27:35.880 --> 00:27:38.500 Microsoft MAI models are not open source. So 00:27:38.500 --> 00:27:41.549 they're a propriety in -house models. Just like, 00:27:41.549 --> 00:27:43.450 well, the Anthropic models are, for example, 00:27:43.450 --> 00:27:45.230 you can only use them through their services. 00:27:45.930 --> 00:27:49.490 I do see that the models can be used from Azure 00:27:49.490 --> 00:27:55.630 AI Foundry. So that's not the same as running 00:27:55.630 --> 00:27:59.450 them locally, of course, but then you can probably 00:27:59.450 --> 00:28:04.369 connect them. Foundry is expensive. I've been 00:28:04.369 --> 00:28:07.289 messing around with some of the GLM open models 00:28:07.289 --> 00:28:12.549 and GLM 5 .2. I think the Foundry cost to run 00:28:12.549 --> 00:28:16.289 that is something like $60 an hour. It's, you 00:28:16.289 --> 00:28:18.930 know, it's not nothing. But it's interesting 00:28:18.930 --> 00:28:21.630 because, you know, one of the challenges I think 00:28:21.630 --> 00:28:25.150 as a defender trying to defend things is that 00:28:25.150 --> 00:28:27.809 a lot of the commercial models have such big, 00:28:27.809 --> 00:28:30.390 like so many guardrails, right? You can't get, 00:28:30.470 --> 00:28:33.569 you know, Claude and Sonnet or any of the current 00:28:33.569 --> 00:28:37.309 Claude models to defend or do anything into in, 00:28:37.329 --> 00:28:41.690 you know. that's cyber -specific because it triggers 00:28:41.690 --> 00:28:45.190 the guardrails so quickly. And so you almost 00:28:45.190 --> 00:28:49.170 have to be looking at how can you use open models 00:28:49.170 --> 00:28:51.349 or use models where you can bring the processing 00:28:51.349 --> 00:28:54.210 of that stuff in -house or use it locally so 00:28:54.210 --> 00:28:57.259 that you can use the full... power of the model 00:28:57.259 --> 00:28:59.779 without tripping the guardrails right yeah so 00:28:59.779 --> 00:29:01.619 this whole thing is going to become very interesting 00:29:01.619 --> 00:29:03.500 because it's great to have the model but the 00:29:03.500 --> 00:29:05.500 model is not useful if if it doesn't allow you 00:29:05.500 --> 00:29:07.779 to do the things that you need it to do um so 00:29:07.779 --> 00:29:10.690 yeah Yeah, and that was also the main reason 00:29:10.690 --> 00:29:14.670 why Anthropic did not release the Mythos model 00:29:14.670 --> 00:29:18.170 a while ago. And then Fable launched, which was 00:29:18.170 --> 00:29:21.089 a more stripped down, a more secure version with 00:29:21.089 --> 00:29:24.369 a lot of guardrails, people noticed. So, yeah, 00:29:24.630 --> 00:29:27.970 that might be also the reason. I'm not sure if 00:29:27.970 --> 00:29:35.329 MAI… why CyberOne Flash is actually being available 00:29:35.329 --> 00:29:38.130 in AI Foundry to chat with and interactively 00:29:38.130 --> 00:29:43.930 interact with. When you look at Project Perception, 00:29:44.049 --> 00:29:46.549 and I put some screenshots also in the show notes, 00:29:46.650 --> 00:29:51.609 you have what they call playbooks. And these 00:29:51.609 --> 00:29:55.359 playbooks are now pre... But I can imagine there 00:29:55.359 --> 00:29:57.440 will be more and there will be probably some 00:29:57.440 --> 00:29:59.279 options where you create your own playbooks. 00:29:59.440 --> 00:30:03.000 So, for example, you can tell Project Perception 00:30:03.000 --> 00:30:05.960 that you want help with an investigation on an 00:30:05.960 --> 00:30:09.440 incident. And it has a blue icon telling you 00:30:09.440 --> 00:30:11.480 that the blue team will actually help you with 00:30:11.480 --> 00:30:13.720 that. And you point it to an incident and it 00:30:13.720 --> 00:30:16.099 will do the whole triage for you and it will 00:30:16.099 --> 00:30:19.160 write you a report. And if there are any remediating 00:30:19.160 --> 00:30:22.460 actions required, it will ask you for input. 