Loading summary
A
Welcome to the podcast. Today, State Farm is under fire for cutting commissions to some of its insuring agents because of AI. We have some news that Allbirds is not going to be called Allbirds anymore. They're focusing, they're shifting their name and they're going to be focusing on AI infrastructure. A wild switch up and pivot for that company, a company called Odysseus, has just hit $1.45 billion in valuation. Amazon and AMD are both betting that their world models are valuable. And only 16% of Americans see AI as a net positive, according to some Pew research. And that's in a specific subset of this research, but there's some pretty incredible statistics there. Threads, Instagram and TikTok are all giving users the ability to adjust our algorithms ourselves and not just have AI pick purely what we want. I want to talk a little bit about where we see that going, especially when one of their competitors, Pinterest, has just launched a brand new app called Ask Print Pinterest. This is a standalone AI shopping app. And it's interesting because this is a big pivot from what Pinterest has very publicly sort of been opposed to AI in the past. I am super excited to announce that AI Box is officially launching an mcp, which means that if you use a tool like Claude and it doesn't have audio and it doesn't have image generation, you can use the MCP of AI Box to get the image generation and audio generation inside of Claude. Or if you're using something like ChatGPT, which is about to kill the Sora video generation, you can, you can get Google's VO3 video model inside of ChatGPT and use our MCP to get access to all of the AI models inside of the AI tools that you already use. If you want to check it out, I'm leaving a link in the description to AI Box AI mcp. This is something we are super, super excited about. It is $8.99 a month and you get access to over 80 different AI models. Image, audio, text, video, music. And you get them all inside of any of the AI tools that you already use. So go check it out. I hope that saves you a ton of time and money. O let's talk about the first story here, and that is State Farm. There's recent backlash. There was a writer, Gene Eaglesham, who said State Farm's AI driven plan to impose a new contract on its 19,000 sales agents is causing an uproar. This was written in the Wall Street Journal and essentially what's happening is there's this huge backlash because there is a change that is going to reduce the income by up to 40% of these sales agents over at State Farm. It's also going to cut their benefits. And, and this is all part of a strategy to adapt AI advancements. And State Farm right now is really trying to take on Progressive, who has overcome or who has basically surpassed State Farm as the largest personal auto insurer. Now, I don't know what part of the world I'm in. I use geico. I just asked all of the different car companies. I was the cheapest one. Whenever I called around. State Farm and Progressive were way more expensive. So I have no idea why. Those are the two biggest. Maybe you guys can tell me if you listen, what I will say is, as someone that is not in this industry or, you know, doesn't know much about this industry at all other than I'm a customer of it. I personally, when I hear that a sales agent is going to get their, their benefits cut by 40% or their, their income or their commission, sadly I actually, that doesn't bother me. I don't want to pay any commission to a sales agent. I just have to get insurance on my car. I really appreciate the way GEICO lets me just go on the website, put what I want, select it, you know, make all of the adjustments I want and just click buy. And I think in some cases actually do have to call the GEICO agents to get stuff set up. But it always feels like this is kind of a bureaucracy. It's always kind of something that has been regulated into existence. I don't really want that regulation. I wish it was more free market with insurance. But because legally I'm required to get insurance. I just want the cheapest insurance possible and that is me. So anyways, if AI can help do things cheaper for the consumer, which is myself, I'm happy now for people that are, you know, getting their income cut. I am sorry, that is sad and you know, I resonate there. But I truly believe in a free market and if we can make this better for the consumer, that is kind of what I would say is the best outcome. So probably not a very popular opinion, but that's my thoughts on that. Okay. Allbirds is changing the name. This was, you know, kind of a trendy shoe brand that came out of Silicon Valley. They're changing their name to smartbirds. This is based in San Francisco. They've gotten a new chief executive that's going to be helping them with this and they're switching from shoes to AI infrastructure. This was not a shift that anyone expected, but essentially what was happening is I believe they were public, their stock was cratering, it was at all time lows, you know, they were becoming unpopular. People were just worried this company