Too late to AI proof your career?

0:28Welcome to Season 3 of Great Chat. and fund managers. Try it for yourself. Visit mercury.com to join over 200 ,000 entrepreneurs who use Mercury to make managing their money unbelievably simple. Hey guys. Hi ladies. Happy season three. Another season. Speaking of seasonal, Sally, I like your sweater. Oh yes. Thank you. Shout out to our dear friend Nagarin for this new sweater collaboration with The New Yorker. I've seen that all over Instagram. It looks super chic. You look great in it too. It's so comfortable. I think I'm going to have to get one. My favorite is that she got Keith McNally to be like one of her model influencers.

1:14And the story of how she did it is just like, it's just speaks to what a determined person she is, but truly like the best, the best unveil. Oh, it's been an incredible journey for her. And it's the New Yorker's 100th anniversary. And so the fact that they are doing a collaboration with her is very, very cool. I love this new apparel sponsor opportunity that we have on Great Chat for video watchers. Stay tuned. I feel like the conversation is all about what does good growth even look like when all the benchmarks are changing. I don't know if you guys saw the clip from Hamant, the leader of General Catalyst's interview with Harry Stebbings on 20 Minute VC, where he essentially said the era of triple, triple, double, double is over, which means growing 3x, 3x, 2x, 2x.

2:13That's no longer top growth. He doesn't even want to see those deals. Companies should be going from one to 10 to 100. And I haven't listened to the full interview, but I imagine some of this is inspired by ReCore, where GC is an investor. And I think they announced the other week that they went from one to 500 million ARR in 18 months, which is truly unprecedented. And we've seen other companies like Anthropic, there are definitely companies in this AI era that are completely breaking all prior assumptions about how fast one could grow. So that's sparked some pretty fun conversation in the timeline.

2:54And I think I saw like both sides of the debate online. And one of the downsides of this or the backlash is that some of these startups are marking up what is not truly ARR into an ARR bucket, right? And so these one-time revenue is getting mixed in. And so these financials are getting tainted. So I think people are being more disciplined about looking into this more aggressively. Right. I even saw some commentary on calculations that our startups are predicting future revenue by assuming every seeded a company is is now like using their product. And so there are some like creative numbers going on, which, you know, has like always been the case.

3:36But now it's now it's at an extreme. Yes, that is the risk with like putting benchmarks as high as they are. I mean, I think the other part of me that I mean, because we've seen earlier iterations of this, maybe not quite as extreme and held to like the Mercure standard. But, you know, I felt like maybe in an earlier episode, we said that like 5 million ARR is like the new 1 million ARR, right? Sort of like the floor requirement for Series A. And I think that is interesting because, you know, oftentimes when you have a slope that's that steep, not enough time has gone by to really understand retention.

4:11And I think that for these investments that are long-term holds, like really understanding the durability of the business is important. And I think that when everyone's now optimizing for this slope, you know, I wonder, you know, what will happen, I guess, overall, especially with these creative sort of ways that we're getting to the ARR projection. A fun narrative violation is people were really worried about that with Harvey and some of their initial growth in the legal tech space. And I saw maybe a couple of weeks ago, one of the Harvey founders sharing their net revenue retention. And it was beautiful.

4:48So, so to your point, Helen, like it can be great or it can be like, it's, it's very, you know, you can't just look at growth and know how quality that revenue is. And so it's actually like kind of fun to see some of these companies either converting that growth into like really healthy revenue, or maybe it's been there all along, but it's just really hard to parse from the outside. I think the other thing is we talked about this in the last season was experimental ARR is still in play with a lot of these companies, right? Like even look at our companies, there's a lot of different tools that we're experimenting.

5:23So even in the same category, we have, you know, three or four enterprise accounts that we probably won't renew all four of them at the end of next year. So that's still definitely in play. I think there are definitely exceptions like the Harveys of the world. But yeah. But I think companies are getting better at experimenting. I think that before there were switching costs are really high. But I think what the SaaS era has trained all of these companies to do is to get better at like sort of trying and pushing out like quickly without completely disrupting, you know, like their flows. And so that needs to actually be factored in to like, I think the arguments of, oh, well, you know, once we're in, once we have our foot in the door, we become essential.

6:05And I just think that like that's also sort of like less true because I think companies are adapting it as they should. Right. So much time has gone by since there was like the explosion of SaaS inside Fortune 500. I do think there's a really interesting question for startups, especially like very early stage startups who haven't yet chosen a path of triple, triple, double, double or 10x. We should all be so lucky as to have those choices of like, how do you project growth into this environment? Right. So, you know, especially at pre-seed or seed, you know, everyone has that like future revenue chart to like paint a picture.