00:30:22.509 --> 00:30:26.430 So it will never interact. It's always humanly 00:30:26.430 --> 00:30:29.390 driven on the decision part. But you can, for 00:30:29.390 --> 00:30:33.289 example, also ask it to protect against a specific 00:30:33.289 --> 00:30:37.670 threat. So there is from the... threat intelligence 00:30:37.670 --> 00:30:40.190 feed obviously a whole list of threats in there 00:30:40.190 --> 00:30:42.670 you can select a threat so i don't know some 00:30:42.670 --> 00:30:45.650 kind of storm or blizzard or you know all those 00:30:45.650 --> 00:30:47.890 you've seen those names before and you can say 00:30:47.890 --> 00:30:50.170 i want to protect my company against this threat 00:30:50.170 --> 00:30:52.950 and you can see a red a blue and green green 00:30:52.950 --> 00:30:55.890 icon stand next to there and there all the agent 00:30:55.890 --> 00:30:58.640 teams will actually come into play and try and 00:30:58.640 --> 00:31:00.980 see if you're actually vulnerable to that threat 00:31:00.980 --> 00:31:03.460 and what you can do to protect yourself against 00:31:03.460 --> 00:31:06.720 it. And also I really liked the green one, which 00:31:06.720 --> 00:31:09.000 is in the screenshot as well, plan for posture 00:31:09.000 --> 00:31:12.339 remediation. So then it will go across your estates, 00:31:12.339 --> 00:31:15.519 also your Azure cloud and all the Defender XDR 00:31:15.519 --> 00:31:20.059 stuff and give you a prioritized list of how 00:31:20.059 --> 00:31:22.619 you can improve your posture as well. And that 00:31:22.619 --> 00:31:26.140 is the MDash stuff you see now as part of perception. 00:31:26.990 --> 00:31:31.730 There goes my joke. Very new, of course. So it 00:31:31.730 --> 00:31:34.549 is in limited private preview, so you can request 00:31:34.549 --> 00:31:38.710 access, but I'm not sure how they will be giving 00:31:38.710 --> 00:31:40.769 out access right now. I can imagine there are 00:31:40.769 --> 00:31:43.849 a lot of constraints on hardware and costs in 00:31:43.849 --> 00:31:49.049 the back as well. Like I said, it's still read 00:31:49.049 --> 00:31:51.589 -only right now, so the agents are not actually... 00:31:51.960 --> 00:31:54.640 making changes for you. But I can't imagine a 00:31:54.640 --> 00:31:58.359 future where that will happen. It's only Azure 00:31:58.359 --> 00:32:00.480 for now when it comes to cloud reconnaissance. 00:32:01.099 --> 00:32:04.839 So no on -prem, no AWS or GCP or whatever. So 00:32:04.839 --> 00:32:08.619 there's a lot of stuff that will be added later. 00:32:08.819 --> 00:32:12.819 And it is expected that parts of this will be 00:32:12.819 --> 00:32:17.299 released during Ignite later this year. Okay. 00:32:17.380 --> 00:32:20.089 I mean, it's such an interesting... We're in 00:32:20.089 --> 00:32:23.390 this place now where things are changing so quickly, 00:32:23.470 --> 00:32:25.589 right? The rapid development in this is crazy. 00:32:25.789 --> 00:32:27.930 So it wouldn't surprise me if in a month from 00:32:27.930 --> 00:32:31.250 now, this has already matured significantly from 00:32:31.250 --> 00:32:33.769 where it is today, right? You know, we've definitely 00:32:33.769 --> 00:32:36.170 seen that type of improvements. And again, it 00:32:36.170 --> 00:32:38.049 goes back to your point that you made earlier 00:32:38.049 --> 00:32:42.000 of like, we really need... as organizations to 00:32:42.000 --> 00:32:44.380 be able to react more quickly to things right 00:32:44.380 --> 00:32:47.279 it doesn't help if you now look at this and you 