Never follow your heroes (on X)

0:05Today on Great Chat, we grapple with the wave of tech leaders who are trading mystique for distribution and engagement maxing. Is this the new world and we're just boomers? I'm Ashley and I'm joined by Helen. Great chat is brought to you by Mercury, radically different banking loved by over 300 ,000 entrepreneurs. We use Mercury to run this podcast and love recommending it to other founders and fund managers. It's free to get started. Try it for yourself and visit mercury.com to learn more and apply online in minutes. Hello. Oh my goodness. Did you see that our good friends at Anthropic pushed the Fable access deadline once more?

0:44Yes. I'm using it nonstop and it's good. I don't know if it's 50 % better, but it's good. When we recorded last week's pod, it was like the last chance to use Fable within your existing paid cloud plan. And then right after we recorded, they extended it to the 12th. My husband's like, do all your workflows, do all your research. I'm like, oh my God, okay. Wait, Gabe? Gabe was like that? Yeah, yeah, yeah. That's so funny. Yeah, yeah, yeah. He's very worried that I'm not using my plan properly. He's like, well, do you have any automations? I'm like, yes, I have automations. Are you guys on a family plan?

1:24No, no. Hell no. I keep my relationship with Claude and Chad GVT. Oh, yeah. My best. That's right. We are not in an – well, I guess it's because they were not in an open relationship, but I guess we are. But like I have a lot of respect for anthropocoms and they've done some beautiful marketing and like gotten a lot of kudos. But this just kind of feels messy. Like, yeah, I'll take more Fable access, but like what is the end game here? Like this is just – it feels kind of reactive to maybe what ChatGPT is doing or like Meta is doing. and I just want some clarity. Whatever little tiny scrap of clarity you can give me about the future, I will take.

2:10That's funny. I didn't think about it through this lens of you're going to lose Fable access. What are the next five moves you make? You know when they announced that cities were going into lockdown during COVID? Yeah. Whatever day that was in March, it's a fun icebreaker question I would ask is like, when you found that out, what was like the move that you made? Like, what was like the final, you're like, okay, the city's going to lock down tonight at 9 PM or whatever. Like, what did you decide to do? We went out for sushi. We were like, this might be the last chance to eat sushi in a long time.

2:49Okay. So like I went to Flora Grub Gardens, if you know what that is in San Francisco. I do not know what that is. I panicked and I went to a nursery and I bought some plants. Oh, that's so wholesome, Helen. I don't know. I feel like that's a good instinct. I was like, I need my raw fish, but I will not be preparing myself. Yes. So Angela jumps on the group chat and she's like, did you guys see the anthropocode? And we're like, I was driving. and I think like most people they saw it for the first time during the quarterfinals between Argentina and Switzerland over the weekend, I think. But the ad actually ran a couple days before that and there was some coverage on it that I completely missed.

3:37But I was so excited. Angela got excited about a commercial. I was like, oh, I can't wait to watch this. And she said it was good. She liked it. She loved it. Yeah, and I thought it was just fine. Did you watch it? I did, but I want to hear your take first. so yes first i was like not knowing anything and trusting angela fully i was like oh yes this is the golden age of tech marketing all these ai companies all these startups are just finally coming around and i'm so excited and i watched it and i was like damn it they just did the same shit and what i mean by that is i felt like oh gosh i mean there are some things that i feel like are more fundamental inexperienced television advertising moves.

4:23It's a 90-second commercial. That's a very long commercial. And so if you were jumping in at any point, or you left early, you didn't know whose commercial it was until the very end where they just had Claude as the end card. And it's like, wow, you spent all that money for most of that commercial to air and have no idea who to attribute that to. So that was tough. And then the second one was, it was beautiful cinematically. I think people loved the music, but if that was a Gemini ad, it could have worked just as well. If that was a chat GPT ad, it could have worked just as well. That's really hard.

4:59You're spending all that money to advertise not your own company, but your category. And so it kind of messed that one up. I thought the URL was strange. It was like hard dash questions. Already Claude is like kind of hard. Some people don't know how to spell it. And if you're airing it. People don't know it has an E at the end. Pro tip for our listeners. Yes. And it's a mass, you know, you're deciding to buy this big media package to do like mass, you know, market awareness. You know, like it could be the first time they ever heard of Claude. So it was sort of tough from a lot of those like standpoints.