was going to get delisted. And so they decided to sell off the rights to Allbirds, their shoe brand, take the money that they were able to get from that and just dump it all straight into AI infrastructure. So Nadia Carlston is going to be in charge of the company. The former CEO Joe Vernancio is going to be leaving in April. They said that they were going to do this rebrand. So I remember this. They actually, I think said they were going to call it like New Bird AI or something, but it appears that they're actually going to be just calling it Smart Bird. But in any case, the new CEO, her background includes a stint as CEO at a data or at a Danish center for AI innovation. Um, and I think there's, you know, they have some people that had experience over at Amazon. So this is going to be incredibly interesting. Here's a post that Nadia, the new CEO, posted on LinkedIn. She said, for many organizations, shared infrastructure simply isn't enough. Whether driven by performance, cost, security or sovereignty needs, they need dedicated infrastructure. But they shouldn't have to become experts in GPUs, networking, data centers and cluster operations to realize the benefit of AI. So they are fully like switching all of their branding, all of their, you know, their speak into just purely data centers. It's crazy just to see a company that, you know, they were publicly traded, they had some sort of assets and they realized that this company wasn't going to be successful long term. They cut their losses, took the money and just started doing data centers. That's a crazy pivot for a shoe brand, if I'm being honest. Okay. Audius has just hit a $1.45 billion valuation. Amazon and AMD are both investing in them. It was a new round which was led by Natural Capital. Um, and then you also had GV that was participating as well. This is a company that was just started in 2023 and they have now raised $337 million in total funding. AWS is now Odysseus's preferred cloud provider, which, you know, this is how we see these deals, right? Amazon's gonna put the money in and they expect that they're gonna be using their services aws. So you can see Amazon, Microsoft, Google, they put money into all of these different companies Nvidia even, right? And they're like, hey, you gotta make sure you're using Nvidia GPUs. But in this case, because Amazon got to them before Nvidia did on this round of funding, they're actually going to be optimizing their models to be trained and to be run on Trainium chips and not Nvidia GPUs. So Amazon is putting, you know, a little heat to Nvidia by investing in companies and getting them to use their Trainium chips instead of Nvidia GPUs. The founders are Oliver Cameron and Jeff Hawk. And they come from the self driving. Cameron was at Voyage and Cruise. Hawk was over at Wave. They have a bunch of amazing angel investors. This is a very serious company. They have Jeff Dean, Elad Gill, Gary Tan, Guillaume Rauch and Cruise founder Kyle Boak. This is a company I would definitely keep your eyes on. And you know, incredible the passing that billion dollar valuation. So there was some research done by Pew and a certain subset of the people they surveyed found that only 16% of them saw AI as a net positive. This is pretty wild if you look at this. They say, you know this, you know, AI doesn't have, it doesn't expect to be a net positive impact on society over the next 20 years. 40% expect a negative impact. According to Pew, 44% of US adults now use ChatGPT, which is double the amount when they did the review or kind of the poll in 2023. So the share is up. Gemini is at 24%, Copilot is at 17%. According to this study that Pew did, 67% don't believe the US government is going to meaningfully regulate AI. 59% don't trust companies to develop it safely. About 25% of Americans use AI chat bots every single day. But about 75% of people that are 65 and older say they never use them. So I mean, this is a big difference. Younger people are heavily skewed to using these tools. But among people that were under 30 years old or only 14% of them think that AI will be a net positive. This is by far the most skeptical age cohort in the survey. And I mean a lot of these people, you have to think about it, are in school, they're early on in their careers, they're seeing how this is becoming harder to get jobs out of school. Junior level positions are not hiring as much. So I mean, this makes a lot of sense and it's interesting to see where this sits. But according to all of this, we also saw that Meta AI was at 14% usage, Grok was at 8% and Claude was at 6% and Character AI was at 3%. Now, what's interesting to me here is, I mean, we hear a lot about Claude. You probably hear me yelling at the microphone all day long. Claude, as Sam Altman once said, is an expensive tool for rich people. Now, I don't know if that's super true, but what you do have to remember is Claude is very useful for a certain subset of white collar workers that are doing things on their computer. Claude is not useful or very used. Maybe, maybe it is useful