6:47of what the future might look like. And of course, it's oftentimes, you know, it's a best guess. It's the best optimistic guess to give people of like, this is the upside case. I really feel for founders right now trying to decide between like, do we give the realistic version of what we think we can actually do? Or in this environment, is there more pressure to tell a story around how we can follow the growth path of Emmercore or Cursor or, you know, some of these companies that are setting new records for, you know, their path to 100 million and beyond. So I did think that Brett Taylor of Sierra made a really great counter.

7:30It was funny because like, Hemant did this clip on 20 Minute VC and then Brett was on TVPN. So just like very of these times. And he was asked about this and he had a great sort of a more nuanced response, which was that oftentimes growth like that actually shows the lack of a moat because if your company can grow like that, maybe competitors in the space can too. Speaking of best amounts of revenue, the information I think was the one that reported OpenAI's revenue was at$4.3 billion in the first half, which is like 15 % more than And their burn. Last year's. Right, yeah. Yeah. Yeah. There was a lot of announcements yesterday on OpenA.

8:18I think the biggest one that everyone was interested in was agent to commerce. I'm curious what you guys think about that. Yeah, I had so many reactions. There were actually a bunch of announcements sort of bundled into one, right? Like I am thinking about payments and the implications there with the Stripe partnership and then also just shopping generally. And I feel like one thing I'm watching people do is like draw sort of an analogy to Google and how, you know, Google Shopping and I guess their affiliate program is such a small sliver of their total ads revenue. And I think there's a lot of chatter right now about OpenAI launching ads on ChatGPT.

8:54And so, okay, do we model it from that world? And we do look at the Google ratio. But I think that, you know, OpenAI is probably making a bet that I think I saw 700 million people in the US are using ChatGPT every day or something, like a really impressive stat. Or maybe it was 700 million around the world. I forgot. It would have to be around the world. Yeah, yeah. I was like, wait, how many people live in the US? Quick calculation there. And I think that they're betting that, you know, this is different, fundamentally different from Google and that that would not be a good comp. And so, yeah, I think it's also kind of like up for the taking and we don't know where this is going to go.

9:31But I do think that one takeaway that people agree on is like GEO is like more important than ever. Um, uh, and then also bundled in that, in that, uh, I thought was interesting was the Shopify and Etsy partnerships versus, you know, I think that if this launch had happened 10 years ago or another like sign of the times to your comment, Ashley, like they would have partnered with like retailers or some like big, big brands, you know, with big distribution. And I think that launching with Etsy and Shopify, which has been, which have been these platforms that sort of like level the playing field for newer brands over the past few years, like what's telling.

10:09And so it's exciting, right? It's exciting. It feels like anything is up for grabs. So those were kind of like initial takeaways. And for those who don't know what GEO is, this is the new SEO, which is, I think, Generative Optimization Engine or AEO, which is Answer Optimization. Generative Engine Optimization. Engine Optimization. Engine Optimization. I haven't had one. The one unique thing or one thing that's going to be a differentiator between OpenAI and the Google Shopping, because I do click through Google Shopping, but once the engines understand my taste really well, right now, they've only partnered with Etsy and then they're going to start rolling out, I think, with some of the other partners.

10:54But once they have the entire commerce sector on the platform and they really know the type of brand or type of quality of T-shirt and sneakers and that I like tennis shows up, then I think it could be really powerful in that memory that Google search doesn't have. Speaking of some of the bigger players' announcements, Meta and Vibes, it got me to download the Meta AI app so that I could scroll through and see it for myself. Have you guys played around with it yet? Was it amazing?

11:33I think you said it all, Ashley. I don't see myself like going there and just spending hours watching, I guess maybe just for folks who haven't followed the news or played with the app. It's an AI only generated content feed, all video. And I guess you can like remix things from other people. I have not tried that yet. But I will say I read a piece from Ben Thompson about it, and I agree with him that it's actually kind of a relief to have to know that it's an AI only feed versus like trying to parse what you see in Reels or TikTok. talk. I appreciate that it is like the goal is not to trick people.

12:27It's to explore the boundaries of AI and imagination. And so, yeah, it's interesting. The reactions have been so interesting because I feel like it's really like, you know, two polar opposites where one, it's like, this is slop. No one wants this. Everyone hates this. And then, you know, on the other side, I think people are like, well, that's what they said about every big sort of consumer shift in the past. And maybe this this is, you know, that maybe that signal this reaction is signal that this is what's going to take off. And, you know, I think, again, like, we don't know what's going to happen.