00:32:47.279 --> 00:32:49.000 go hey this sounds like a good idea sounds like 00:32:49.000 --> 00:32:51.519 something i want to use in my organization and 00:32:51.519 --> 00:32:53.640 it takes you a year to plan to be able to implement 00:32:53.640 --> 00:32:56.519 it right because one year from now it's going 00:32:56.519 --> 00:32:58.440 to be a completely different landscape and and 00:32:58.440 --> 00:33:01.380 there's you know so the quick iteration you know 00:33:01.380 --> 00:33:04.359 fail fast mentality is is and mindset is really 00:33:04.359 --> 00:33:06.799 going to help um organizations stay ahead so 00:33:07.289 --> 00:33:11.289 Very cool. Yeah, and the company I work for has 00:33:11.289 --> 00:33:14.029 been a design partner on Project Perception for 00:33:14.029 --> 00:33:17.170 a while now. So we have access. So I was able 00:33:17.170 --> 00:33:20.140 to play around a little bit with it. Unfortunately, 00:33:20.180 --> 00:33:22.759 I couldn't take any screen captures or videos 00:33:22.759 --> 00:33:25.460 of that because it's obviously a private preview 00:33:25.460 --> 00:33:29.640 or limited public preview, I must say. But I 00:33:29.640 --> 00:33:32.200 really recommend looking at the introduction 00:33:32.200 --> 00:33:34.980 link in the show notes and watch the video there 00:33:34.980 --> 00:33:36.420 with some demos, and then you'll get a better 00:33:36.420 --> 00:33:40.400 understanding of how this is going to work. Yeah, 00:33:40.480 --> 00:33:43.980 and it's interesting times with all the AI stuff. 00:33:44.059 --> 00:33:47.119 I still sometimes have some hard times when it 00:33:47.119 --> 00:33:50.059 comes to... how we now interact with each other. 00:33:50.160 --> 00:33:54.079 I'm getting fully automated emails from HR by 00:33:54.079 --> 00:33:58.259 AI nowadays, and it makes me also sigh a little 00:33:58.259 --> 00:34:01.160 bit inside. But on the other hand, it's really 00:34:01.160 --> 00:34:04.880 interesting what AI can bring you. Not only it 00:34:04.880 --> 00:34:08.360 enables me to be a sort of programmer nowadays, 00:34:08.579 --> 00:34:11.360 I'm creating a lot of cool stuff I wasn't able 00:34:11.360 --> 00:34:13.639 to do in the past, but also with the whole defending 00:34:13.639 --> 00:34:17.260 stuff. I think it's really important to embrace 00:34:17.449 --> 00:34:20.010 the ai side of things there because otherwise 00:34:20.010 --> 00:34:23.090 like you said you're too late and the tackle 00:34:23.090 --> 00:34:26.550 has already taken place yeah and you're not you're 00:34:26.550 --> 00:34:29.590 not wrong um and i think that's a it's a good 00:34:29.590 --> 00:34:31.869 segue right because um well it's a good table 00:34:31.869 --> 00:34:34.409 for two reasons but let me comment is i i'm getting 00:34:34.409 --> 00:34:37.829 so much linkedin ai like i literally have people 00:34:37.829 --> 00:34:40.489 reaching out to me on linkedin um and it's all 00:34:40.489 --> 00:34:43.030 ai bots right because you know when you when 00:34:43.030 --> 00:34:44.980 you respond to say please know don't contact 00:34:44.980 --> 00:34:47.119 me again or thank you for the contact but you 00:34:47.119 --> 00:34:48.539 know because i like to try and be as polite as 00:34:48.539 --> 00:34:51.260 possible but after i've politely declined you 00:34:51.260 --> 00:34:53.219 three times and then you reach out to me a fourth 00:34:53.219 --> 00:34:55.380 time like that's just you know that's just rude 00:34:55.380 --> 00:34:58.179 right and it's because it's all ai so um you 00:34:58.179 --> 00:35:00.960 know um but i was gonna say uh it's a good segue 00:35:00.960 --> 00:35:03.239 because you know to talk about our community 00:35:03.239 --> 00:35:06.679 project