for everybody, but it's not very used by the entire general population. According to the survey, only 6% of people are using Claude, where you have a much bigger percent, 44% that are using ChatGPT. So ChatGPT has way more usage. But CLAUDE is making, you know, possibly potentially more money because people are paying $200 a month for Claude plus API credits. So, you know, Cloud has a very high, you know, they do have a $20 a month tier, but they have a lot of people on their $100, $200 a month tiers. So Claude is making a lot of money from, from that, which is a big jump. Okay. Threads and Instagram and TikTok are all giving users a little bit more control over the algorithm that they see. Instead of this just being purely, you know, AI picking that for you, although, you know, the AI is still doing the, all of the algorithmic work, you get to pick a little bit more about what you see. I think there is something. So Threads just, just launched something called Year Algo today. And basically it lets you privately tell Threads to show you more or less of a specific topic in the way that they're doing it. It's like, hey, show me more of this topic for three or, you know, three or seven days at a time. That is so weird to me. Like, hey, I want to see more about, like, cars for the next three days, or I want to see more about, you know, slacklining for the next seven days. Like, that's so weird. If you're interested in a topic, I probably just toggle it on in the settings. Like, hey, I'm interested in like, these 10 different things. It's like the basic onboarding you expected from every, you know, web service over the last 10 years that asks you, like, what you were interested in. And, and then of course, TikTok just kind of guessed it and figured it all out. But it's interesting that they're only letting you pick that for a few days. I'm sure whatever you interact with is just gonna get permanently baked into the algorithm. You'll probably keep seeing some of that stuff, but it's interesting. They're only letting you tune it for that amount of time. There was a different thing that came out with Instagram earlier this month called your algorithm, and that works for Feed, Explore Reels, and this happened at the same time. I think they're kind of basically trying to compete with TikTok, who has a tool called Manage topics. I think that's like two years old. TikTok's been doing this for a long time. But TikToks is pretty interesting the way it actually works. They'll let you pick a whole bunch of different keywords that you're interested. You could say, you know, I'm interested in like sports, travel, humor, current affair, dance, food, like, you know, all of these different topics and it's going to show you more of those. Now, I think this kind of stems from maybe some criticism that TikTok got over the last couple years. I don't know how accurate or true it was, but I've heard for many years that if you go to watch TikTok in China, it's going to be, you know, all these videos about people learning different STEM, science, academic things. And then if you watch TikTok in America, it's just, you know, degenerate gas station brawls and, you know, that's like China trying to use the algorithm to destroy the American mind. Now, I don't know how much truth is in all of that. We kind of have the, you know, TikTok has been divested from China to some degree. I don't really know exactly how much separation has happened with all of that. You know, although that's been going on for years as well. What I will say is, regardless of the truth to all of that, there was definitely a lot of bad PR and, and you know, people were talking about that. And so I think they are like, look, hey, if you're worried that you think we're trying to destroy your, your brain with the algorithm and give you TikTok rot brain, look, we'll let you pick your algorithm. Now, is it going to destroy your attention span to watch 15 to 60 second clips on repeat all day long, what? Probably. But now at least you get to pick how your attention span is destroyed. Just kidding. But I think this is a good move in the right direction, allowing people to pick what they want to see. Okay. The last story I want to cover today I think is fascinating because this is a company that has taken a completely different approach than basically all of the other social media platforms we've talked about here. And that is Pinterest. Pinterest has been very famous in saying, like, hey, look, we're trying to not really outright ban AI generated content on our platform, but basically they're trying to demote it, they're trying to tag it, they're trying to flag it. They really don't want you to see a lot of AI generated stuff and they've been pretty clear about that. Something that's interesting about Pinterest is a pin has a very long life. A life, I don't know, expectancy, whatever you want to call it, right? If you post something on TikTok, you can expect that thing to be popular for like a week, maybe max, three days, probably two days. And after two days, you basically run the