13:01It's it's all up for grabs. And I totally agree that at least it's clear, you know, that this is that this is AI generated. I think that the reaction that I had was, and I saw somebody else had this too, where there are so many movies, like children's movies that have been in production using AI that sort of like match the style of, you know, what we're seeing on Vibes and the feed and sort of like what we got a preview of in that launch video, that the reaction from like movie studios might be like to actually move away from that. And I think similarly, like I'm craving like beautiful cinematography that is so, you know, uniquely captured in a very analog way.

13:47And so, you know, maybe that's like a boomer comment from me. But also you're seeing a lot of like kids, you know, have nostalgia over like these sort of like 2D cartoons or stuff like that. So I don't know. It could be interesting and spawn creativity sort of in these mediums of the past. I have some thoughts on this too. So that Wired is also reporting that OpenAI is rolling out a similar feature like this with their Sora 2, right? Which is like their new competitor to Meta. I think like if you look at how the transition of like social media content has evolved, we went from, you know, filtered Instagram posts, like at the very earliest stages of when Instagram rolled out to very high cinematic.

14:31And now and then we went through the phase during staff where everyone had face filters. And that was kind of a fun thing. So I do think uniquely there is a interest in AI generated slot to your point, Helen. And then the quality of these images and videos are going to get much, much better. The other thing I think I thought was interesting was, so the meta one is between Midjourney and Black Forest Labs are the ones that are powering Meta's platform. And so I'm curious how this impacts Discord because Midjourney, I think, was like 13 % of Discord's traffic in 2023. And then they got rid of their exclusivity, but still it's a big portion of that.

15:17These models have no places to host that's more consumer facing. So a lot of the people that want to generate images on MidJourney or on Black Forest Labs needs to go on, you know, FAL or Korea to create these. So having this consumer interface will be a huge boost for these image generation platforms. So I thought that was interesting. Yeah, the distribution point is so spot on, right? Like Meta's advantage is that every time they launch these apps, it's like they have this built in, you know, user base that they can tap so easily where they don't even have to go through an aggressive signup process.

15:54because it's like, oh, are you this person on Facebook or Instagram? Great. Like, here's your friend graph, right? Your follow graph. And here's all of this content that we can actually serve you and fix the cold start problem because we have all of this data on you anyways, right? So, yeah, that does seem like an advantage. Big news was that threads surpassed X in terms of total users. Which is so crazy to me because maybe I'm just like, I have a horrible threads experience. I absolutely like want threads to be a viable alternative to Twitter for us. But like, I know my feed is so bizarre. What do you, what do you guys experience?

16:32No, no, no, mine too. And then the content that threads sends me in my Instagram feed is so weird. And I'm partially to blame. It's so weird that sometimes I click cause I'm like, what is this? And so I'm probably training it that I want to see like weird things that I need to just like, let it go. I think it's a good reminder how much of a bubble access. I was in New York last week with Ashley and I was at a brunch and a few of my friends who work in tech actually had never heard of TBPN. And they're growth stage companies. And so that was an interesting data point. A friend at that brunch who has a startup also hadn't heard about 996.

17:11And I got to explain it to her. I have seen people say that it's a very, yeah, it's a very like also like San Francisco bubble thing. Right. And like, you know, investment bankers have been doing 996 or something more extreme for a really long time and never felt the need to be so performative about it. I think it was a valid point. I was randomly reading John Grisham's The Firm, which is about this guy who joins the I've seen the movie. Totally. And he put 996 to shame. He was working seven days a week, 100 hours. I'm like, we did not invent this. I think that's one thing that's funny. I think tech always thinks it's inventing new cultural norms.

17:56It's like, no, no, no, no. It's cute. But we did not. The others have come before us. One of my former employees did say she was laughing at 996 because she's like, you know what? In reality, she's really doing 8-12-7. And she's like, it's a real vibe. I hope for her sake it's 8 p.m. to 12 p.m. All right. Well, maybe that's a good segue into our main topic for today. Something that we've been like swirling about, you know, within this episode, but also just like in our group chat, and has been a good theme throughout all of our seasons, which is just AI and how it's impacting, you know, career paths and what to study and the job market.

