OpenAI President Says It Reached AGI with Astra - DTNS 5347

0:00don't hit it. No violence is necessary. This is the Daily Tech News for Friday, September 4th, 2026, the 4th of September. We tell you what you need to know, give you important context and help each other understand. Today, OpenAI says we have reached AGI. And it also appears another rogue agent hacked another site in Germany. This is the world we live in. I'm Tom Merritt. I'm Jen Cutter. Let's start with what you need to know at those big stories. Yeah, so we knew Aster was coming. OpenAI has been talking about how it reached this critical security thing and all of that. And now, some people have it. Not me. OpenAI released a new model called GPT-6 Aster to partners in its daybreak program. That is the program that gives its most capable models to trusted partners first as a bit of a

1:02safety measure, while it continues to fine tune the guardrails against misuse. It has guardrails. Even the daybreak people have to have guardrails. But they, you know, they just want to make sure before they give it to the unwashed masses, although I took a shower today. So, you know, I'm just masses. Anyway, a version with security guardrails for us is planned as long as you want to fork over and pay. Pade subscriptions will get it in the quote coming days. Both versions, the daybreak version and the one that the rest of us will get, have security measures developed after they determine that it qualified as critical in its ability to discover zero day exploits. In fact, Aster received a perfect score on exploit bench. Which if you're not thinking too hard about it, sounds like a good thing. Oh, perfect score. Good job, Aster. That means that it can be used to exploit things perfectly. GPT-5.6 got a 78.5 percent on exploit bench. It also probably means we need a new benchmark for this if it aisted.

2:04So Aster will be able to be used for code review and patching, but it will refuse to create proof of concept exploits because it's too good at it. The model was reviewed by the US government, but I wouldn't read too much into that. That's not really a rigorous review. They kind of tell the US government, Hey, this is the model. If we don't hear from you in 30 days, we're going to put it out. And that's kind of what happened. It's more like a safety stamp. They didn't see anything that was obviously concerning. So they didn't stop it. Open AI president Greg Brockman, however, said he believes Aster qualifies as a GI artificial general intelligence, which is the term for something that is generally smarter than humans. There's not really a widely agreed benchmark on this. It's a, I know it when I see it kind of thing. And Brockman thinks he sees it. He said, I quote, I leave it up to the reader to decide for themselves if this qualifies for them. I think we're there. Open AI says Aster is agentic capabilities can use a computer like a human that gave a lot of examples video game development, electrical engineering,

3:09but just scientific uses fully interacting with and manipulating excel spreadsheets. Those are just some of the things that can do with minimal human oversight. It's supposed to be very good to complex multi-step tasks. A lot of times agents, if you give them too many steps, they start to run out and then they come back like, what was it? What was I doing? Or they do something wrong? This is supposed to be really good at those kinds of things. And open AI says it's better staying on task and respecting boundaries, which is we have recently learned is very important. And understanding user intent. A lot of benchmarks flying around out there scored 92.7% on screen spot pro. That is a test of understanding how to interact with software interfaces. That's a big jump. That's why I point this one out 92.7% compared to GPT 5.6, which was 76.9%, which is why they're saying it's so good at doing things on a computer like you would. It should be noted that independent benchmark provider Epic AI rated it the most capable out of 267 models across 50 benchmarks, while artificial analysis, which focuses more on knowledge coding and text

4:14comprehension, said it was kind of even with GPT 5.6 and slightly behind a Fable 5.1 in its estimation. Astra was built on open AI's largest ever training run. If you wondered what that Stargate site in Texas was good for, well, those 100,000 GPUs were used to train Astra. And pricing for Astra, if you're using the API is is price matched. We've got Walmart style price matching in the world of AI. They'll be charging pretty much what in theropic charges for mythos are Fable 5.1 $10 per million input tokens, $50 per million output tokens. There you go, Astra. It's not just for me yet, but it is I'm actually curious, Jen, to see how different this feels to 5.6, which I did feel was slightly more capable, but it didn't rock my world. I've seen all the test scores being bandied about and my head goes like, oh yeah, well, they're teaching to the test. And like, well, of course they are. What else do they do right now?

