In short
The episode covers (1) Netflix’s open-source VOID video model for object removal with physics correction, (2) rumors and examples of improved “GPT Image 2” text/image rendering, and (3) agentic AI tools and access limits (C-Dance 2.0, Google Gemma 4, and OpenClaw/Claude).
Guests
Addy and Joey (co-hosts; both discuss camera/VFX details and AI tooling).
Key claims
VOID removes objects and also updates interactions so remaining objects behave correctly (hands removed, top stays; weight removed, pillow stops sinking; dominoes removed, final domino doesn’t fall; person removed, blender still off). VOID uses a two-pass method with flow-warped noise for stabilization/cleanup.
Notable examples
comparisons to Apollo/blue marble and Artemis camera gear; “YouTube UI” screenshots allegedly generated by GPT Image 2; OpenClaw revoked OAuth, forcing per-usage credits; C-Dance 2.0 rejects any face input and caps output length/resolution.
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Chapters
Tap a time to open that second in VOArtemis Moon Missions and Imaging Technology
0:45 to 4:28
A discussion about the Artemis missions, camera technology used, and the surprise role of GoPro and Nikon.
“Yeah, their main camera was the D5, which at first when they met, they kept saying the D5.”
The Pseudoscience of Space Writing
4:28 to 5:28
Exploring the myth and reality of using pens vs. pencils in space, including safety concerns.
The Evolution of Video Technologies
5:28 to 7:16
Discussion about the differences in video quality and imagery over time, including comparisons of iconic images.
“The warp noise thing is also their own white paper.”
Netflix's VOID Model Explained
7:16 to 11:58
In-depth analysis of Netflix's VOID model, its capabilities in object removal and physics correction in videos.
Open Source AI Video Editing
11:58 to 14:00
Discussion on the implications of Netflix's open source model for the future of video editing and collaboration.
“and they would add, like, hey, Linux doesn't have a clock.”
Open Source Models Empowering Innovations
14:00 to 14:50
Explore how open source frameworks enable advancements in AI models like Netflix's VOID.
“Yeah, I like to ship it out in the world and see what happens.”
GPT Image 2 and Its Stunning Capabilities
14:50 to 18:10
Learn about the new GPT Image 2 model and its impressive text and graphic rendering features.
“Somebody will build something else that's open source on top of this model.”
Realism in AI-Generated Images
18:10 to 19:50
Discuss the realism of AI-generated images and the nuances that make them believable.
AI Art's Impact on Society
19:50 to 23:30
Examine the societal effects of AI-generated content, especially in politically sensitive contexts.
“I mean, my sort of like simpleton brain is thinking like, this is just like a template, like a YouTube template.”
C-Dance 2.0 Rollout Challenges
23:30 to 26:10
Delve into the issues surrounding the rollout of C-Dance 2.0 and its restrictive features.
“It was like, this is the most reverse rollout.”
Show all 19 chapters
Google's Gemma 4 and AI Advancements
26:10 to 28:00
Discover Google's latest open-source model, Gemma 4, and its potential applications.
“I mean, do you remember when the Tom Cruise Brad Pitt video came out?”
Google's Gemma 4 and Local AI Capabilities
28:00 to 29:00
Learn about Google's latest open-source model, Gemma 4, and its potential uses.
“So it's an open-source distilled version of Gemini 3 that can run a couple flavors depending on your memory that can run locally.”
Apple's Potential Use of AI Models
29:00 to 30:14
Discuss theories on Apple's use of AI models like Gemma for Siri and app integration.
“can do that much, but it would probably be good for just on-the-phone stuff.”
Challenges with Siri and AI Efficiency
30:14 to 31:48
Explore the limitations of Siri and the desire for improved AI efficiency in apps.
OpenClaw's OAuth Controversy
31:48 to 33:08
Discuss the recent changes to OpenClaw's OAuth access and user frustrations.
“You also probably have, I don't know, 100 apps on your phone, right?”
Fragmentation in AI Tools and User Experience
33:08 to 35:14
Analyze the fragmentation of AI tools and the impact on user experience.
“Is it just called computer or machine or something?”
The Future of AI in Task Management
35:14 to 38:20
Speculate on the future of AI-driven task management and virtual assistants.
“What do you think is the near-term future of OpenClaw and sort of at-home agentic capability?”
Unifying Knowledge Across AI Tools
38:20 to 39:28
Discuss the challenges of unifying knowledge across multiple AI tools.
“And because it's already kind of connected to stuff, it does that.”
Interactive Avatars and Future AI Developments
39:28 to 40:18
Explore new developments in interactive avatars for AI and virtual meetings.
“And hopefully some of our viewers are also living vicariously through you.”
Transcript
Automatic transcript. May contain errors.0:00Joey:Boomers are really cooked now. It's so hard to tell what's real and fake. And if you think you can, I think you're overrating your abilities.
0:14Joey:All right, welcome back to Denoised. Addy, how you doing? I'm doing good. How are you, Joey? I'm good, man. Did you watch any of the Artemis going around the moon on Monday?