5:39And then, you know, I mean, full disclosure, like I don't think I've made it any secret that like I'm such a Claude loyalist. Like, I mean, since Great Chat season one, I've been loyal to Claude, but I don't know. It was just a little tough for me. And then I read a little bit about it. They got 120 ,000 people submitting questions for like what kind of questions they would ask. But then it was like, they were too on the nose, kind of heavy-handed on the editorializing of picking which questions made it into the commercial because it's like, let's find all the questions that hit our corporate communications talking points around safety.

6:18And yeah, it was just like... Maybe we should describe what the ad was like for people who haven't seen it because the the sort of tagline was, there's hope in hard questions. And it felt like what Claude was trying to do was like tackle some of the big critiques slash fears around AI head on, right? Like, you know, what happens when AI takes all the jobs? Who makes sure that this thing doesn't go too far? And like they were flashing these images of like a graveyard. So it was very dramatic. and I actually didn't know the backstory in these questions. So it's, yeah, it sounds like they solicited them and then I think I did see that they actually had the people who asked the questions, like, read them for the audio and so it was like a different voice for each question.

7:10It was a very dramatic, heavy ad that then tries to take you on a journey where it's like, oh, there are all these hard questions but, like, actually AI can be amazing for people and then, you know, and then it ends with, like, Reveal, this is the Claude ad. Yeah, it's hard because I think we've talked a lot on this podcast about how we need the leaders of these massive AI companies to actually engage with these questions and not just have doomsday narratives and actually speak to the broader public about what to expect and what are the reasons for optimism and without gaslighting people or giving false hope.

8:00And I felt like this ad was actually very surface level. They get points for engaging with the hard questions, but I didn't feel like any of them were actually answered. Yeah. And so that felt like a miss to me. It was almost like a, look, we're willing to tackle the hard topics, but they weren't actually tackled. They were just acknowledged. And they were acknowledged in ways, again, like if you flash a graveyard on a screen while talking about the dangers of AI, it's just like, oh, holy shit. I really did not like the ad, which is funny because this is the kind of attempt I want these companies to make.

8:43But it really fell far short for me. Yes. Again, so I worked at Quora for a little bit. And after we had hit some 300 or 400 million users, and so it's like a lot of people. And you get to kind of see, well, what are the most common questions asked? And it's like, trust me. When you get to mass market, the questions are not that level of sophistication. So it was just sort of like, oh, this could have been like a Google ad, a Google search ad from 10 years ago where there was some other overarching theme. Like when there was maybe a market correction or something and the mood was low or something.

9:25And it just didn't hit for me. But I think to acknowledge how much of a contrast there was in Angela's reaction, I actually know many people who said that they thought it was great. I think that they were speaking to Angela. I think they're speaking to people within Silicon Valley and other parts of the country who are extremely plugged into AI and asking these questions. That's so funny. I read it so differently. Who do you think that they were talking to? If they were targeting a more like insider audience, I would say it was a huge fail because my feed hated it. Like if you look at the tweet from Claude with the ad that like unveiled over the first time and read the replies and read the quote tweets, it was like universally despised.

10:18Damn, Angela is the only person who liked this ad. Angela's like an actual contrarian. So let's give her some, let's give her some thoughts. By the way, like Twitter is not real life. And like, you know, people are, the prior Claude ad that we've talked about, their Super Bowl one, which was sort of like a dig at ChatGPT and, you know, and oh, well, we won't do advertising. So our answers are better, blah, blah, blah. Like tech people loved that and including the Twitter set, right? Like we're like, ooh, this is fun. this is like, you know, a little edgy. And Anthropic was still like the challenger underdog in that relationship.

11:00You could argue that actually now it is the category leader, at least in revenue. But still not close compared to ChatGPT in terms of consumer. And so to make making the choices they did on the media buy, it really does show they were speaking, they were looking for consumers, right? Like this was not a... Oh, I'm not saying it was good outside of tech circles. Well, outside of tech circles, people did not like it. Yes, yes. Insiders loved the prior ad. Which is so interesting. It's the same ad agency. You know, same creative team. Totally, but I think... And yeah, totally different mood though, right?