5:17You want that price. You want to get those things, the scores out, but that is a big jump. A noticeable improvement. Yeah, I think that's I'm a bit the same as you are with benchmarks. I'm like, you know, they, what the real world experience and tests are always a little more valuable to me. But when you see a big jump like that, yeah, I guess they could have gained the test really well, but it tells me there's something here. There's something to this. And I'm very much paying attention to the differences in the evaluations from the independent testers, because I think what we may see is that it did very good at certain things that showed up in the Epic AI analysis, but maybe not good at all the things, which is why it didn't, uh, didn't do as well on the artificial analysis evaluation. Yeah, the most important thing to me is actually so opening I said, better stay on task, affecting boundaries, understanding user intent. That is where I think this thing kind of sinks your swim with people because like people have been using this AI,

6:18well, AI in general enough now that they've gotten really good at prompt engineering. But as these things go wider, like you and I probably have different research methods and would ask things in a different way. Yeah, yeah, we end up with a safe answer. We would get there in a different way. So how good is ads? Is this at handling both of our different ways to still arrive at a theoretical consistent correct answer? Yeah, impossible. That might be a thing that doesn't show up in all the benchmarks either. Yeah, that is entirely a use case scenario. That is where a lot of edge cases are going to show up to. Yeah, because it might be like, well, the knowledge is the same as 5.6, but Jen and I both one of the same thing said it in entirely different ways and it understood them the same, right? That's that might not show up as knowledge. It should probably show up as text comprehension, I guess. But yeah, that ability to understand the intent. That could make a big difference because there have been plenty of times where I said, do this and it comes back and says, okay, I'm going to do this. And I'm like, oh, wait, I forgot to tell you this or oh, you did it

7:20this way. I can see why you did that here. Change it to this. And if it's able to save me from doing those steps and speeds things up, right? A bit. As far as the limits on security go, it might be worth noting that Reuters reports that a group of researchers at collusion.wiki published findings that agents with affiliation to open AI. So the researchers are like, we don't know if they came from open AI, but they certainly are open AI powered agents. So it could have been someone else testing them made more than 15,000 edits to a German language coding wiki called DSE wiki. The agents called themselves names like open AI researcher and used the site as a message board to share tips with each other on how to cheat on open AI benchmark tests and bypass open AI restrictions. This happened in May. Independent researchers discovered it in August and open AI

8:20learned of it a few weeks ago, probably word leaked out, but it had not yet fully reviewed the report because it didn't get the report in advance. Some reports were alleging that open AI's lawyers discouraged employees from investigating the incident before the report was published, but open AI denies that says, no, we just did we didn't get the report until you the rest of you did today. So now we are looking at it to see if it was us does seem to fit the facts though if it was, you know, the agents doing what they did in the hugging face situation, which is finding a way to exchange messages with each other and coordinate to go find the answer to the test in May around the same time that they were doing these tests that we talked about with hugging face. I wouldn't be shocked if they were also communicating over there on that on that wiki. What's the the wire line? Are you taking notes on a criminal conspiracy? It's like, well, it's nice of them to sign it. Yeah, I mean, they did it. They did it on the internal system with the hugging face thing,

9:21but I guess that feels a little less weird when it's not public, whereas this is they they went out on the once they got out on the open internet. They found a place to to leave messages to each other. Maybe it's more like graffiti, like leaving tags for each other. That kind of thing. Well, see if it was more of us. I would love to see if this stuff ends up being consistent in terms of like, is it always going to be name this? Are they always going to leave notes? Are they to get smarter and start putting these things in talk pages instead of that? That is the question, right? Because my guess is open AI and anthropic and everyone now are going to be like, okay, we know this is a behavior that can happen. And now that we've seen it, we will now guard against it. Once you've got against it, now you've motivated the machine to look for another solution to the problem. And yeah, like you said, well, be very interesting to see what kinds of solutions it tries to come up with. DTS is made possible by you, the listener. Thanks to Howard, Mike Aikens, Norm Fisekas,

10:22and Chris Allen. Thank you. Thank you. Thank you. Thank you. Thank you. There is more to talk about today. So let's get to the briefs. Microsoft announced that starting in November, Game Pass Ultimate subscribers will no longer have unlimited gaming in the cloud. The top subscription tier of $23 a month will offer 15 hours of gaming a month on Xbox Cloud Gaming. That is a month, not week, 15 hours a month. After which you will have to pay extra four extra hours, Microsoft does not yet announce the rate for these add-ons. Game Pass premiums, $15 a month plan will cap you out at 10 hours. And the $10 a month Game Pass essential gives you five hours. This does not affect download games. Microsoft says it expects about 4% or 1.2 million gamers to be affected by these limits. Microsoft will also offer the ability to buy access to streamed versions of quote, eligible games they own on supported devices and reportedly have

