October 16, 2026
2026-10-16
https://www.youtube.com/watch?v=drWniTIWqDI

[0:09] We go live in ten.
[0:23] Every morning, the future moves faster.
[0:27] The models, the money, the power, and the people building what comes next.
[1:06] From Austin, Texas to the world.
[1:13] AI News, Real Context.
[1:16] This is the Austin Wilson Show.
[1:19] I'm Austin Wilson, your human in the loop.
[1:28] Good morning, and welcome to the Austin Wilson Show where we're breaking down the AI headlines that matter to you.
[1:35] Today, I'm Austin Wilson to you coming, uh, coming to you live from the AI capital of the world, Austin, Texas.
[1:42] It is Friday, October 16, and we're excited to get into it.
[1:48] It's almost the weekend.
[1:50] Let's go.
[1:51] Today, we are talking about ChatGPT ads that know your phone number, Amazon $75, uh, device, wearable.
[2:01] The Fed says AI is freezing hiring.
[2:05] Meta's free Muse Spark open model.
[2:09] Uh, Accenture's $29 Claude allowance, and then ChatGPT's voice model has some new accents.
[2:17] Excited to get into it with you guys today, cover the top breaking AI news stories.
[2:24] Let's get into it.
[2:25] Starting up first, advertisers can upload your email or phone number and aim ChatGPT ads at you, and a new site just tested it.
[2:37] Four zero four media bought ads warning that humans read chats.
[2:42] Opting out of personalized ads doesn't stop them.
[2:45] Ads can still follow what you type in the current chat.
[2:50] Oh, alright.
[2:51] Here we go.
[2:51] ChatGPT's free tiers now work like Facebook ads.
[2:55] Your contact info follows you into the chat.
[2:57] OpenAI says advertisers never see which people matched and uploaded lists are usually deleted within a day.
[3:05] Lists need more than 25,000 people, and one tracker has logged over 43,000 ChatGPT ads.
[3:12] Still, OpenAI wouldn't say whether it scrambles raw emails and phone numbers the way that Meta does.
[3:18] Uh, this is coming out of four zero four media who, uh, tested this.
[3:23] Alright.
[3:23] We're getting into it.
[3:24] Ads and ChatGPT, we've we've kinda seen this coming for a a while and, um, have maybe seen some of it pushing, but maybe not to the extent that that we're at now.
[3:35] And this is the point where we kinda see...
[3:38] We've talked about this quite a bit lately, but is OpenAI compute constrained, or are they just money constrained?
[3:44] Right?
[3:45] And I think at this point, they need to generate revenue, especially on the heels of Anthropic reportedly coming out this week with, you know, record numbers, uh, for the quarter, but also being profitable.
[3:59] Uh, we, of course, don't know, uh, what OpenAI's status is from that standpoint, but they've gotta be, you know, in that same realm where they're competing to be competitive, uh, or or competing on on a price and cash flow and looking at future IPOs and all of the above.
[4:17] Uh, and so they've got to generate revenue wherever they can and as much as they can to continue to fuel this fire.
[4:25] Uh, you know, projected spending through 2030 is up to $700,000,000,000 at OpenAI reportedly.
[4:33] And so you've gotta imagine the folks over there trying to figure out, hey.
[4:38] Either we've gotta save money, and I don't know where we're gonna do that, or we've gotta figure out ways to make more money.
[4:44] And so, uh, the, uh, the ad platform is kind of a natural route to take, especially when you have, I believe, over a billion weekly users on their ChatGPT app.
[4:58] Uh, what do you guys think about this?
[5:01] Is this one of those things where, hey.
[5:04] We get used to ads and it's cool, or is it one of those where, uh, man, now they have ads.
[5:12] They've ruined the experience, and we're done.
[5:15] Let us know what you think in the comments.
[5:18] Alright.
[5:19] Uh, moving on.
[5:20] Amazon has launched a $75 wristband that listens to your conversations and then turns them into summaries and to do list.
[5:31] B Frontier lasts a week per charge.
[5:34] That's the name of the device, the b Frontier.
[5:37] Uh, you have a $100 phone back version.
[5:40] The b Eclipse will come, uh, later.
[5:43] And then b Pro, the subscribe...
[5:45] Subscription model is $12 a month.
[5:49] What do we think about this?