11:37Very different mood. I would kill to read the brief for this ad. And like the fact that the Super Bowl one was maybe a win with insiders, but amiss more broadly with like the general public when we looked at the ranking of Google ads towards the bottom. People were like, wait, what's AI? Like, I don't even know how to engage with this like advertising dig. And this time by hunches, they were actually trying to reach a broader audience of people who are worried about data centers and like the price of electricity to be going up and jobs going away. I mean, there are all these surveys about how like people fear AI and hate AI.

12:20But I just, again, I just don't think they actually engage with it in a substantive way. It was like dramatic images, dramatic questions, but like, let's see. Where's the hope in the hard questions? Like, I don't know. I mean, I'm an optimist too. I'm not a doomer. And I was like, the ad made me kind of bummed out. Okay, something that's still swirling on the timeline is the 1X Neo robot review by Wired. I cannot believe this is still a story. Like, it's, I don't know, lasting several more days than I thought. But basically, Wired had the exclusive, I guess, on like this new, you know, on the new Neo robot.

13:03And that's like that soft, human-like robot that we've seen before and we've talked about on the pod. And it would have been fine. It would have just been another article and there would have probably been other coverage on it, whatever. But the founder took to X and said, I feel betrayed. We gave the exclusive to Wired and they sort of twisted the story. and they're sort of talking about how it's sexualized. And it was long. And, you know, I think I forgot who, maybe it was Angela asking for our reactions on this one. And mine was just sort of like, I think he knew exactly what he was doing because it's Twitter.

13:49We've already seen, you know, sort of a wired backlash just in 2026 and deservedly so in some of those cases. But this one, I was just sort of like, okay, like if you give someone your exclusive, like rule number one is you have no idea how that's going to turn out. That is the sort of thing that you trade when you give someone the exclusive. And so he's jumping on pretty much immediately after the article came out, right? Like none of us had read it. And it's behind a paywall. So unless you're paying wired, you have to take his word for it. But you did eventually pay for it. And like, what was your take?

14:28I didn't pay for it. Okay, you got a copy of it. Like some archived thing, which I shared that link too. And I was informed by someone in media that they don't like it when people share those links. And I'm like, I don't care. Yes, we know. I can't pay for all of these publications. That's right. Okay, I have a really fun fact check for you, Helen. Why? For you and like most of the people who engage with this tweet, this guy is not a fucking founder. He is the VP of marketing, design, and product. And I think the whole thing reads so differently. If you're like, oh, it was your team that fucked this up.

15:06I was just going to say, wait, that seems really likely that the comms reports to him, right? 100%. And he throws his comms team under the bus in the tweet. And I like, it was so funny because I actually, I'm one of the few people who like decided to read the full article. I also watched the video. So basically the two things that Dar, the executive, was objecting to is one, this idea that this robot is sexualized. And then two, privacy concerns because while some of these motions – so the focus of this launch was on the hands and how dexterous and fast the fingers are. And that was the point of the launch.

15:51are, yes. But also it's like, if you think about it, like a lot of the limitations of a human-related robot are like what it can do with its hands. Can it like load the dishwasher with a glass of wine without breaking it? Can it, you know, do like more complicated things? And the technology does seem amazing. However, some of it is fully autonomous and some of it is remote operated by like humans overseas who are like looking through the robot and are actually like - Oh my God. Yeah. And so those were like, so that's one reality that was already actually like known about the Neo robot before that this is a thing and it's still a thing.

16:31And it's like, I don't have like a problem with that. Like, I think that's like an interesting topic to engage with as we're in sort of this like adolescent or like toddler stage of robotics. I don't know enough to know where we are in that, but not that's the end of it. And then also they did this like product video where it's just like the robot but also the hands doing a bunch of different tasks and this is where the sexualized um claim comes to them because one of the tasks is like unzipping a jacket or hoodies very slowly on a human um and then it's like doing some other like uh like you know like right after that it's like lifting some weight with a finger and like if you look at the youtube comments it gets like very dirty very fast and it's funny because it's like sexualized but it's like not remotely sexual so anyways so all of this like the i thought the wired article was actually very positive on the technology and balanced on the question like privacy and autonomy versus remote operating i thought the company actually had good answers around transparency and how like actually customers choose when to let a human, um, you know, come in, uh, come in like, you know, remote operate their Neo.