i'd um i'd sort of said to everyone uh 00:35:06.679 --> 00:35:09.960 last month that i would i would update on uh 00:35:09.960 --> 00:35:12.320 connect 365 and kind of where we are and you 00:35:12.320 --> 00:35:14.139 know a lot of what i've been able to achieve 00:35:14.139 --> 00:35:17.940 working on this project has been because of the 00:35:17.940 --> 00:35:22.280 help of AI. And I conceptualized this stuff such 00:35:22.280 --> 00:35:27.380 a long time ago, but I struggle to find time 00:35:27.380 --> 00:35:30.360 to learn new things, right? And especially when 00:35:30.360 --> 00:35:32.579 it comes to learning to code in Rust, right? 00:35:32.639 --> 00:35:34.800 Stuff like that. I'm super interested in it, 00:35:34.900 --> 00:35:37.539 but I don't always have the time to spend. It's 00:35:37.539 --> 00:35:40.510 like learning to speak Dutch. love to do it don't 00:35:40.510 --> 00:35:43.409 always have the time to spend to um to uh to 00:35:43.409 --> 00:35:45.730 to commit to it right i'm not sure if i agree 00:35:45.730 --> 00:35:47.969 with that comparison chris because learning rushed 00:35:47.969 --> 00:35:51.889 you can use ai for that so you just tell ai to 00:35:51.889 --> 00:35:53.949 do that talking dutch you have to do yourself 00:35:53.949 --> 00:35:56.710 well both are equally useful though right so 00:35:56.710 --> 00:36:01.389 um but uh anyway so um Essentially, for those 00:36:01.389 --> 00:36:06.150 folks who don't know, Connect 365 is sort of 00:36:06.150 --> 00:36:08.190 a PowerShell -based, or it used to be a PowerShell 00:36:08.190 --> 00:36:10.530 -based tool that I put together. I think I wrote 00:36:10.530 --> 00:36:12.730 the first version of this thing back in sort 00:36:12.730 --> 00:36:17.130 of 2010 or 2012. And it was just a way to very 00:36:17.130 --> 00:36:19.630 quickly and easily, at the time, connect to Exchange 00:36:19.630 --> 00:36:22.409 Online. So, you know, you had to remember a whole 00:36:22.409 --> 00:36:24.449 sequence of commands in order to connect to Exchange 00:36:24.449 --> 00:36:28.510 Online via PowerShell back in the day. Instead 00:36:28.510 --> 00:36:32.449 of just stringing them together into a PS1 file, 00:36:32.650 --> 00:36:36.150 I wrapped a GUI around it so that you could execute 00:36:36.150 --> 00:36:38.190 it and then it would give you a login interface, 00:36:38.369 --> 00:36:42.429 basically. Now, back then, we used basic auth, 00:36:42.469 --> 00:36:44.610 so it was username and password. And so it was 00:36:44.610 --> 00:36:46.449 easy to just take the username and password, 00:36:46.530 --> 00:36:49.730 put it into the field on a GUI, and then just 00:36:49.730 --> 00:36:52.440 pass that through to... connect to login but 00:36:52.440 --> 00:36:54.679 over the years you know mfa became a thing and 00:36:54.679 --> 00:36:56.559 we we you know we don't use basic auth anymore 00:36:56.559 --> 00:36:59.239 and all of that type of stuff and so i've i've 00:36:59.239 --> 00:37:01.579 updated the script you know over the years to 00:37:01.579 --> 00:37:05.519 to include more modules and the most recent version 00:37:05.519 --> 00:37:08.340 now there's a few um out of date modules in it 00:37:08.340 --> 00:37:10.920 but it basically at one point in time had everything 00:37:10.920 --> 00:37:15.760 you needed to to administer m365 from from powershell 00:37:15.760 --> 00:37:18.820 so it had links to all the modules um and essentially 00:37:18.820 --> 00:37:22.340 you You launch the script, you get an interface 00:37:22.340 --> 00:37:24.760 and you tick the services you wanted to connect 00:37:24.760 --> 00:37:26.599 to. So if you want to connect to Exchange Online 00:37:26.599 --> 00:37:29.679 and, you know, Entra, tick those two things, 00:37:29.800 --> 00:37:32.440 pop in your username. It will then authenticate 00:37:32.440 --> 00:37:35.820 you through, you know, through Entra, MFA happens, 00:37:35.940 --> 00:37:38.420 all of that stuff. And you end up with a session 00:37:38.420 --> 00:37:41.000 that has all of the correct modules imported. 