0:23Addy:I'm a nerd, so yeah, I've been following all that stuff. All right.
0:27Joey:Okay, you too. All right. And then when they kept talking about how they trained to photograph the moon, and they did simulations, and also that they kept talking about GoPro. And I kept thinking, this is the best advertising GoPro has had. For a while. And Nikon, too.
0:43Addy:The main commander was like, I only use Nikon Z9s. I was like, that's pretty specific.
0:48Joey:Yeah, their main camera was the D5, which at first when they met, they kept saying the D5. And I was like, are they mixing up 5D? Is it the Canon? And then I was like, no, it's the D5.
0:55Addy:D5's their old school.
0:56Joey:10-year old camera. Yeah. Yeah. The Zs are the mirrorless. Yes. And they did bring a Z. And then apparently, well, I'm thinking because it takes forever for stuff to get like space certified. So it was probably like the D5 was already certified and really good at low light. That's right. But they said that they were to convince them to bring the newer cameras as well.
1:14Addy:The whole point of a spacecraft is to make things like habitable inside of a space. Like, can't you just bring anything? Am I too simple minded to understand that?
1:27Joey:There was some story of, I don't know if it was like the original NASA days, and I also don't know how true the story was, but that it was kind of a joke where it was like NASA spent$50 ,000 to develop a pen that could write pressurized. So they could write upside down and in space and like underwater. And then the joke was like, the Russians just use a pencil. And that was a joke. But then there was clarity where it was like, you can't actually use a pencil in space because it like leaks graphite. And then the graphite particles will like float around and like get into stuff. It'll short circuit the switchboards.
1:56Joey:yeah it was like a major yeah so that kind of story is why stuff has to be vetted to bring
2:04Addy:into space is my guess yeah i just think they overthink it just send that up just take a gopro just take why did they take a it's the 360 take an iphone yeah they did have iphone that was also another good iphone commercial but i thought it was funny because the thing they kept dealing
2:18Joey:with on the radio was they kept having glare in the window and they were like we need to put shirts on the window and i was thinking like they're like those little hood things for 15 bucks that like suction clump to the window and you put the lens in it so a polarizer filter types the glare i mean i'm sure these things already have polarizers on them but i was like glare glare is the biggest issue and out of all the stuff they thought about and they couldn't get the little hood thing but they look amazing so they figured it out they look amazing yeah and
2:48Addy:uh they they i was watching a video on youtube where they were comparing the blue marble from 1972 with today's blue marble yeah and the internet is going crazy because today's blue
2:57Joey:marble looks more polluted and yellow and gunky i saw one that was more commentary on like comparing that to how the film frames and how like everything is just like less saturated and flatter today it's like the netflix look yeah on artemis yes let me try to find this image yeah i mean my thought was the image we were looking at from apollo that was shot on film brought back to earth processed right the one we're looking at instantly is the one that people are sharing that was an iphone photo transmitted over their um wireless signal so i'm guessing
3:39Addy:compressed sure sure i think the dynamic range would hurt there for sure um but like the nikon images that came back shout out to nikon i mean those are crisp so the reason that the blue marvel looked gunky and yellow is because it was a nighttime exposure yeah what you're looking at there this so the one on the right i think is the new one and the left is 1972 yes because so the reason the one that looks a little flat and gray is because that's a that's a night time exposure like that's the earth at night so you're saying they were in the shadow of the oh because i could
4:13Joey:see the sun peeking out on the other side exactly right so yeah so like that's how good our cameras
4:17Addy:have gotten that we could go to like 57 000 iso and still get this amazing image that's crazy how
4:24Joey:is it so bright if that was night if this was in the shadow of like right how is it that's how good those nikon z9s are how is it so blue oh so this was a nikon i thought this was one of the iphone
4:33Addy:this is a quarter second exposure and yeah i'm such a nerd for knowing this quarter second
4:38Joey:exposure with 57 000 iso at f4 okay all right cool thanks for explaining that that helps a lot but yeah this was the meme going around also comparing how flat devil wears prod and harry
4:51Addy:potter reboots are compared to that's funny that's funny that's pretty good it always wins with the jokes like bring back more dampers to hollywood yeah should not fire them all let's talk about actual like ai stuff that's nerding out okay so netflix came out with their first
5:08Joey:ai model open source called void video object and interaction deletion so it's basically you want to remove an object from the video that's been doable before you can like remove something but what void does is it not only removes it but it corrects the physics for this object being gone in the video wow so in this case let's see what we got here they're spinning some tops and some hands are moving the tops they remove the hands and then the top stay the same this one they have a weight on a pillow so the pillow's sinking they remove the weight but then the physics of this pillow sinking is corrected the pillow doesn't sink anymore whoa that's crazy the domino one is this one yeah like i can't rewind this video but roomba was like knocking over a couple of dominoes and then they remove like three out of five dominoes and then the physics were corrected for like the domino at the end if they remove the middle dominoes the domino at the end doesn't fall whereas in real life it did fall because of the other domino effect that's crazy pretty wild that it is like something that would be common in film where it's like i need to remove this but i also need to fix all of the physics so it makes sense so like this this shot's crazy this girl is blending something up in a blender it removes the girl but then it also keeps the
6:24Addy:blender off yeah it's doing a lot more than in painting there's definitely some reasoning going on where it's figuring out like for example what the blender is actually doing and then what it would do without the person there and then generating that and placing it and where the
6:38Joey:old blender was yeah let's see it's doing two passes voids first pass generates a physical plausible counterfactual video with the object and its interactions removed if the model detects object morphing, then an optional second pass reruns inference using flow warped noise and stabilizes and cleans it up. So it goes through twice. Yeah.