17:50It's not like people are spying on you and you have no, like, and there's like something that lights up for, you know, and, and, um, uh, and then the author also had watched the product video and made some comments about how it's pretty sexualized. Like, Dar did not like these questions or these comments. And I was just like, either you really fucked up because you didn't know that these were a poor topic slash possible interpretation of this video that you made. Or, yeah, you just think that you can, like, that your marketing isn't fair game for journalists to comment. Like, it just felt embarrassing or opportunistic.

18:35Maybe both. This launch got a lot more attention because of the tweet you mentioned. I mean, Mark Andreessen's quote tweeting it and saying, very concerning, dot, dot, dot, dot. Like, I'm pretty sure he didn't read the article. I tweeted about it. And I have, like, people in the replies either saying, like, oh, yeah, I, like, I actually hadn't read it, but I, like, some of the stuff Wired's written lately has been, like, really, like, you know, really egregious. And so I was just like, oh man, we're just not in an era of nuance and doing the work. It's so tribal and it's so ideological. Corporate comms and crisis comms were having a big week.

19:13So the FIA story, some good segue to that. There was a Bloomberg article. They had spent weeks investigating this and they were accused of cookie stuffing, which is something that other e-commerce platforms. But I love it. Yeah, it's basically to like, you know, sort of like misattribute and take the attribution for, you know, some clicks. And if you're in the industry, it's actually more common. However, the way the article was written, it just seems like they were committing fraud, right? And so there were people I saw sort of pointing out like, you look, Honey did this. and they were like a$4 billion company or so-and-so.

19:59These brands were also caught doing this and this is what happened. But I think the Phoebe Gates having the name, it was just sort of like a very clicky article and one that took off like wildfire. And the thing that, and Ashley, I'm very, very interested in your take on this one, but I think the sort of knee-jerk reaction I had was, okay, cool, like crisis comms mode. Like what, you know, they are very well funded. They have great advisors. Like what are they going to do? Because it seemed like they really leaned into, because, you know, people pointed out like, well, they really set themselves up for criticism when they like created like a Coachella poster to like announce all of their celebrity investors and all these things.

20:43So people were having - Sydney Sweeney on your cap table. Yeah, yeah. And they're like, how much diligence did this person do? And it's like, come on, like that's not, you know, How many of these folks really invested because of that and not because this was such a zeitgeisty moment to be a part of? So my question, because immediately, almost immediately, both the co-founders, they locked their social media profiles, right? Like these tweets are protected or something. And so they couldn't get that mentioned. And then they weren't able to be sort of tagged on social. And when I saw that, I was like, ooh, that is actually usually like step one in like a crisis comms playbook that I no longer think is really relevant.

21:27And step two is you put out a statement. And I didn't see them put out a statement. You know, today's the 14th. I don't know if maybe they eventually did. but I did see that the next day they issued a statement in response specifically to the Bloomberg article. And they said, like, within the last 24 hours, we were made aware that there was basically like a bug and it was causing misattribution from a subset of users and that they fixed this completely. I don't know how well that landed. So I just, I thought it was more interesting. Like, I didn't think anything was like that interesting except for the, oh, they used like a pretty, in my mind, like an outdated crisis comms playbook.

22:09So curious, since you definitely have more experience than me running crisis comms. I've seen my share of crises. What you thought about that? Yeah, exactly. I will say, I don't think locking down your social media accounts has ever been a gold standard crisis comms thing. I've seen it in my experience, yeah. Oh, I've seen it for sure. I think it's - No, no, no. Like I've seen the comms, I've seen the crisis comms folks say, like basically instruct that that should be the first thing to do. I've dealt with this at companies and then like, you know, guiding companies and it almost always just makes people angrier if they can't engage.

22:50So that feels, yeah, that I guess, I guess I disagree with that, with that playbook, even in a prior era, a 2026 playbook would be to like say nothing, be totally unapologetic, and then like announce your next revenue milestone or your next like valuation increase. I thought the statement was a huge mistake because it was just trust eroding. Like either, no one is going to believe that you were accidentally, like you accidentally, you know, had this sort of like cookie backdoor. It's not like they treated it like it was a bug. And that is very hard to believe because to your point, this is actually like a longstanding, you know, shady practice in this category.