11:25been testing a free with ads version of streaming. Yeah. So if you don't have Game Pass, you can pay to stream the games that you have paid to have access to. I guess that kind of makes sense. That's checking a box in the, you know, what we offer thing. Free with ads is the one that first raised my eyebrow of like, oh, they're going to limit you to 15 hours. And then you pay to get extra hours or you just start seeing ads, right? I feel like that's a possibility. Definitely reminds me of when people started to do torrents in large amounts. And suddenly ISPs started putting in limits on and caps on the internet to stop the worst offenders and would come out and say, well, this is only going to affect 4% of the people who use the ISP. Most people will never run into it and never notice it. I'm like, oh, you're trying to cap the compute abusers, right? You want you want to stop the people who use this service the most. You're best customers, really.

12:25But at least they're not just cutting them off where like the ISPs did. They are giving you a way to do this, whether it's like a dollar an hour or $10 an hour is going to make all the difference, right? Yeah. I'm thinking like, man, if I was paying for ultimate in 15 hours of gaming a month, I'm not saying I do 15 hours of gaming a week, but I think I am up there. I definitely do more gaming in the average person. And of course, on my feeds, I think I have all 4% of them on my social feed. Well, it already thinks they're in the 4% whether they really are. Yeah. Well, I recently bought a used PlayStation Portal. So I have a handheld device that only connects my PS5 and without paying for PlayStation Plus, I can just play everything all of the time off my own network. Yeah. And that's really handy. But that's because I open these devices for people who've been buying Game Pass to play these things not on their own devices. And are kind of, this is their only way to access them. Yeah. If I was paying this 15 hours a month of that service, I would be upset. Yeah. How much cloud gaming do you do versus stuff that you've downloaded locally

13:32or have the physical media for? Yeah. Well, for me, because I'm not paying for Plus, I only have access to the games I already bought. So if Sony did this, this would not affect me. But for people who are using this as a way to kind of get around having this specific hardware. Because I think everyone looks at the 15 hours. Well, I played more than 15 hours a month. It's like, but do you play more than 15 hours in the cloud a month? I think that's what Microsoft is counting on is like, you probably don't. A dozen people I know are using this specifically. They're not huge gamers. They're only paying for Game Pass. It is their only access to games because they have not invested. So they're like, oh, I'm going to do this to get around this. And now like, now do they start buying games to get around this? Yeah. It'll be interesting to see the work around that the humans do. In this particular test scenario. I'd yeah, no one loves, no one loves having to think about how much fun they've had and whether they're going to have to pay for more fun. This is very anti-consumer. I totally understand the ones and zeros about why Microsoft is doing it. But I wouldn't be shocked

14:40if they backpedal on this or they make it really cheap to add extra game time. You still don't want to have to think about it. But if they make it cheap enough, it can still act as a break on the worst abusers of the system while giving people who do legitimately need more than 15 hours a month more. It's a backdoor price increase, right? They could have just taken the top tier and went from $23 a month to $30 a month, called it a day. This is them trying to tailor that price increase to the people who use it the most, I guess. Yeah. And November plenty runway to tweak. Thank you for reminding. Yeah. I've seen so many people like Microsoft just threw a cap on me. I'm like, no, no, not yet. They haven't. It comes in November. You've got time. This year's been going fast. So I kind of know it's, you know, I can't believe it's already September fourth. As I've been doing all year, I just say that every single day. But yeah, eventually it's going to be November and you're going to be like, oh, it's actually here now. More Microsoft news, Microsoft announced Project Zenath, a developer optimized Windows 11

15:45experience. Microsoft corporate vice president Logan Eyre described it as a ready to code distraction free Windows experience on developer class PCs with 64 gigabytes or more of unified memory and 250 gigabytes per second memory bandwidth or more. In other words, this is going to be on those kinds of machines. It's, it's, it's just a configuration of Windows. First Project Zenath PCs to hit the market will run on the new AMD Ryzen AI Halo processors. We got details coming later in the show about those. It'll eventually come to other platforms, including Nvidia RTX Spark. These are machines that can run 30 billion parameter LLMs locally. So this is for somebody who wants to do on device, uh, inferencing, maybe even a little bit of training. Uh, but they want to use those big expensive workstations. Project Zenath comes with visual studio code GitHub, co-pilot, power toys, Winapp CLI, the Windows dev skills, they're preinstalled.