6:57Addy:The warp noise thing is also their own white paper. Remember we covered that flow. Yeah, I guess. Go with the flow.
7:03Joey:Yeah.
7:03Addy:Yes. That's cool, man. Shout out to Netflix for dropping an open source model. Yeah, that's great.
7:07Joey:That's like an actual super, that's like goes in the bucket of corridor crew, the green screen and keyer cleans up the spill of just like very practical real world things you need to clean up in a video that would cause a lot of headaches to like correct all the physics and everything and now it could just do it in one pass and you're working with real footage and
7:28Addy:material yeah very vfx friendly like it's something we all can understand and try to solve and historically all these things are hard right like chroma king well is hard and in this case not just removing object but reinserting with a similar object is also hard i'm looking at i can
7:46Joey:compare it against other models interesting yeah so there's a comparison like runway you say remove and this t-rex is knocking a t-rex over and the t-rex still falls over i'm trying to find out what the output potentials are i don't really see that of like are you of like what you could give in and
8:04Addy:way and give out i'm hearing something really interesting on that what a lot of ai filmmakers are doing now is they have an entirely different pipeline they're calling finishing pipeline okay where they're figuring out like what the upscale resolution needs to be what the color space needs to be and if there is dynamic range ups conversion and all that and that's like an entirely separate thing versus like generating the frame we tend to kind of think of those as the same because we come from physical production right like the camera sensor resolution is your resolution like everything is just below that down sample of that are you saying they like are
8:40Joey:thinking of processing their footage and then they get the like 1080 output back and then some pipeline to like up res that again yeah it's like you know there's a combination of topaz and
8:51Addy:magnifique and a bunch of other things to like get it to where it needs to go but that's like
8:56Joey:a separate portion of your pipeline are these people i mean yeah that is a deal but are they like dealing with hybrid footage or real life footage or is this more for like a fully generative
9:06Addy:pipeline i'm guessing it's more for fully generative but i don't see why it wouldn't apply to hybrid because when you put camera footage through this video void video model you're getting something way less pixels than what you're putting in so you're gonna have to go upscale
9:24Joey:again yeah but the thing with the upscale models like the more the the diffuser models and stuff like um topaz's starlight is there more for like there's they add a little weirdness if you give it real footage like they're good they're really good for yeah you have that ai output footage and you're working sort of in a full ai pipeline and then you need to like go up but they're still not the it still can add weirdness if there's like if you run it with real footage they've gotten better i mean i'm still thinking though like and i know this is more of our like yeah in production world of of if you shot you know you show it with like airy or black magic and you have your original footage with like b-raw and you give it to this mod and to this model you like as much as you can you still want to stay in that space and have that latitude because when you fix your vfx shot you still might have to do other elements might have to do some more compositing have to send it to color when it still have that color range and latitude to color it with all of the other shots that are surrounding it so it's still important like the corridor model will give you the what like what 16-bit exr outputs and stuff like it'll still you can still stay in that high-end space i'm curious what the void model is i'm trying to find an answer to that i would guess i mean you know like obviously netflix is looking to use this or is already using this in real productions so that so i am sure there's a way to put this in a pipeline and not degrade or destruct right source footage and be locked into a shot that has the physics fixed but now you're
10:57Addy:in an 8-bit hd video that you're let me ask you this why do you think they released this as an
11:05Joey:open source model and i was wondering that what would they get out of it i'm the only fan of the I think I can think of as attracting talent to come work at one of their labs.
11:18Addy:My guess is this is just one tiny aspect of an AI-native production. They're putting it out there to see how other people cobble this together or modify it and turn it into something completely different than what they're thinking. So it's getting developer time, production testing time for free if they put it out there and just kind of watch it evolve over time.
Read the full transcript
11:40Joey:Okay, so they got kind of this kernel of a thing and want to see what other people do with it so they can adapt it to their own stuff or adapt it to other internal things they're working on?
11:49Addy:Yeah, I mean, that's how Linux became what it is today is when Linus open-sourced all of that. Other developers were working on it on their own time, and they would add, like, hey, Linux doesn't have a clock. Let me add a clock feature. And now it's open-source with a clock and so on. And after decades of evolution, Linux became this really enterprise-grade-ready stuff, all because of individual micro-contributions from developers working on something open source. So I think in that same sense, you put this thing out there, and then people who are actually going to use it are going to fix and modify things that it needs, and it's just going to happen organically.