00:37:41.139 --> 00:37:43.280 So you can use one window to do all the stuff. 00:37:44.440 --> 00:37:46.400 That's kind of where we are. But, you know, the 00:37:46.400 --> 00:37:48.929 challenge with it. has been that it's always 00:37:48.929 --> 00:37:54.329 been native PowerShell. And originally it was 00:37:54.329 --> 00:37:59.670 using some WinForms UI components and the current 00:37:59.670 --> 00:38:03.110 version uses WPF, which none of the stuff works 00:38:03.110 --> 00:38:06.969 cross -platform. It doesn't work in PowerShell 00:38:06.969 --> 00:38:09.690 Core. So if you want to use it on Mac or Linux, 00:38:09.809 --> 00:38:12.789 you cannot. And so for probably the last three 00:38:12.789 --> 00:38:15.949 years, I've been wanting to rebuild this and 00:38:15.949 --> 00:38:18.909 make it cross -platform. I myself stopped using 00:38:18.909 --> 00:38:22.349 Windows many, many years ago. And so I haven't 00:38:22.349 --> 00:38:24.070 been using my own tool because I don't really 00:38:24.070 --> 00:38:26.170 use Windows anymore. And it's not really something 00:38:26.170 --> 00:38:30.110 I used. And so I wanted to redo this. And I've 00:38:30.110 --> 00:38:32.469 been sort of contemplating how I make this work. 00:38:32.869 --> 00:38:35.840 And I didn't want it to be... a script anymore 00:38:35.840 --> 00:38:38.800 for some reason. And I'm still playing with the 00:38:38.800 --> 00:38:40.960 idea of having a 2E version that is all text 00:38:40.960 --> 00:38:44.179 -based and PowerShell native still. So it's not 00:38:44.179 --> 00:38:46.840 impossible that that may happen. But I wanted 00:38:46.840 --> 00:38:50.800 to essentially build an app that handled all 00:38:50.800 --> 00:38:52.659 of the connectivity for you. And so that's essentially 00:38:52.659 --> 00:38:56.019 what I've done is I've spent the time to firstly 00:38:56.019 --> 00:38:58.960 update all the modules. So we are recent again 00:38:58.960 --> 00:39:02.960 with modules that you need. And there's a bunch 00:39:02.960 --> 00:39:04.980 of them now. So if you think about things like 00:39:04.980 --> 00:39:08.539 Graph and Entra, they're all there. Teams and 00:39:08.539 --> 00:39:10.460 Exchange Online, it's all there. But I've also 00:39:10.460 --> 00:39:12.940 included third -party modules. So PMP PowerShell, 00:39:13.000 --> 00:39:17.219 which is quite a popular one that is not a Microsoft 00:39:17.219 --> 00:39:19.800 module, but lots of folks use it. So that's included 00:39:19.800 --> 00:39:24.719 as well. One of the things I've also thought 00:39:24.719 --> 00:39:27.820 about that, you know, I know frustrates a lot 00:39:27.820 --> 00:39:32.260 of people is when you connect to Graph, you have 00:39:32.260 --> 00:39:34.860 to define your scope, right? The scope of the 00:39:34.860 --> 00:39:37.340 permissions and the rights that you want to use 00:39:37.340 --> 00:39:41.320 during that session. And so not everyone always 00:39:41.320 --> 00:39:43.300 understands or knows what exactly it is they 00:39:43.300 --> 00:39:47.320 want to do. And so what I've done is I've created 00:39:47.320 --> 00:39:50.079 some quick scopes so that if you are connecting 00:39:50.079 --> 00:39:54.019 and you click Graph, it will... prompt you to 00:39:54.019 --> 00:39:56.579 to select your scope so you can go if you if 00:39:56.579 --> 00:39:58.079 you want to there's an advanced option where 00:39:58.079 --> 