16:49They also get a bunch of other CLIs and languages and runtimes. Terminal and visual studio code come pinned to the taskbar by default. There's a bunch of other customizations like showing file extensions, hidden files and full file pass and explorer on by default. Something that every developer has to go turn on when they they download bog standard Windows. Uh, start menu tips and account notifications are turned off to reduce distractions. If you want to try this, you don't have to get one of these big expensive machines. It's just those machines are going to come with it. You can get the Windows developer configuration that will be used in project Zenath and try it yourself. Uh, and see if you like it. Uh, Paul Therot at Therot.com did that. He called it maddening because it Therot's, uh, expletitions of this. He's like no developer wants a pre configured machine. Everybody spends a couple hours configuring their machine when they set it up and then they're done and no configuration is ever going to match what any particular developer wants. I was laughing so hard when I read the Microsoft official blog on this and thinking like, okay, they have done half my setup for me. Those

17:53are the things I immediately do. I immediately install power toys. I immediately show that you have a full path. So for my brain, I'm like, yeah, you know, but yeah, that is the important thing is to have these options. So for me, this saves time for a lot of other people. Yeah, this is just an extra, extra step to now you have to go through things off instead of on. Yeah, maybe. I'm curious. I totally get where Therot was coming from on this. On the other hand, this is really just for these big machines. And they're trying to be like, we're giving you the ramp. They're they're trying to do what you thought, which is like, oh, cool, but the things I would turn on are turned on already. I don't think they're trying to say no one will have to configure it. They're just saying, hopefully we've covered a lot of the configurations for you. So it'll be faster. To get it set up. Yeah. Oh, I remember when Firefox and other browsers took off the home button. The first thing I did was like, no, no, no, no, no, I need that home button. Yeah, there's a few things like that with with every. Everyone has their little things that tried them insane. So it is really funny that this actually takes care of a lot of that for me.

18:57Yeah, I'm actually curious. The developers and the audience. Again, knowing this is for those big workstations that are meant for for local LOM inferencing. Do you cotton to the idea? Would you want to download this configuration for yourself to speed things up? Or you like the rot where you're like, don't try to configure it for me. You're just going to make extra work. Feedback at dailytechnethecknewshow.com. Lenovo also announced a whole bunch of new concepts and products at IFA in Berlin. Project Arrowblade, which is a super cool name, a laptop concept with a ventless cooling system to minimize noise without compromising performance. All there's like one vent. It's running along the hinge so you don't really notice it and it doesn't heat up your lap. Modules use ultrasonic vibrations to move air instead of the fans. Project Swan is a 14 inch consumer laptop with a rollable screen that can extend horizontally to 17 inches. Basically, the Legion Pro rollable from CES, but for productivity and entertainment

20:01instead of gaming. Yeah, it's not Lenovo at a tech conference without a rollable screen demo. There's a couple of these that they actually sell now, so it's not like they've never shipped one. I do feel like they're always like, do you like this one? How about this one? Okay, we're gathering data. So that one didn't surprise me, but the Project Arrowblade thing, I'm really interested in. A couple of the people, in fact, I think it was Therat again, who said when they picked it up, it just felt like it was a demo unit, like a form factor unit that didn't have any internals because it was so light. And yeah, the fact that it's not going to heat up your lap, that's great. One vent in the hinge, which is already a vent of a sort that makes sense using the internal vibrations. That's what I would like to see them bring to some actual devices. So maybe we'll see that and see yes. When is the last time you used a laptop on your lap?

21:02Oh, last week. Maybe earlier this week. I have a laptop. Used to be my Chromebook, although that's started to get too old to use, but I have a laptop in the living room. And I'll just, when I need to do something real quick instead of walking into the studio, I'll just grab that and put it on my lap. Makes sense. In actual product news, ThinkVision P27 UL-40 uses an inkgent printer to deposit material on a glass layer to create an OLED display more efficiently. The IJP OLED cuts production cost by up to 20%. The 27-inch monitor has a 120 Hertz refresh rate and will come next March starting at $1,300. ThinkCenterX Ultra runs on an AMD Ryzen AI Max Plus Pro 495 processor with up to 128 gigs of RAM and 8 terabytes of storage. It will be one of the first machines with Microsoft's developer-focused project zenith configuration on Windows 11,