course of how many people are going to see that organically in their algorithm. Pinterest, on the other hand, you can get something, you know, five years later, someone might still be looking up your, your mushroom soup recipe or whatever and looking and repinning that. And that's actually a platform I've played around a lot with my first startup, Self Pause, which by the way, my wife just rebuilt and relaunched, completely redesigned with new features. She vibe coded on Claude right before she had her baby a week ago. So if you're interested in checking out Self Pause, subtle plug to that. It's got an AI life coach. It's a AI. Well, it's really just a meditation app with real voices and real meditations. But we did add an AI life coach on there. That's kind of cool. You can also record your own voice. But in any case, something that was interesting I've done a lot with Pinterest is I actually used Pinterest to promote that Self Pause, that app when I first, when I first started launching that company maybe like seven years ago. And I very quickly was able to get to about a million monthly views on my pins that I was posting for that app. So Pinterest has a massive potential to really grow and push things if it's, if you got the right audience on there and if you're, if you're interacting with it correctly. But all of that was, you know, not AI generated content. And when all the AI generated content came around the, you know, the name of the game with Pinterest, the way I was able to get a million impressions back then was I was pinning, I mean, I was using like Canva and I think I had virtual assistants that were making like pins for me. My strategy in particular was I would just kind of make a nice background and I'd have like a positive affirmation quote on that background. And then I had all of these different boards that were categorized and I would have, you know, like, I think like 20 to 50 pins going out every single day. I had virtual assistants that were making them. We had, you know, big databases of positive affirmations that we had writers writing. So that was. It was easy to just copy and paste that onto like a nice background. And I had virtual assistants that could pump out a ton of them, but it was all about the numbers. That's how we were really growing on Pinterest, was how many things we were posting. So today, when AI can virtually do an unlimited amount of that content for even cheaper, even faster, they really wanted to cut back and try to give people just, you know, organic human made content. And so it's pretty interesting to me that they have now launched a new standalone AI shopping app, which is, by the way, is all based off of Pinterest's taste graph. So they have like all their database, all their data is used for this. But they didn't want to put all of these AI features into the Pinterest apps. They made their very own standalone app for this. And it is basically a conversational AI shopping app. So you can chat with it about what you want to get and it can bring up information and different products. And it's all based off of everything that's already on Pinterest. It's limited access right now, but even Pinterest, who completely swore off AI, is testing the waters with AI. I'm excited to see where this goes because I see so much potential with generative AI and shopping. And Pinterest has such a huge opportunity with shopping. You see what Amazon's doing with AI and shopping. I think Pinterest sees that they don't want to miss out, but they don't want to feel like they're maybe selling out to AI. So this should be interesting. They are rolling out AI in a bunch of other ways. They have a new MCP that people that are advertisers can use to help interact with their advertising platform. They also have a new performance plus creative AI model that went live globally to basically select the best performing ad creatives. So when it comes to ads, they're using AI and algorithms and mcps. But for the users, they're kind of trying to push away from that branding until this brand new standalone app. So I'm curious to see how this will perform. Thank you everyone so much for tuning into the podcast today. If you enjoyed the episode, make sure to leave a rating and review wherever you get your podcast. And as always, make sure to go check out the brand new AI box mcp. If you want to get access to images and audio creation and inside of Claude and ChatGPT and you know, be able to pick through 80 different AI models. You're able to also build tools on AI box and call them from inside of Claude. So if you have a workflow where you're, you know, generating different thumbnails or different assets, you can create a tool over on AI box that's going to get it perfectly. You know, it's going to get it done perfect every time. And you go and have Claude call that, spin up the asset, grab it and build it into your workflow. So it's amazing. I'm super excited about it. You can go check it out. The link is in the description to AI Box AI slash mcp. All right, catch you guys all in the next episode.