[5:50] Amazon wants AI that remembers your day, and it plugs straight into Alexa and Amazon Shopping.
[5:56] Amazon Amazon says no audio is ever stored, and only you can read your transcripts.
[6:02] Frontier sells now for $75 with up to seven days of battery and a light that shows when it's listening, but the people around you never agreed to be transcribed.
[6:17] Man, we've seen a lot of devices like this, Claude and Pocket and a handful of others.
[6:26] Uh, and and I think, um, we saw, I believe, an acquisition.
[6:30] OpenAI acquired another wearable company maybe last year or earlier this year that that had a similar use case as well.
[6:39] Uh, these kind of...
[6:40] These wearable things, maybe it clips onto your lapel or, yeah, you you...
[6:45] It sticks to the back of your phone, uh, or you can put it in your pocket.
[6:49] And, uh, you can turn it on and it'll listen to you and transcribe, uh, what you're doing.
[6:55] What's the value in this?
[6:56] Well, you can give the AI context about everything that's going on kind of in your non digital life.
[7:03] Right?
[7:03] So, you know, you're talking to your spouse about what's going on this weekend, and we need to schedule a babysitter for Saturday night because we're having a date.
[7:12] Well, that's not gonna automatically get into your agent world and and do that without having some sort of transcription or recording device.
[7:22] Right?
[7:22] So you're able to record that, and then your model, your agents would be able to, you know, action that item on on your behalf.
[7:30] So there's interesting use cases.
[7:32] There's a lot of value in being able to, uh, give agents more access to your life or more, uh, information about kind of what you're doing and what you're working on and really interesting to, you know, see the different, like, kind of little things that people find to do with this.
[7:52] That being said, from Amazon's standpoint, it seems like there's been a lot of these have already come out, and maybe they're a little bit behind on the the device space, or why not just go acquire another one of these companies?
[8:07] And and may...
[8:08] Maybe they did.
[8:08] I'm I'm not aware, but, uh, it doesn't seem like a groundbreaking thing coming out of Amazon.
[8:15] And, also, it's, uh, you know, from a personal standpoint, I'm like...
[8:19] Wait.
[8:20] They're like, hey.
[8:20] Yeah.
[8:21] We promise.
[8:22] No. Uh, you know, we're not gonna store any of the data, record anything.
[8:26] It's like, alright.
[8:28] Sure.
[8:28] Uh, sure that's the case.
[8:31] Not not to, uh, accuse him of anything, but I just get a little bit speculative, uh, right, about, you know, maybe a large tech company and whether they're they're going to try to data mind you if you're using their device.
[8:46] I don't know.
[8:46] What do you guys think about listening devices?
[8:50] Do you think these are gonna be really valuable?
[8:54] Would you use one?
[8:55] Would you use the the Amazon, uh, listening device?
[8:58] Also, this is kinda funny, but earlier this week, we're talking about a a bug that was planted in OpenAI's offices.
[9:05] A spy planted a bug that was a listening device.
[9:09] And now, you know, later this week, we've got Amazon coming out with a listening device that's for personal use.
[9:15] Really funny.
[9:16] Tell me what you guys think.
[9:17] Would you use one of these?
[9:19] If you have one of these, how do you use it?
[9:21] I'm curious.
[9:22] I haven't tested one yet, and I'm curious to see if if it's really worth it.
[9:31] Alright.
[9:32] The Federal Reserve's latest survey of businesses found AI is quietly shrinking jobs, not with layoffs, but by skipping new hires.
[9:42] Five of 12 regional fed districts reported firms cutting staff through attrition, delaying hiring, or expecting fewer jobs after investing in AI.
[9:53] In Dallas, AI agents ran one hundred and eighty hours of job interviews in two days, for example.
[10:00] We're starting to see this more and more.
[10:02] If you're job hunting, especially entry level, the door may be shrinking before anyone gets fired.
[10:09] Minneapolis staffing firms said AI isn't causing layoffs, but is dampening new job creation.
[10:17] A Dallas accounting contact said big firms are delaying entry level contracts partly because AI makes them more efficient.
[10:26] But it's anecdotes from business contacts with no job account, and overall employment was roughly flat.
[10:35] Wow.
[10:36] We've covered some some stories like this.
[10:38] We've talked about the the fact that, you know, earlier this week, talking about call centers where it's, hey, not necessarily the the the fact that we are completely replacing humans with voice agents.