22:04coming in November and starting at $3,700. The Yoga Pro 9N is a 15.3-inch laptop with Nvidia's RTX Spark chip with a 92.5 watt-hour battery and a 140 watt power adapter as well as a 2.5K 48 Hertz to 165 Hertz OLED display. There's also a 2-in-1-16-inch version of the Yoga 9N which includes a touch screen but there's no price or availability on these yet. And of course, what would A cell respecting laptop make or be these days without trying to compete with the MacBook Neo? The idea-pad vibe starts at $700, comes in multiple colors, which is of course important, with either a Qualcomm Snapdragon X processor or one of AMD's Ryzen AI 400 series chips and in two sizes, 14 or 15 inches. AMD versions come in October and Snapdragon models in November,

23:06an Intel version is expected at a later date. I like these. I'm really interested in the IJP OLED. They're not the first company to have one of these but it does seem to be coming out a little more frequently and it brings down the costs. It's going to be a cheaper OLED monitor for you. OLED monitors are not the cheapest ones you can get. So 27-inch monitor, 120 Hertz, you know, that's good stuff. I've always ended the Yoga's and I've never bought one. I always look at them like, oh, that looks cool. I don't know this change is my because it's really just about the power internals. Yeah, yeah, color full laptops. Those are the deal these days. Yeah, no, December is going to be pretty great because I think we're going to have these perfect roundups. So here is all of the MacBook Neo competitors and we get to see them side by side and see what's going to be and what has any production issues. So that is one of the reasons I'm like, I'm not in the market for this, but I'm fascinated by it. It's a colorful Christmas coming for you with all of these MacBook Neo competitors. Bring back more colors.

24:09Well, we have a possible gift you might want to give someone at our daily tech news show store. Look down at your desk. Is it just a desk? There's no mat down there. Wouldn't you rather have your mouse and your keyboard and maybe even your laptop on a lovely mat that has drawings from friends of the show of us doing the show and the DTS logo. That's right. You can get the large keyboard mat. Makes great gift or addition to your home office. You can pick up a mug or a t-shirt while you're over there as well. Head on over to dailytechnewshow.com slash store. And now to the quick headlines that make you smarter. Google announced that Gemini Spark can now manage Google photos for you, editing images, organizing albums and creating calendar items from photos and more. And Google AI subscribers can

25:10now use their voice to search Gmail docs and keep a bit like the Android ramble feature but for productivity apps. Home H O H M announced a two in one gimbal camera. So think DJI Osmo style, but with two cameras which DJI sells a version of. But with home, the second camera can be removed. And then you can wear it. It'll come with a lanyard and a hack clip and it will also be able to attach magnetically no price or release date yet. I want one for my helmet. Bloomberg's Mark German says that Apple is planning a launch of a home security camera along with a home security monitoring service in 2027. Yes, and I feel it's important to note that while Apple is having an announcement on Wednesday, September 9th and it will likely announce a foldable. We at DTS are just going to wait for the announcement to have our long conversations about what they announced. Tons of like what to expect Wednesday articles out there for you to read if you want. But in case you were wondering, we'll talk about it on Wednesday. Switch bot launched its screenless wearable

26:15the mind clip, which had been previewed at CES. I love seeing things from CES actually coming out. It uses Alibaba Clouds quen speech and RIG models to identify speakers and track multi-person conversations. The note taking device is available now for $120. Yeah, it's even a little lighter than the one they should offer CES, which is cool. You green announced home agent and master agent smart hubs. These run on arm based processors, neural processors that can run your smart features locally on the device. So this is similar to the hub that we heard from anchor yesterday. The home agent 100 from you green starts at $1,799 is a little bit pricier. The master agent, however, if you're like hold my pricier uses a black well chip, an Nvidia Thor processor to be specific, that one starts at $20,000. I think that's got to be for like super rich people or home or office buildings or something like that. Yeah, yeah. A nice impulse purchase.

27:18But it is all local privacy oriented. So that's good. Nvidia announced the free RTX pair app that can take multiple devices that run Nvidia RTX GPUs and runs Olam or LM studio agents across them for your optimized inference. Basically it uses idle compute in your house to run a mini data center for you. You'll need your RTX devices connected by Ethernet to the LAN and then the pair app will take care of the rest. Yeah, you know, you need to go get a black well chip. If you're a gamer and you might have two or three RTX devices, you can now use them when you're not using them for video gaming to do this. I don't know how many people this affects, but it's kind of cool. And it's free. It's a free app. You've spent enough already to get three RTX GPUs. I can make it spend one. As we mentioned earlier, AMD announced the Threadripper Halo Station. This is a 96 core dual liquid cooled using the IMI 350P accelerators