Podcast: Open AI
Host: AI News
Date: June 17, 2026
In this episode, AI News explores the disruptive effects of artificial intelligence across various industries, focusing on how State Farm’s AI-driven changes are impacting insurance agent commissions. The host also covers tech pivots (like Allbirds rebranding to “Smartbird”), fast-growth AI startups, shifting user control of social media algorithms, new data on public attitudes toward AI, and Pinterest’s cautious entry into generative AI shopping.
[05:30] State Farm’s rollout of AI-driven tools has led to new contracts for 19,000 sales agents, provoking backlash due to up to 40% income and benefit cuts.
Host Reaction:
“When I hear that a sales agent is going to get their, their benefits cut by 40% ... that doesn’t bother me. I don’t want to pay any commission to a sales agent. ... If AI can help do things cheaper for the consumer, which is myself, I’m happy. ... I truly believe in a free market and if we can make this better for the consumer, that is kind of what I would say is the best outcome.” — Host [06:50]
[10:25] Once a trendy Silicon Valley shoe brand, Allbirds is now rebranding as “Smartbird,” moving entirely into AI infrastructure due to financial struggles and declining shoe sales.
“They realized that this company wasn’t going to be successful long term. They cut their losses, took the money and just started doing data centers. That’s a crazy pivot for a shoe brand, if I’m being honest.” — Host [12:45]
[14:30] Odysseus, founded in 2023, just hit a $1.45B valuation after raising $337M; Amazon and AMD are major backers.
“Amazon is putting, you know, a little heat to Nvidia by investing in companies and getting them to use their Trainium chips instead of Nvidia GPUs.” — Host [16:30]
[19:00] New Pew survey highlights that only 16% of Americans view AI as a net positive for society; 40% expect negative impacts over the next 20 years.
“Among people that were under 30 years old, only 14% of them think that AI will be a net positive. This is by far the most skeptical age cohort in the survey.” — Host [20:30]
“Claude, as Sam Altman once said, is an expensive tool for rich people. ... It’s very useful for a certain subset of white collar workers... only 6% [of] people are using Claude.” — Host [23:15]
[25:00] Threads, Instagram, and TikTok now enable limited, time-based user tuning of content algorithms.
“If you go to watch TikTok in China, it’s going to be ... STEM, science, academic things. And then if you watch TikTok in America, it’s just, you know, degenerate gas station brawls ... Now, is it going to destroy your attention span to watch 15 to 60 second clips on repeat all day long, what? Probably. But now at least you get to pick how your attention span is destroyed. Just kidding.” — Host [29:00]
[32:00] Pinterest, known for restricting AI-generated content due to the long “life expectancy” of its pins, has surprisingly launched “Ask Print Pinterest,” a separate AI-powered shopping app driven by Pinterest’s “taste graph.”
“Even Pinterest, who completely swore off AI, is testing the waters with AI. I’m excited to see where this goes because I see so much potential with generative AI and shopping.” — Host [36:00]
State Farm’s Free Market Rant:
"If AI can help do things cheaper for the consumer, which is myself, I’m happy ... I truly believe in a free market." [06:50]
Allbirds’ Unthinkable Pivot:
"They realized that this company wasn’t going to be successful long term. They cut their losses, took the money and just started doing data centers. That’s a crazy pivot for a shoe brand, if I’m being honest." [12:45]
AI Adoption Stats:
"Only 16% ... saw AI as a net positive ... only 14% of under-30s think AI will be a net positive. This is by far the most skeptical age cohort." [20:30]
Algorithm Satire:
"Now, is it going to destroy your attention span to watch 15 to 60 second clips on repeat all day long? Probably. But now at least you get to pick how your attention span is destroyed. Just kidding." [29:00]
Pinterest’s Generative AI Turn:
"Even Pinterest, who completely swore off AI, is testing the waters with AI. ... potential with generative AI and shopping." [36:00]
This episode illustrates rapid, sometimes unexpected, transformations as AI penetrates legacy industries, consumer brands, finance, and tech platforms. The host provides sharp, personal commentary and a range of statistics, while surfacing tensions between innovation, market forces, and public skepticism. AI is creating winners and losers fast—and even the most “AI-cautious” platforms are now hedging their bets.