12:29Joey:I can't find an actual number for resolution just through some quick search in the actual paper. But in their conclusion and limitations in future work, future work could obtain better training data sets beyond rendering engines. The generated video links are still in the range of a few seconds and resolutions could be further improved. So it doesn't say what the resolution output is, but I'm guessing it's either 720 or 1080.
12:48Addy:If it's nothing to brag about, they're not going to say it.
12:51Joey:I mean, yeah.
12:53Addy:The resolution could be better. Yeah. And it's also probably not the concern for them. like they're trying to solve the actual problem of physics based replacement in in painting right like that is the core of the problem that they're trying to solve so like yeah resolution that's another research paper that's another thing we're working on we're going to tie those two things
13:12Joey:together the the use kits could also still be like if you're in post and you're still like you know doing rough cuts and assembly cuts uh or even just later cuts but hey we want to see if this like shot works without this thing or this physics thing happening or this person removed do a quick run you still have it in your rough cut and then just make the determination like oh okay does the shot work in the story or what whatever needs we have before then we commit and you know send it to vfx uh but you can basically like filtering out and making sure like yes we definitely want the shot we know it works and before sending stuff to vfx that that gets worked on that ends up getting
13:50Addy:cut uh because you don't need it so i love that yeah that would be really useful that that could
13:54Joey:be another use where it's like yeah we're not going to ship it with this but we just want to make sure that we're using VFX time to the best of its best capacity. Sure. Okay, cool. Yeah, I like to ship it out in the world and see what happens. I mean, the same thing happened with the corridor screen removal, where people took it and then figured, or within a few days figured out how to make it run more efficiently and did all sorts of other things with it.
14:20Addy:Yeah, and quick shout out to the open source models in general. I think the Netflix void model was built on cog video X, if I'm not mistaken. That's an open source framework that you can use. And there's another one called vase, which Alibaba makes. And if it wasn't for these open source foundations for them to build models on, this stuff would not be open source in the first place. So again, you see where I'm going with this. Netflix built this thing on top of an open source model. Somebody will build something else that's open source on top of this model. and it just gets better and better and more specific and more niche yeah yeah okay next story
14:59Joey:ready yeah gpt image 2 potentially there's some new stuff showing up on elmarina new model that was doing some crazy stuff with graphics and text rendering uh people suspect it's most likely the next version of gpt image 2 i think it's been pulled down since then but most likely something is coming i thought it would maybe be this week but maybe it's next week i am already hearing
15:22Addy:rumors of this being a nano banana killer and all that stuff i mean i've heard this rumor too here
15:28Joey:possibly better than nano banana pro killing the banana it looks like it's done really well as like this one shot youtube interface like a lot of just really like complicated screenshot type stuff and layouts and text that looks like real stuff that's crazy that's generated yeah this whole thing like screenshot that looks like a youtube screenshot was generated that's generated shot that looks like a like the most boring mall photo bath and body works yeah this was generated this map i don't know why we need this generated but i mean that's that's insane this human anatomy body thing yeah let me try to find something that actually looks there were some better ones that i saw that like someone made a it was like a fake thumbnail but in a youtube screenshot UI.
16:20Joey:And so the whole thing was generated, but it looked like, oh, they took a screenshot of like a YouTube video.
16:24Addy:That's impressive if these are true.
16:26Joey:Yeah. Uh, it had, I guess there's three models. It had three code names, masking tape, gaffer tape, and packing tape.
16:33Addy:The, the one that's most impressive to me is the Bath and Body Works image. I mean, like that is reality to me. Like there's, yeah, this looks like a iPhone pick of like a closed store. Of like somebody who's just bored at the mall. I mean, I can't wait to run my 1970s New York City street test on this.
16:50Joey:So let's go. Let's go, ChatGPT. This one, handwritten, looks like a shot of handwritten notes on a notepad. Oh, people are doing some comparisons with other models here. GPT image 1.5 looks like an AI image.
17:05Addy:Wow. Two. Wow, look at the lighting on her and the hard shot. Even the shadow from the photographer on her. Look at that. Yeah. yeah that's crazy dude yeah and the weapons details in the background yeah i don't know people look not warped yeah i mean the fence is a little unrealistic but i'm nitpicking but that is so much closer to reality than the other one flower slop is that intentional i think that's the
17:28Joey:the prompt from the person okay yeah that's their handle let me see the dirty pantry thing the that's that's not that's not really it's just somebody's phone it's from 2023 okay it's not uh This one also claims it was GPT Image 2. Now, of course, these could all be.
17:46Addy:I mean, the Obama-Trump selfie, if you go back, like the skin detail there is spectacular. And the Oval Office rendition in the back, I mean, that is the wallpaper for the Oval Office.