00:39:59.840 you can go and pick exactly the scope that you 00:39:59.840 --> 00:40:02.440 want if you know it but if you don't let's say 00:40:02.440 --> 00:40:04.500 you just want to do a user administration i've 00:40:04.500 --> 00:40:07.150 got a user administration profile tick that and 00:40:07.150 --> 00:40:10.210 it's going to correctly get, you know, your user 00:40:10.210 --> 00:40:12.329 read -write -all, group read -write -all, directory 00:40:12.329 --> 00:40:14.309 read -all. So you're going to, you know, it's 00:40:14.309 --> 00:40:17.070 just a quick way to kind of help you get to the 00:40:17.070 --> 00:40:19.250 things that you need, right? So I've predefined 00:40:19.250 --> 00:40:21.530 five that I could think of were ones that were 00:40:21.530 --> 00:40:24.349 sort of commonly used. So there's some stuff 00:40:24.349 --> 00:40:26.869 for application management and device and Intune, 00:40:26.869 --> 00:40:31.059 stuff like that. So it handles sort of that heavy 00:40:31.059 --> 00:40:34.019 lifting or figuring that stuff out for you. And 00:40:34.019 --> 00:40:38.079 then the other thing that it does that I think 00:40:38.079 --> 00:40:40.500 is really, really helpful is it actually can 00:40:40.500 --> 00:40:44.539 manage modules for you and the module installation 00:40:44.539 --> 00:40:49.199 and module updates. So there's a prerequisite 00:40:49.199 --> 00:40:51.519 checker like there was on the PowerShell version. 00:40:52.159 --> 00:40:54.539 But now the great thing about this is, you know, 00:40:54.559 --> 00:40:59.150 you can actually install and update. the, you 00:40:59.150 --> 00:41:01.369 know, the required modules that you might want 00:41:01.369 --> 00:41:03.289 to use right from the app. And it will launch 00:41:03.289 --> 00:41:06.730 PowerShell and sort of do that all for you. So 00:41:06.730 --> 00:41:12.750 it is actually a native sort of app cross -platform. 00:41:12.989 --> 00:41:16.010 It's actually built in Rust, launches PowerShell. 00:41:16.050 --> 00:41:19.050 So you end up still being, you know, once executed. 00:41:19.710 --> 00:41:21.590 you still are dropped into a PowerShell window 00:41:21.590 --> 00:41:25.309 with the correct modules and all the stuff as 00:41:25.309 --> 00:41:27.429 before. It's just the user interface and all 00:41:27.429 --> 00:41:32.929 the app logic now runs as a Rust app. So my goal 00:41:32.929 --> 00:41:39.070 is that this will be correctly assigned code 00:41:39.070 --> 00:41:43.829 as applications that you can download and install 00:41:43.829 --> 00:41:47.289 on Windows, Mac, and there's a Flatpak version 00:41:47.289 --> 00:41:51.530 for Linux as well. So still a few little bugs 00:41:51.530 --> 00:41:54.670 and things for me to test and tweak out. I've 00:41:54.670 --> 00:41:56.769 been testing it pretty thoroughly on my Mac, 00:41:56.909 --> 00:41:58.409 but I've still got to do some Windows testing 00:41:58.409 --> 00:42:01.989 and certainly Linux testing. But hopefully by 00:42:01.989 --> 00:42:04.010 the time you all are hearing this and listening 00:42:04.010 --> 00:42:07.949 to this, this will be available in GitHub and 00:42:07.949 --> 00:42:11.079 you'll be able to take it for a test drive. provide 00:42:11.079 --> 00:42:12.800 some feedback if there's anything you find that 00:42:12.800 --> 00:42:14.980 uh i'm sure there will be bugs and things that 00:42:14.980 --> 00:42:17.079 don't work this it's actually quite a complex 00:42:17.079 --> 00:42:20.699 um application now compared to um what it used 00:42:20.699 --> 00:42:23.360 to be right it used to just be you know a few 00:42:23.360 --> 00:42:25.380 hundred lines of powershell code now it's actually 00:42:25.380 --> 00:42:29.039 a very you know complex