[10:51] Right?
[10:52] But for instance, if, uh, you have to call in or if you got a a team of callers that, uh, need to call in and confirm payments and they're checking accounts receivable and they're they're trying to figure out how to get to accounts receivable at a company, well, you could have a voice agent call and do that.
[11:10] That makes that person that much more efficient, and then you don't need to hire as many people.
[11:15] Right?
[11:16] You don't need as many, uh, staff hours, working hours, uh, of people doing these jobs.
[11:22] And so this is, I think, of the beginning of something that we'll continue to...
[11:27] Continually see both from the standpoint of the models being intelligent enough and capable enough, but also from humans trusting and, um, allowing these models to begin to take on more and more work.
[11:42] I think we'll start to see more and more of this in different use cases.
[11:46] Of course, this doesn't translate to everything yet.
[11:50] So we're, you know, seeing this at lower level, entry level jobs, anything that's repetitive, um, kinda manual stuff, deterministic, meaning that we know, okay, the steps to do this job are a, c essentially, and 90% of the time, that's what you do is a, b, c.
[12:10] Right?
[12:10] So if you can define that really well, uh, and then have a way to confirm that the AI is doing that job correctly, it's, like, pretty easy, um, or or not easy necessarily, but simple, um, to install an agent that can go...
[12:28] An AI agent that can go do that job.
[12:31] So, um, we're seeing it, um, kind of more and more in, uh, obviously, like voice agents, lower level stuff.
[12:39] But but I think the next thing we're gonna start seeing is in the, you know, financial and accounting space, we're starting to see people augment, uh, with accounting tools that basically can can do most of the bookkeeping for you.
[12:55] And I don't know that people are fully trusting these models yet.
[13:00] Again, um, that's a much more complex use case and, uh, higher stakes if you get it wrong, but I think we're, you know, gonna start seeing humans adopt it.
[13:11] The intelligence is there from what the benchmarks have shown is that these models are capable enough to handle more advanced accounting and, uh, tax use cases.
[13:21] And so are we going to start seeing companies adopt more and more of, uh, these these models and and the, uh, super intelligence in those use cases?
[13:31] It'll be interesting.
[13:32] Um, you know, are you guys experiencing this in your lives?
[13:38] I, um, I have not necessarily been, you know, uh, personally affected or know anybody that's been personally affected in this specific way.
[13:51] So I I can't speak to it directly, but are you guys seeing any of this kind of shrinkage where it's not layoffs, but maybe we're we're slowing down hiring?
[14:01] Um, maybe, uh, we don't need as many hours for for a human to work in this.
[14:07] And then what do you guys do?
[14:09] Do you do you go repurpose those people into, uh, other more complex tasks?
[14:15] I...
[14:15] It...
[14:15] It's interesting to follow and see where this goes.
[14:17] And and, of course, we don't want anybody to get displaced, and so I'm I'm interested to see how do we reposition people, retrain people.
[14:26] Uh, that's why I'm a big advocate for training and studying and and learning how to use these AI tools and how to become really proficient in them because I think it'll be a superpower as we continue to develop and adopt more superintelligence as as the the superintelligence progresses and gets better, and and we get better at using these tools.
[14:50] Um, I guess, would you do a job interview with superintelligence?
[14:57] Yes or no?
[15:00] We'll see.
[15:02] Meta just gave away a top AI model that you can run on two Mac studios if you have roughly $38,000 lying around.
[15:11] Uh, the model is called the MetaMuse Spark Open.
[15:16] It's built for coding and long tasks, and Apple tuned its software so the model runs on its desktops.
[15:23] Meta's best model, Muse Wonder, stays locked up.
[15:27] It's still a closed source model.
[15:30] Uh, a rival level AI living on your own desk means your data never has to leave the building.
[15:39] Meta calls it the most capable American AI model anyone can download for free.
[15:46] Meta and Apple ran it on two linked Mac studios where it fixed bugs in a real app by itself.
[15:54] But in Gadget, pegs of working Mac setup roughly $38,000, and Chinese free models still beat it at some coding.
[16:03] Interesting to see here.
[16:04] Hey.
[16:04] We're gonna we're gonna try to get some input here from, uh, our producer producer, Ron, and see if he can give us some feedback here.
[16:14] Let me see if we can get Ron in here.