17:58Joey:Also, you know, we could do Nano Banana and, you know, comparisons. But A, if this is true, but B, just GPT Image 1.5 versus 2, because 1.5 still had the like waxy AI skin yes this has fixed a lot of that AI look so yeah maybe
18:20Addy:maybe it is a banana killer man maybe nano b has its days coming I don't know yeah it looks pretty
18:25Joey:solid here's not a banana pro 2 right right right and we dropped that I mean at this point we're
18:33Addy:just taking a given that text is gonna come out flawless which is crazy like just a year or two
18:39Joey:ago text was so impossible i know and now we're like oh cool the big text is sharp but like i need every text all the way in the back here to be like super sharp too yeah like on the bottles
18:49Addy:like no dude that's out of focus the other the other commentary about a lot of this stuff was
18:54Joey:just like boomers boomers are really cooked now boomers are cooked with 1.5 like with the boomers man that's not fair i think everyone's cooked like so this like is it's so hard to tell it's so hard to tell real and fake and if you think you can't i think you are over overrating
19:15Addy:your abilities yes yeah yeah you're no better than the rest of us i just like think every i'm just suspicious of everything now which just made me suspicious of every image we're so cooked as a society i mean like it's the some of the war stuff that's ai generated you and i talk about that it's
19:32Joey:crazy got the lego things coming out of the fake netanyahu stuff with the six fingers and the
19:37Addy:pockets and then um i mean the lego stuff it's funny it's entertaining and uh it's stylized
19:44Joey:but it's also like about the craziest it's literally two countries at war they're like doing diss tracks about each other with legos it's like yeah and they're trying to win like
19:54Addy:the viral meme moment of the day which is crazy like they're competing not just at a war level but at a like a social media level yeah social media yeah it's a weird timeline yeah it's what
20:09Joey:a world we live in yeah all right i can't find the one i'm thinking of but all right oh wait literally right here oh oh this is one of ours this one of ours oh that's you yeah it's a vp yeah the vp lamppost this you this well it kind of got cut off but this whole ui screenshot thing with this uh you travel to the middle no not me but this was the image this was the image i was trying to find this was someone made this image allegedly with gpt image 2 on elm arena and it was more of just showing like in one shot they made the entire it replicated the entire youtube interface and this you know if image of them like as if they're vlogging from the middle ages that Yeah, that was the one I was trying to find.
20:53Addy:I mean, my sort of like simpleton brain is thinking like, this is just like a template, like a YouTube template. Anytime your prompt has, make a YouTube, da-da-da, it just goes into the template. It's not even using AI. It's just inserting text into a template.
21:09Joey:Oh, like it already just has the YouTube UI already. I know, it's like, yeah.
21:14Addy:It's like, it's more web design than it is generation, you know?
21:17Joey:Yeah, that part of my brain is just like, I mean, okay, cool, it can replicate the entire UI, but you can also literally screenshot it and just cut out the image and put the image in there. Yeah, I mean, what would be impressive is if you're like, design me the UI for a new app that I'm trying to do.
21:35Addy:Exactly, exactly.
21:36Joey:And how well it does that.
21:38Addy:That would be impressive. And my guess is that's where Figma and who's the competitor, Canva, and some of their AI tools are going to do. Yeah.
21:46Joey:Speaking of the deep fake and brain fried, this is going to be a boon for those millionaire influencer people who like show their PayPal or their Stripe account things. Be like, look how much I make in a day. And they show their like Stripe dashboard. This is going to be instant for that. Be like, give me a Stripe dashboard where I'm making a million a day.
22:06Addy:Exactly. Yeah. Yeah. I mean, I, I, gosh, I can't believe I'm saying this on the podcast, but I have, I have faked some school. notes for my kids schools before i mean i think it's okay if you're doing it and uh not them yeah it's like a way they were they they forgot a doctor's note for that day let me just generate it with nano b and i do it and they've accepted that it's fine is that bad no i i think schools
22:36Joey:have gotten more ridiculous i remember i remember when i was like super young i don't know we went on like trips and they we would just get my school work ahead of time and the schools were like cool they're like yeah yeah here's the work he's gonna miss like you know have fun and now they're like so picky about attendance and it's like what yeah i think the conspiracy theory is
22:52Addy:that they get paid schools get paid for per student per day basis yeah so like it's they lose literally lose money if somebody doesn't go to school they're like no no no no remote work for
23:02Joey:you it's returned to office for it's returned to office for you kids yes exactly five days a week we're paying for this beautiful uh office space it's about the community here and the camaraderie. We need your children for the money.
23:18Addy:We need to harvest them.
23:21Joey:Okay. Next update. C-Dance 2.0 has been rolling out, and every time I would see a company be like, we got C-Dance 2.0, and then in the fine print it'd be like, except for the U.S.
23:33Addy:It was like, this is the most reverse rollout.
23:35Joey:I think because, you know, they had such a splashy launch with Brad Pitt and Tom Cruise, and then got into the, like, legal, you know, everyone threatened to see them. Like all the studios started to prepare their lawsuits.