thing um but i'm thinking 00:42:29.039 --> 00:42:31.860 about all sorts of other things you know um being 00:42:31.860 --> 00:42:33.679 able to do session management and all sorts of 00:42:33.679 --> 00:42:36.619 things with it as well so um first version probably 00:42:36.619 --> 00:42:38.480 won't have everything all the bells and whistles 00:42:38.480 --> 00:42:42.380 but uh it will hopefully function well for what 00:42:42.380 --> 00:42:46.119 it needs to do. So I got some screenshots in 00:42:46.119 --> 00:42:50.119 the show notes. You even got dark mode, Chris. 00:42:50.639 --> 00:42:54.280 You know, I personally prefer the light mode, 00:42:54.440 --> 00:42:59.820 but I know that everyone else in the world likes 00:42:59.820 --> 00:43:01.800 dark mode, and I knew that I would be getting 00:43:01.800 --> 00:43:04.440 a lot of drama if I did not include dark mode. 00:43:04.480 --> 00:43:07.840 So dark mode was one of the first UI things I 00:43:07.840 --> 00:43:11.409 did. Great. Well, you have to make sure that 00:43:11.409 --> 00:43:15.690 you share this properly aside from mentioning 00:43:15.690 --> 00:43:17.869 it here in the podcast, because I think a lot 00:43:17.869 --> 00:43:20.670 of people will be very happy with what you did 00:43:20.670 --> 00:43:23.170 here. Yeah. Yeah. Look, we'll get it functioning 00:43:23.170 --> 00:43:25.610 and working properly first. And then, yeah, definitely 00:43:25.610 --> 00:43:29.050 get it shared out. It's been, you know, I've 00:43:29.050 --> 00:43:32.110 been working on this project now for, you know, 00:43:32.110 --> 00:43:35.150 15 years or so. So it's been my sort of labor 00:43:35.150 --> 00:43:37.869 of love. But I'm really sort of happy to be. 00:43:38.199 --> 00:43:40.679 um updating it and bringing it into the sort 00:43:40.679 --> 00:43:43.079 of you know some new functionality and giving 00:43:43.079 --> 00:43:45.900 folks on on max and linux uh the ability to do 00:43:45.900 --> 00:43:47.440 it too i'm i'm you know i'm hoping to eventually 00:43:47.440 --> 00:43:49.360 just have the thing be available in all the stores 00:43:49.360 --> 00:43:51.199 right so you can just download it from the store 00:43:51.199 --> 00:43:54.860 or whatever um but uh you know baby steps i guess 00:43:54.860 --> 00:43:58.719 yeah we'll make sure uh people you follow chris 00:43:58.719 --> 00:44:02.480 on linkedin and he'll uh he'll share his updates 00:44:02.480 --> 00:44:05.460 there i can imagine so we mentioned linkedin 00:44:05.460 --> 00:44:08.380 by the way earlier with with all ai content there 00:44:08.380 --> 00:44:11.139 i also saw that linkedin is now providing you 00:44:11.139 --> 00:44:13.579 an option in your feed when you click on the 00:44:13.579 --> 00:44:16.420 three dots you see an ai post you can click on 00:44:16.420 --> 00:44:18.940 the three dots and you can click on this seems 00:44:18.940 --> 00:44:22.119 like ai slop they even call it ai slop i like 00:44:22.119 --> 00:44:27.539 it so People know that this is there and fine 00:44:27.539 --> 00:44:30.519 -tune your timeline so that you only see some 00:44:30.519 --> 00:44:32.960 more human and relatable content there. And then 00:44:32.960 --> 00:44:37.159 that will ensure that Chris updates content will 00:44:37.159 --> 00:44:39.860 not get lost in the AI slot. Yeah, fair enough. 