[16:17] Uh, okay.
[16:18] It looks like Ron went to get some copy, uh, so we'll have to touch base with him, uh, later.
[16:24] This is great.
[16:25] Again, we're seeing more and more open models, advanced intelligence, uh, in these open source models, which is really cool.
[16:33] The the price tag on the Mac Studio is the $38,000.
[16:37] That's just the demand for the hardware has increased so much that these are really, really expensive.
[16:44] And this is a larger model, so it's gonna take a a significant amount of hardware to run it on.
[16:51] You're not gonna run it on, you know, a simple MacBook.
[16:54] You you need a little bit more punching power there for this new meta model.
[17:01] Uh, that being said, what's really cool is this is open source, uh, or or open weight.
[17:06] Right?
[17:06] So, uh, what you can do is you could run that on a virtual private server, run it in the cloud, and be able to, uh, use that at at actually a much lower cost.
[17:19] And and for, you know, the everyday use cases, the average person doesn't need to spend $40,000 on on a Mac studio or two Mac studios to run this deal.
[17:28] But really cool, again, to see, um, more open source intelligence coming out.
[17:33] They're pushing the envelope, more open source American intelligence.
[17:37] It's funny that, like, maybe a year ago, just the Chinese open source models were phenomenal.
[17:43] They were the the, you know, best open source models that you could get, and nobody was really even close, um, just in terms of the rest of the world.
[17:52] Right?
[17:53] The Chinese models were leading in a big way.
[17:56] And, of course, Anthropic, uh, Anthropic and OpenAI had better, you know, frontier models, but closed source.
[18:04] The the open source community, there's nothing in The United States.
[18:06] Fast forward to a year later, and we see now we've got multiple open models coming out of The United States that can compete and and in some cases, you know, do a little bit better.
[18:19] Some cases do a little bit worse, but really great to see the competition.
[18:23] And really interesting from Meta, the fact that they are...
[18:28] You know, have the the muse wonder model that that is, you know, frontier level, pushing the frontier, really, really phenomenal model that's closed source, but they also have their open source offering, which is awesome as well too.
[18:43] And to see what the Alexander Wang and and Marc Zuckerberg and team have done over there is, uh, really, you know, just encouraging and exciting to see.
[18:54] Uh, I think, you know, referencing, if you didn't see last night, uh, crazy roadster reveal from Tesla.
[19:04] They flew this car.
[19:06] Like, this car can do crazy things, and you and you see the way that we are pushing things forward from a technological standpoint and what Americans can do, especially when you kind of put us in a corner speaking about the open source thing, or, you know, just things like Elon Musk as well, and and the Tesla folks, and, you know, SpaceX for that matter, just have a chip on their shoulder and are gonna go out and, you know, especially when you tell them you can't do something, they're gonna...
[19:37] It's kinda like my toddler.
[19:38] Right?
[19:39] You tell them you can't do something, they...
[19:41] They're gonna jump right in.
[19:42] So really interesting to see what we're able to produce as a society, and this is no small feat from Meta and really good for the the superintelligence community, good for intelligence on the whole and pushing the the the whole level of a technology forward.
[20:04] We'd love to see it.
[20:05] I I guess, is anybody in in the in the audience going to go out and spend $40 on a home setup for Mac Studios and run this locally?
[20:17] Or do you already have a local setup?
[20:19] If so, tell me about it.
[20:21] I would love to nerd out about Mac Studios.
[20:26] That sounds awesome.
[20:27] I would love to see that.
[20:28] Alright.
[20:29] Moving on.
[20:29] We are talking about Accenture reportedly told staff that the company now only covers $29 a month of Claude after, uh, some employees burned through a thousand dollars of Claude credits.
[20:45] Since October 1, anything over that must be billed to a project or team, and one worker said their access was cut off entirely.
[20:53] Oh, man.
[20:55] Can you believe that?
[20:56] Even companies selling AI to everyone else are flinching at what their own people actually use.
[21:02] Accenture says it makes AI available to all its people with the right tool for each job.
[21:08] Workers told four zero four media they regularly spent 500 to $1,000 a month, dwarfing the new $29, uh, allowance.
[21:19] The catch is Accenture sells AI advice to clients while rationing it for its own staff.
[21:27] Uh, this is an interesting story because $29 a month is very low, especially when you're talking about enterprise AI or enterprise API cost for, uh, for AI.