23:49Addy:Like, hey, man.
23:50Joey:And then we're like, we'll roll it back. But then in the meantime, they were holding off on rolling it out in the U.S. and then started rolling it out in other territories. But they have technically rolled it out in the U.S. But I messed around with it on fall. I don't know if other tools have slightly more open access, but um it's pretty it's highly it's highly restrictive it's redacted if you have anything in your inputs with a human face it will reject it so it's kind of honestly it's pretty useless right now it's um it will reject that it caps you at 720p output so it's also kind of useless all of the whole like a can generate like multi-minute you know coherent short films i mean i'm sure that is possible in the version they ship that's not possible it's a 15 second cap there's no multi-shot prompt understanding like how cling you can has a structure where you can you know prompt for multiple shots in a single generation it doesn't do that a lot of stuff that was like in the initial announcement i am sure it can technically do it it's just not they don't let
24:57Addy:you do it right now well that's just a big can of disappointments man yeah and i messed with it a
25:03Joey:bit and uh i emailed a contact at fall they said that about the face thing because also it was like when they initially rolled it out it was like you had to verify your business and promise some things um and the file was like we're the there should be less restrictions network next week we're
25:23Addy:working on it okay okay okay because they know this is gonna die if they keep this up
25:28Joey:i mean preventing faces period is kind of pretty extreme right just any face it's just like nope No. Like even when the Sora, the mobile app, they allow faces, right? Then they prevented you if it was like an IP face. But like we're just giving it, I'm trying to test it with inputs.
25:44Addy:Yeah, what about a synthetic person, synthetic face? No, any fail.
25:47Joey:Because that's what my, yeah, I wasn't testing with IP. I was testing like an image of a fictional human character as an input. And then I was like, no, if it has a face, no. No bueno.
25:58Addy:Yeah. All right. We'll give it more time to sort out the kinks, I guess. Give it more time.
26:02Joey:And it's, yeah, pretty extreme rollback. But, yeah, obviously we've seen, we know what it's capable of. It's just haven't put that power into the hands of everyone yet.
26:14Addy:Yeah. I mean, do you remember when the Tom Cruise Brad Pitt video came out? Around that time, a lot of, like, bigger AI YouTubers had, like, early access to C-Dense 2. Yeah, that's possible. And some of the stuff that they generated was pretty solid.
26:29Joey:I have not gone down the rabbit hole of like, yeah, let me try to turn on a VPN and try to go to some party tool who has access with ByteDance to get like a less filtered version of a model. I'm sure there are ways to dig into that. Yeah, I've only been interested in like above the board cleared legal ways. Also because of, you know, like clients and stuff. I can't. Yeah. I mean, you're doing it for commercial reasons.
26:55Addy:Yeah, you can't. Yeah, like, it's something where it's like,
26:58Joey:yes, we can legally access this through file. We're not skirting around anything.
27:01Addy:But Joey, the weekend guy who's just having a beer and he's on his phone, can that Joey the weekend guy generate stuff through a VPN?
27:08Joey:That's what I'm saying. I'm sure Joey the weekend guy, if he had more time, could probably figure that out. I'm sure it's possible. I just have not gone down that rabbit hole.
27:15Addy:Okay, all right, all right. Yeah, we're going to have to have a showdown. Joey the weekend guy has to enjoy his bourbon. i think when c dance 2 is out for good like it has all the bells and whistles you and i should do a showdown like a quick yeah yeah yeah yeah something with more features because like right
27:34Joey:now with like what's publicly available cling is still in my opinion like the best and there's control and most options um c dance isn't uh cling three yeah cling three is very good and then until you know maybe once cdance like rolls out and it's like you can do more stuff than like the next day vo4 is going to come out google's just waiting and they're like yeah and drop so speaking of google last update uh last week they released their biggest open source local model uh gemma 4 uh so these are built from the same world-class research as gemini 3 gemma 4 brings breakthrough intelligence directly to your own hardware for advanced reasoning and agentic workflows.
28:22Joey:So it's an open-source distilled version of Gemini 3 that can run a couple flavors depending on your memory that can run locally. And they have a small version that's designed to run on smartphones locally.
28:39Addy:Wow, really? Yeah. Is this what Apple's going to license for Siri? I thought they were licensing Gemini or something, right? That's pretty cool that they did that. I think they are.
28:51Joey:I think, good point. Possibly.
28:55Addy:That makes sense, I guess,
28:57Joey:for smaller tasks. I don't think the local phone model can do that much, but it would probably be good for just on-the-phone stuff. Like, yeah, create a calendar event or call this person. Yeah, very basic stuff.
29:11Addy:You don't have, you want something fast, you don't have, maybe you don't have connectivity. Or if it has access across all your apps, you can have a do things in those apps, right? Like it'll click for you and swipe for you. And I hope, I mean, I hope Apple's thinking along those lines.