23:41And like, you know, honey was like faced a lot of legal issues and reputational issues over a period of years. Like this is like a serious thing. And so no one's going to believe they accidentally like, you know, got attribution for, you know, for all these products that they didn't actually recommend. And so to say, oh, yes, we were made of the bug and we fixed it, either signals that you're lying or that as leaders, you are like totally unaware of what product decisions are being made at your company. I actually think they would have been better off not commenting, at least you know in the context of that article and so yeah so i think like i think it you could hurt them the interesting thing is the broader conversation around this piece was really about her being bill gates daughter and like yeah some people read that as you know that means she should be you know under more scrutiny and you know she thinks she can get away with this because she's like a billionaire's daughter.

24:50Some people were saying it's unfair. She's being targeted because she's Bill Gates' daughter and because she's pretty and whatever. And so the argument around this, this piece also got a lot of chatter, but it was more like the meta commentary around whether Bloomberg should have written this in the first place. My favorite detail is that like Capital One was like a main character here in like taking down Fia. Like I guess they have a competing product called Capital One Shopping and saw what was happening and shared this information with other researchers and brands and whatever. And I'm just like, oh, this is so funny.

25:29Capital One was salty and took down the shiny, buzzy Bill Gates daughter company and it's not a dynamic you see too often. So that brought joy to me. I worked at Plaid and Capital One was definitely like a foe for a brief period. They most recently invested in a series A of a company that I invested in in a previous round. So I got a little sense of their diligence process. Somebody explained this to me about Capital One and it made a ton of sense. Capital One was like the kind of original disruptor bank. Like they think of themselves as like this startup bank that like was a disruptor. So like even 20, 30 years later when there are other disruptors, they're like, oh, hold up.

26:17That's so funny. It explains kind of like their ethos and like a lot of the things behind some of the things that they do. I love knowing that because I think of them as like the stodgy brand that's like not on my lawn, you know. Totally. And they're like a challenger brand in their minds. That's hilarious. All right, well, go Capital One. Just like do your thing. A favorite, you know, sort of often discussed character in our great chat, great chat at least, is Nikesh Arora. And if his name doesn't ring a bell, he's the current CEO of Palo Alto Networks. He's the former president of SoftBank. And he was a longtime executive at Google.

26:56I think he was there for 10 years. And when he left, his title was like SVP and chief business officer. But he essentially ran like global sales. He has been like a new Twitter character for those terminally online. and tweeting a lot, engaging with users. He's got a ton of super fans. I mean, I'll count myself as a super fan. Like I've been following his lore for decades now. But the thing that we have been sort of like focused on around this is like individual and personal brands and what makes them powerful over time. And sort of like watching this moment where somebody who I would consider and many would agree, this like incredible personal brand with a lot of power and a little bit of mystery to them because they're inaccessible.

27:44Right. And like, also I really want to say, I don't think I'm imagining this. Every photo you could have ever Googled of him, like found on the internet of him was like him looking like really stern and like powerful and serious. And now every photo I see of him, it's him like with this giant toothy smile, just kind of like, what's happening? But anyways, yes, the joke is that I'm crashing out over this because to me, he's trading this like really powerful personal brand for followers and Twitter clout. And so it got us thinking about like, well, we're actually watching this happen in like, you know, sort of in this phase.

28:21And maybe it's just like, I'm too much of a boomer. And like, this is just, you know, Nikesh and Benchmark and Sequoia and all these like legacy, like powerful brands in tech. decide very intentionally, like, yeah, the new power is reach. The new power is sycophantic followers, you know, by the several hundred thousands and you have to be accessible. You can't just be gatekept behind, you know, articles or exclusive interviews or closed doors, you know, events and lore. It really has to be like front and center. You're tweeting all day, you're tweeting all of your thoughts. Yeah. You know, I thought that would be a fun topic to kind of poke at, frankly, ourselves or me on like, why is this such like a never meet your heroes moment?

29:12And like, why it's almost like making me rethink everything I ever thought about, like building personal brands. And was mystery like a moat? I know that sounds strange, but it's like, to me, mystery is part of like, you know, like what makes a brand sort of, ooh, inaccessible, right? It's not for everyone. And, you know, an example just to kick us off is like the benchmark website, you know, that has famous, the benchmark website is famously used as an example of like, it just being like plain. There's no bios of the GPs. It's just literally just a dark blue website that says benchmark and it's got like an address on the bottom and that's it.