[21:39] You know, I don't know that you're getting much done for $29 a month in straight API cost.
[21:45] And especially when you contrast that to, like, AI researchers and advanced folks, you know, reportedly using tens of thousands of dollars of API, uh, in in AI tokens a day, uh, if not...
[22:04] You know, in some cases, it's like millions of dollars a month in compute that these advanced researchers are running.
[22:11] And so then when you talk about Accenture, they are throttling this back.
[22:17] And even from a thousand or $500 a month does not seem like, you know, astronomical numbers.
[22:23] So, um, I wonder where is this coming from.
[22:28] Is Accenture just in a, you know, desperate place to cut cost and to continue to grow their margin?
[22:36] I feel like they're in a place that's pretty competitive right now from...
[22:41] Especially what we're seeing in SaaS software and and Accenture doing a lot of consulting in the software business and a lot of revenue coming from implementation of software, and and maybe they're, you know, seeing some of that go away, uh, and maybe they're just trying to cut costs and save from there.
[23:00] The other angle here, though, is that they are...
[23:05] They're really just not getting the value out of it, which if that's the case, that's extremely bearish, uh, for the the AI market as a whole.
[23:14] And what does that mean if, you know, you've gotta imagine Accenture is probably more of a power user?
[23:22] You know, they're they're a technology business.
[23:24] They're in the technology business.
[23:26] If they're not getting value out of it, what does that say about the broader kinda ecosystem?
[23:33] And if you're not getting a return from what you're paying for, how long can you continue to pay for this?
[23:40] You know, we see Anthropic's revenue growing quarter over quarter.
[23:46] We're seeing astronomical CapEx announcements from these leading technology businesses, continued investment, and projections of spending hundreds of billions of dollars into the next couple of years for for kinda each of these tech companies.
[24:04] Uh, if Accenture is saying, hey.
[24:07] Actually, we're...
[24:07] It's not worth us, you know, worth worth...
[24:10] It's not worth more than $29 a month to us, then then what does that mean for the the broader business?
[24:17] Or is this signaling that just Accenture is trying to penny pinch here?
[24:23] I don't know.
[24:24] I don't really have an...
[24:25] Strong opinion on this or a strong feeling.
[24:29] Actually, you know, to take that back, I get tons of value out of my AI.
[24:36] That being said, you know, I'm on on subscription models, so it's heavily subsidized.
[24:40] If I was paying for enterprise API, it would cost me thousands of dollars a month.
[24:45] I don't think I would spend that at all.
[24:47] But certainly more than $29 a month, I would certainly get value out of that.
[24:52] So I don't know.
[24:52] It'll be interesting to see what this leads to or or what this is a precursor for, but I think it's important to pay attention to what, you know, Accenture and and Deloitte are doing and and these consulting firms and understanding if they're the boots on the ground using the tools, what what value are they getting out of that, and how does that translate?
[25:14] And, again, I'll go back to...
[25:15] I've said this many times, but I think we're going to see kinda some major breakthroughs before the end of the year in terms of not just how intelligence the how intelligent the models are and and what they're capable of doing, but, uh, but that that capability and the the second order effects that we'll see coming from that intelligence, I think, is where we see the big breakthroughs.
[25:41] Right?
[25:41] Where our kind of paradigm will probably shift quite a bit in how we see things and how we view the world and what we're able to work on, what we're able to accomplish, I think will change quite a bit.
[25:55] And I think for the better, like, think this is an exciting reason to, uh, use superintelligence tools and and and to get really good at them because I think the implication is huge.
[26:09] And so we'll we'll continue to follow what's going on in these tech businesses.
[26:15] Uh, you tell me, could you survive on $29 a month of Claude or ChatGPT?
[26:23] Would that be enough for you to to get done what you need to get done?
[26:30] ChatGPT's voice mode got new voices and better accents, and it should it should stop saying checking so much.
[26:40] Oh, that's hilarious.
[26:41] OpenAI's update.
[26:43] Gptlive1.1 adds voices in French, German, Hindi, Mexican Mexican Spanish, Korean, and Japanese.
[26:52] Non English speakers finally get voices that sound local, not translated.
[26:57] OpenAI says the update brings better accents and less checking chatter while it searches.
[27:03] It's rolling out to every ChatGPT user starting Thursday, uh, yesterday, right in the voice button.