29:25Joey:I think that would also be a matter of the apps building something or Apple developing their own version of like an MCP for apps. Yeah. Because for for a model to like load up a website or an app and then look at the interface and then click and then do the thing. like as a human it's very slow and very like resource intensive whereas if it had the sort of like an api but with mcp like the because the access points are already defined so that it can access and do things and it's like way way faster right right no i mean look you know now we're talking about that i would that's probably i would highly expect something like that to happen like on the app developer side for them to be like hey app developers we have like a i am sure that i can use mcp i'm sure they'll make up their own thing but it'll be something like that for the ai chat bots to access and do things right i think don't they sort of have that with siri like a way for siri to access if you wanted to find it like siri can access things in your app i don't i don't use
30:30Addy:siri at all like so i turn it off what's the weather what time what's the weather yeah it's my alarm set a timer yeah super limited so whoever the they announced not announced but they the media has identified the successor to tim cook i forget who that is it's like a senior executive
30:50Joey:at apple i hope when that person comes online like these are the things that they're thinking about
30:56Addy:because ios can go from like an operating system to like a vibe coding platform where you are literally building apps within your phone all locally to do the very specific thing that you're
31:08Joey:trying to do yeah or even just an easier way to like because they have the shortcut tool which i know is kind of powerful but it's a little clunky it's clunky so if it was yeah there's just like i need you know when i tap this thing you do this thing and i was like okay here it is don't get
31:22Addy:addy and joey started on apple talk i think we're gonna go off on it we're too big of apple fanboys
31:29Joey:yeah no to what you're saying like because look my open claw thing is sometimes useful sometimes like i want to like slam my head into a wall but one of the nice things is like when i have it loaded up on telegram and i just hit the mic button and i'm just like blah blah blah blah blah and then it gets it and it like does the thing right that's when it's like uh that's powerful it's like one of those tastes of like uh like this is how the thing should be like this is how series should be i shouldn't have to start doing this through telegram so yeah that's you know
31:57Addy:what we would hope they are going towards. You also probably have, I don't know, 100 apps on your phone, right? You don't want to be going through each one of those and modifying things and doing... Who has time for that? Just talk to it and tell it what to do.
32:10Joey:I will say update. This is going to be a slight rant. I kind of get it. But in the OpenClaw world, the biggest upset this week, or maybe it was last week, has been Clawed has revoked OAuth use on OpenClaw. So basically, I have Clawed Max and I was able to sign into CloudMax and use my CloudMax plan with OpenClaw, which solved the cost issue because then it was just using my monthly plan limits, which is great because I don't have to worry about the API jacking up every time I used it. And then they sent out an email being like, we're going to revoke OAuth access to OpenClaw. You're going to have to pay per usage on OpenClaw.
32:51Joey:We'll give you complimentary$200 of credits to use for it but then once you're out of those you're gonna have to pay to use it um and i'm also convinced that the free credits they gave are from the dumbest version of claude i've ever used because open claw has also gotten so much worse since they switched this and i don't know why it's like and my theory my conspiracy theory is they're they are uh given the crappy version
33:19Addy:of claude to open claw yeah just just so you're frustrated and you pay for it you pay for the nice
33:25Joey:Just so you're frustrated and you pay for it, or you give up and you go use their OpenCloud competitor tools that they've been building. Is it just called computer or machine or something? Well, that's the thing. There's like three different verticals with Cloud, and they all do slightly different stuff because there's Cloud the app. There's Cloud Cowork, which can run on your computer. There's Cloud Dispatch, which can control a Cloud on your computer from your phone. there's Claude code in your terminal but now you can also connect terminal to your telegram and run it there and there's Claude and then there's remote control which can control terminal from the Claude app on your phone so that as I've explained this you're probably like that's a lot of things there are a lot of things and there's no one central unified like open claw hey it's just a chat thing I can talk to and you can control my computer and it's centralized it's very fragmented right now
34:20Addy:they're going after different markets with each different products but they haven't thought of somebody who would come to the walled garden in the first place the initial evolution made that
34:29Joey:could add cloud code which is like a little bit more techie and then they had cloud code work which is sort of like a watered down version of cloud code yeah for non-techie people and just make it like a similar capabilities but in a more uh yeah it still uses local files and that like
34:45Addy:it accesses your hard drive it can work with your stuff on your computer but you don't have to code
34:48Joey:It's not in Terminal, it's in the app, yeah. And then I think, you know, OpenClaw shot off, and then they've been wedging features from OpenClaw into the app that were not initially thought of or thought out. That's my theory. Like, these extra things have been, like, added on to, based on how people responded to OpenClaw, but they weren't part of whatever longer-term vision they had of what the Clawed universe would be.
35:15Addy:What do you think is the near-term future of OpenClaw and sort of at-home agentic capability?
35:22Joey:When it works, it's good. And it's still the best thing I've seen out there of just like, hey, keep like, I want to do this thing. Can you find it? And then it like adds a feature or finds an ability or adds onto itself and can sort of literally like Claws, like kind of connect and do things. it's just can forget things and go off the rails and not remember what it was supposed to do I've been trying to schedule NAB meetings with it
35:52Addy:how's that going?