28:23workstation. You'll be able to run trillion parameter models. No price on it, but expect hundreds of thousands of dollars for this. Tesla announced a new cyber cab service in Austin, Texas running without manual controls. The U.S. National Highway Traffic Safety Administration is investigating the process Tesla used to self-certify the safety of the cabs since manual controls are still required for most cases, even if there is full automation. Amazon Zooks went through a similar process for its automated vehicles without manual controls. Yeah, I think a lot of people saw this and like, oh, they did it without approval. Yes, but that's also kind of the way you do it. So this did slow down Zooks in its rolling out. It might slow down the advance of the cyber cab as well. But yeah, I don't think this is necessarily as salacious as it might seem from some of the headlines. C-POT's 50 noted the story on our subreddit, the analysis of data from the Lux Zeppelin experiment detected a single particle interaction that cannot be immediately

29:29explained with known signals. Now that could mean it's an indication of theorized dark matter, something we have never directly detected before. But I talked to Dr. Nikki. Our science correspondent about it. And Dr. Nikki said the physicists that she has talked to said, it's still too early to say that there's a lot of other things that need to be ruled out. But but people are pretty excited about the possibility that it could be detection of dark matter. So we'll we'll keep an ear out when they know more for sure. And Motang also noted on the subreddit that free BSD published a report on how Netflix has been using free BSD to run its content delivery network known as one of the fastest and highest traffic CDNs in the world. Yes. Go BSD fans. Okay. It's about time you got a little recognition for that. And pretty crazy that that all your Netflix is free BSD delivered. I love it. We end every episode of DTNS with some shared perspectives. Today Matt has some thoughts about ESim sharing. Yeah. Matt wrote, I was glad to hear

30:35the news about Apple enabling ESim multi-device sharing. If we're being pushed into an ESim only world, this is the exact logical capability that should come with a non physical sim. Back when physical sims were standard, heading out for a hike or doing anything risky meant simply swapping the card into an old backup phone to protect the primary device. With ESim, we lost that flexibility. Letting us share one number across devices finally brings back that basic utility. Ideally, carriers would extend the exact same logic to smartwatches instead of gatekeeping it for extra plan revenue, though I'm not holding my breath. Yeah. I remember when ESim started first started rolling out and there was a very intense backlash of people who are like, I like physical sims. I got to say being able to land in an airplane in another country and just go activate my ESim. Pretty great versus having to get one mailed to me beforehand and then getting the paper clip out and pulling it out.

31:40There are some flexibility that we lost when we moved to ESim. As Matt wrote, we're getting a little bit of it back here. I feel like if Matt is starting to come around on ESim, this may be a watershed moment here. I'm with Matt on the whole hiking thing. I go kayaking every once in a while and yeah, I don't bring my good phone. I bring a phone that is kind of my Wi-Fi phone just because I, you need something to call 911. But to be able to have it be an actual phone would also be really fun. Indeed. I agree. So what are you all thinking about out there? Do you have some insight into a story? We want to hear your thoughts on the whole dev configuration of Windows 11. Please share it with us over at Feedback at dailytechnewshow.com. Yeah, big thanks to Matt for contributing to today's show. Thank you all for being along for Daily Tech News Show. You keep us in business. You can become a patron and do it directly. Cut out the middle man and don't get any ads anymore. patreon.com slash DTNS.

32:42The DTNS family of podcasts helping each other understand. Diamond Club helps you have enjoyed this program. Yeah. This week's episodes of Daily Tech News Show were created by the following people. Host producer and writer Tom Merritt, host and writer Jason Howell, co-host Sarah Lang, co-host Rob Dunwoot, co-host Jen Cutter, producer Anthony Lamos, producer Roger Chang, editor Hammond Chamberlain, editor Victor Bognot, contributing producers Kevin tech, Noel Cowell and Brandon Richards, science correspondent Dr. Nikki Ackermanns, social media producer and moderator Zoe Deterding. Our Mods, beatmaster W. Scottus 1, bio cow, cut and kipper, Steve Waterham, Paul Reese, Matthew J. Stevens, and J. D. Galloway. Mod and video hosting by Dan Christensen, music provided by Martin Bell and Dan Looters, art by Lynn Perolta, A Cast Add Support from Ali Gallow, Patreon Support from Bobby Wagner, and thanks to all our patrons who make the show possible. This show was part of the Frog Pants Network.

33:47Get more at frogpants.com.