[27:11] But staged rollouts mean some people won't see the new voices for a few days.
[27:16] This is great.
[27:17] The the accent thing is really good, really interesting.
[27:21] The big piece that I love about this is the checking thing.
[27:24] If you've used ChatGPT voice, I love it.
[27:27] It's very, you know, very intuitive.
[27:29] It feels like you're talking to a human in in a in a lot of cases.
[27:34] Um, sometimes I put it on in the car, you know, where I'm driving somewhere just to brainstorm.
[27:38] I'm I'm trying to think about ideas, and it's awesome.
[27:41] Like, you've got an assistant there.
[27:42] Like, you go research everything.
[27:44] The most frustrating thing, and I'll open every chat with this and tell it to to turn this off, but you you ask it a question, and it'll go, I don't know.
[27:53] Let me check on that.
[27:54] You know?
[27:54] It's like, dude, you don't need to preface every single statement with, you're thinking about it.
[28:00] Just say something or don't say something.
[28:02] You know?
[28:03] It's it's really funny, but it tries to fill in kind of the the empty space, and And I have to tell it every time, don't say or preface any statements before you say them.
[28:13] That being said, it sounds like they fixed that.
[28:15] So maybe this is a...
[28:17] This is something that was a common issue, and they've they've worked that out.
[28:21] And then more accents, which is interesting because I saw a a competing voice model earlier this week that it sounded like they they dialed accents in, um, really well, where it was in the yeast case that I saw, it was a...
[28:38] The the model was speaking Japanese with a Chinese accent, apparently, which is really crazy.
[28:45] Crazy to think about as, you know, you know, primarily, we're just using, uh, the the English versions of the models.
[28:54] Anyways, uh, interesting.
[28:56] Love to see the developments in the voice models.
[28:58] I think, um, voice is gonna be a really predominant, you know, mode of communicating with our AIs and seeing how, uh, how...
[29:12] You know, really how to create a personal interaction with your AI agents is probably gonna be through voice.
[29:20] And it'll be interesting to see as it gets better and better how that works, uh, you know, across the world, not not just in American accents.
[29:30] And then fixing that, uh, that bug where it's constantly thinking about things.
[29:35] Do you guys talk out loud to your agents, or are you typing the majority of the time?
[29:43] Are you using WhisperFlow to dictate?
[29:46] How much is voice if you had to give it a number, a percentage number?
[29:50] How much are you talking to AI versus typing to it?
[29:55] I'd say for me, it's actually probably less than 50% of the time I'm, uh, actually talking to the agents, but it's much faster, uh, when I can dictate or have a a voice conversation and brainstorm.
[30:09] I really love that feature.
[30:12] I just don't use it a ton, uh, as sometimes it's more efficient and succinct to type.
[30:19] Uh, that being said, the the, uh, the dictation using Whisper Flow is awesome when you're in a private office.
[30:28] Great.
[30:29] So, yeah, let me know what you guys think about voice agents.
[30:32] Do they sound natural to you too?
[30:34] Have you guys...
[30:35] Do you have an issue with them kinda sounding robotic?
[30:39] Curious to hear your experience.
[30:43] Well, that brings us to the end of the show today.
[30:45] What an awesome week.
[30:46] We covered lots of AI news and lots of breaking developments every day.
[30:53] I think this is a really important time to be paying attention to what's going on in the technology space, to understand it, to have a grasp of it, both from the work that we do in our day to day lives, but how this technology will impact us in the future for the better.
[31:09] And I think it's a really exciting time to get to be in on these developments and and have a lot of fun doing it.
[31:16] I thank everybody for tuning in and watching the show.
[31:19] Pre...
[31:19] Please, um, uh, give me your perspective.
[31:23] Uh, give me your feedback.
[31:24] I love to hear it and love to look for ways that we can improve the show, make it better and better, uh, every single day.
[31:33] You can find us on taws.live.
[31:35] That's taws.live or taWS daily at taWS daily on, uh, social media, Instagram, Facebook, x, and YouTube.
[31:48] And then we are the Austin Wilson Show on Spotify.
[31:52] Be sure to tune in here live every weekday morning where we're covering AI news, real context, and I'm your human in the loop.
[32:01] Austin Wilson, we will see you Monday morning.
[32:08] Let's go.