35:55Joey:I've had to send a couple apology emails
36:00Addy:that was my agent not me sorry
36:04Joey:we don't want to waste time with your booth cat here
36:08Addy:oh my god
36:09Joey:No, the worst it's done is like it's jumped the gun and like sent a time that that didn't approve, like run by me first. So it like has sent time to be like, yeah, that's great. And I'm like, no, you have a set of rules. We have to like check the schedule and then like check the rules and then like confirm with me. And then you email them and it'll just like email them first. And then I'll be like, well, you don't check with me. It's like, you're absolutely right. I forgot to check with you. It's like training. It won't happen again.
36:32Addy:Training an entry level admin who just joined the workforce and they know nothing about scheduling meetings.
36:38Joey:you're like no no no you you have to check the calendar like yes yes joey yes uh next time yes absolutely would you like some coffee yeah or in like uh parks and rec when uh what was like aubrey plaza was uh was ron swanson's assistant who like didn't really care about her job and then she's like so i would just schedule everyone for march 31st because i didn't think it existed and then but today but today is march 31st but it just feels like that yes level of oh yeah i I forgot about that. Contempt, yeah.
37:06Addy:I'm too smart to do this.
37:07Joey:But when it works, and I'm like, hey, I got it like, where's this company's booth? And like, what slot, you know, do I have an opening? And again, can we email them? And when it works and it does that, it's cool. I'm like, okay, great, working. So when it works, it's great.
37:22Addy:But it needs handling. I mean, here's a hypothetical question. Now they have a service. I mean, they've had it for a while where you can hire a remote admin for yourself for like a minimal. I mean, they're just probably people overseas. Is it cheaper to do that than to just figure out cloud code and open claw and all the credits and the tokens? You just hire somebody overseas.
37:48Joey:I think it depends what your level of like patient. I went into this knowing it was going to be faulty and that I would still, as an experiment, want to spend the time to try to tweak it and get it to work. I went into this not expecting it to work 100%.
38:04Addy:Okay, okay.
38:05Joey:And I think that if it comes in from your user experience and of like, what do you want out of this? Yeah, probably if you're doing this day in, day out and just didn't want to deal with tech stuff, getting a virtual assistant is probably still easier. I still think that would probably be more expensive that's a few thousand a month I don't know
38:27Addy:I mean just kind of thinking about it long term like where are we going to go with this hopefully all the kinks are worked out and we all have our super agents at home that are basically multitasking on behalf
38:40Joey:of us I mean when it knows the stuff it's good like I've gotten all of the videos transcribed so if I'm like find you know where are the video clips where we talk about X company it can do that. And because it's already kind of connected to stuff, it does that. So when it works in those cases, it's good, and you can kind of see where it's going. I mean, the problem I'm having now is OpenClaw knows some stuff, Cloud knows some stuff, ClickUp knows some stuff. I use some other note-taker apps. So the problem I have is just trying to unify everything so that no matter what tool I use, it has the same knowledge and context that all the other tools have.
39:16Joey:And that's also been a roadblock of just having a central source of knowledge. It's Wild West, man.
39:23Addy:No, I'm living vicariously through you, so thank you for the update.
39:26Joey:Yeah, I will suffer through this and try to report back. And hopefully some of our viewers are also living vicariously through you. Yeah, or if you're doing this yourself, too, and you have found anything that works, let
39:35Addy:me know in the comments. Hit us in the comments. Joey will try the wild experiment on your behalf.
39:43Joey:I didn't save it here, but Pika launched something that was interactive avatars, so it was like a live you can create an avatar and then it can connect to a live feed and so the avatar can like act basically can act on like open clause behalf so you could have a zoom meeting with your um ai avatar that's creepy in real time that's my next that's my next thing i'm gonna
40:05Addy:mess around with the uncanny valley uh like brain and face from the uncanny valley yeah yeah yeah all right good place to wrap it up links for everything talked about at denoisedpodcast.com thank you for a wonderful comment on spotify we appreciate you and um if you are on apple podcast we haven't gotten a five-star review in a while hint hint drop us one please thank you thanks everyone catch you in the next episode
From the publisher
Netflix just open-sourced an AI model that removes objects from video and corrects the physics — and that's only one of several AI updates worth tracking this week. Addy and Joey break down Netflix's VOID model, early leaks of what looks like GPT-Image-2, the underwhelming public rollout of Seedance 2.0 in the US, Google's Gemma 4 open-source local model, and the honest, unfiltered experience of running AI agents for real-world scheduling and production workflows. Plus: Artemis moon mission camera nerding.
--
The views and opinions expressed in this podcast are the personal views of the hosts and do not necessarily reflect the views or positions of their respective employers or organizations. This show is independently produced by VP Land without the use of any outside company resources, confidential information, or affiliations.