00:44:39.960 --> 00:44:41.679 That's a good one. You know what? Actually, while 00:44:41.679 --> 00:44:42.920 we're talking about social media, I've actually 00:44:42.920 --> 00:44:46.059 been thinking about going back. I haven't logged 00:44:46.059 --> 00:44:50.059 into Twitter or X. Oh, at least two and a half 00:44:50.059 --> 00:44:52.659 years, I think. I've been thinking recently that 00:44:52.659 --> 00:44:54.320 I should, because I don't have the app installed 00:44:54.320 --> 00:44:56.059 on my phone anymore or anything like that, but 00:44:56.059 --> 00:44:58.119 I've been thinking that I should log in and take 00:44:58.119 --> 00:45:01.360 a look. I do feel like I miss out on some news 00:45:01.360 --> 00:45:04.000 and stuff, right? I feel like I used to get a 00:45:04.000 --> 00:45:07.929 lot more interesting before. it got completely 00:45:07.929 --> 00:45:11.030 ruined. So I kind of wonder what the status or 00:45:11.030 --> 00:45:14.269 the state of, of Twitter is now. Um, because 00:45:14.269 --> 00:45:16.289 tech Twitter, especially cyber Twitter used to 00:45:16.289 --> 00:45:18.530 be really good, man. When you, if you follow 00:45:18.530 --> 00:45:21.150 the right people, you got some really good information 00:45:21.150 --> 00:45:23.590 there. And I try to recreate that on the other 00:45:23.590 --> 00:45:25.949 platforms, you know, um, blue sky and stuff, 00:45:26.010 --> 00:45:28.289 but I just never caught on quite the way Twitter 00:45:28.289 --> 00:45:30.690 did. So, you know, what is there? Yeah, well, 00:45:30.750 --> 00:45:33.309 that's it. So I might, um, I might do that in, 00:45:33.309 --> 00:45:37.159 excuse me, in the next month. Report back on 00:45:37.159 --> 00:45:38.860 what I find, right? Whether the thing is just 00:45:38.860 --> 00:45:42.280 all crap again or whether it's with a try. I 00:45:42.280 --> 00:45:45.619 did notice that a lot of my, when colleagues 00:45:45.619 --> 00:45:50.480 are sending me AI stuff about, like, for example, 00:45:50.480 --> 00:45:53.800 how to work with cloud code and how to implement 00:45:53.800 --> 00:45:58.579 loop and skills and tasks and whatever, a lot 00:45:58.579 --> 00:46:00.980 of the content they are sharing is through X. 00:46:01.139 --> 00:46:04.159 So a lot of videos on there, which I couldn't 00:46:04.159 --> 00:46:06.860 even find a YouTube account. part for so it made 00:46:06.860 --> 00:46:10.199 me log into twitter as well or x again to access 00:46:10.199 --> 00:46:12.380 the links they're sending me so apparently the 00:46:12.380 --> 00:46:15.659 whole ai technology development stuff is also 00:46:15.659 --> 00:46:19.480 hot on on x uh -huh well excellent well that's 00:46:19.480 --> 00:46:22.530 good We'll see where we go. But, you know, as 00:46:22.530 --> 00:46:24.809 we talk about that, here we go to the end of 00:46:24.809 --> 00:46:27.809 another episode. Man, time flies when you're 00:46:27.809 --> 00:46:30.630 having fun. Guys, thank you for sharing all about 00:46:30.630 --> 00:46:32.630 Perception and all of the cool stuff there. It'd 00:46:32.630 --> 00:46:34.389 be interesting to get my hands on that at some 00:46:34.389 --> 00:46:37.780 point. Thank you for sharing that. I think that 00:46:37.780 --> 00:46:40.519 looks very, very promising. Folks, thank you 00:46:40.519 --> 00:46:43.239 for listening. Really appreciate you taking time 00:46:43.239 --> 00:46:45.380 to listen to us. And, you know, of course, we're 00:46:45.380 --> 00:46:47.980 always open if you have any feedback or anything 00:46:47.980 --> 00:46:50.340 specific that you'd like us to look into and 00:46:50.340 --> 00:46:52.360 address for you, reach out to us, either one 00:46:52.360 --> 00:46:55.519 of us or both of us. We can be found pretty easily 00:46:55.519 --> 00:47:00.500 on the socials. And until next time, Cos, enjoy 00:47:00.500 --> 00:47:05.329 your remaining bits of summer. I will look forward 00:47:05.329 --> 00:47:07.190 to seeing you and speaking with you again very, 00:47:07.269 --> 00:47:10.050 very soon. Yeah, see you later, man. Thanks for 00:47:10.050 --> 00:47:11.050 listening, everybody. Bye -bye.