18:36I think that sort of all of that has bubbled to the surface. And we're starting to see some actual data about, you know, what jobs, No longer just the fear mongering of AIs coming for our jobs, but I think we're starting to see some real evidence that, you know, a lot of entry level roles that existed before might not exist now and that we're seeing layoffs in a more concrete way. There was the jobs report that came out. And so that got us thinking if we should have sort of like a, you know, an episode dedicated to what we would do in this environment, you know, if we were earlier in our careers and also sort of like what we do now, because being mid-career, you know, doesn't mean that we, you know, have have sort of like cemented our destiny and sort of like we are not inoculated.

19:23That's right. That's right. We can change. And so we thought that would be a fun topic. And so I think just generally, like, how would we design an AI-proof career? I saw this question and, you know, I was thinking about, so I was a journalism major. I went to NYU undergrad. One of the best things I think the NYU program does is if you're a journalism major, you're required to double major in something else, a specialized major. I did start with poli sci because I thought like a lot of journalists who start off wanted to be a war correspondent and then pivoted to Christine. You said that. Yeah, absolutely.

20:01I grew up looking at her and saying this is my hero. And then I pivoted to finance and business. And frankly, that was that was how I got into CNBC and business journalism is because I started to understand the technical nuances of the things I was reporting on. And I think if you look fast forward to today, the journalism and writing skill set is obviously declining. You know, you can have I've seen so many journalism type newsroom tools today that integrate ChatGPT, where I think newsrooms around the world are starting to embrace where you don't need to hire 10, 20 generalists. but you'd rather have specialized people that are focused on understanding really the auto and mechanics world or space or climate.

20:50And so, yeah, I think if, again, the lens of if I was going into journalism today, I would actually focus on something that's more specialized than journalism major. And Helen, I know you're also a journalism major, which I only recently learned. in our last episode of season two. Yeah. Yeah. I think like that's, it's interesting because I think that there's the topic aspect of journalism, right? Like what, where are you spending time and specializing? So if it was political science, I think that's like a natural sort of like coupling of journalism. But I think a lot of programs have sort of seen the writing on the wall when newspapers were kind of going out, right?

21:31So I went to the University of Illinois. It was a huge, our program was a huge feeder into like the Chicago Tribune and the Sun Times, which were like collapsing in real time as we were graduating. And sort of one observation I had was, you know, these great programs like NYU and Columbia and Northwestern, like they were able to actually quickly figure out how do we supplement, you know, the degree with something that, you know, will make them more sort of future proof. And so I'm curious as to how these programs are adapting today and how quickly these students are acquiring these skills to have the readiness to couple that interest and that area of expertise with something that is more fitting of the reality of where we are right now.

22:21But my question, Sally, as you were talking was like, were people saying the same thing about calculators? You know, like, oh, well, you know, there won't be need for mathematicians. There are calculators and computers capable of doing all this computing. And I think it's like the thought and the process and maybe even dare to say like the art of a story and how that lands and how that will resonate with people. Like, I feel like that maybe is still, you know, AGI will take it on. But like, I still think that that's a valuable thing, you know, and not so focused on the output of like, oh, a perfect essay or an op-ed is now written so well with AI.

23:00I don't know. What do you think? I have yet to see an op-ed written well with AI. I have yet to see anything written well with AI. So I will just say in defensive writing, maybe not in defensive journalism. as a career path because I think it's like a very hard one right now, given the state of the industry. But like good writing and good thinking has like probably never been more differentiating, especially as all the AI slop continues. And who knows, maybe in a year I'll be eating those words because, you know, ChatGPT and Claude will just be like so good at this. But yeah, I'm just going to defend, I'm just going to defend the skill sets that one might learn in a journalism degree.

23:45But also like if you were going to be a journalist, you'd also have to learn about like distribution and audience building and so many things that I'm guessing were not a big part of your majors back in the day. Like it's just a different toolkit that's needed. It got me also thinking about, we spoke to Rachel Prophet, who's the CEO of Cooley in season one, season yeah season one and these services industries like investment banking or law you know we always talk about these tools like the harveys of the world that are eliminating a lot of the entry-level jobs but you're gonna need the high-skilled md or the big partner at these firms that are winning the deals and bringing these people in so i was thinking about and i don't know the answer like how do you develop the like people skills to are you the like top one percent of the law you know of law school and then you you know learn your people skills and your sales skills or do you go about it a different way I don't know it got me thinking about that I love that we talk about that a lot um at my fund in a health care context like the idea that what is valuable is either the people part of things like the doula who like shepherds you through a healthcare experience.

25:04And then there's like the surgeon who has like very specific technical expertise. And I'm, I'm curious if like that dynamic plays out in other industries as well. So maybe the people who are really good at sales or really good at customer relationships and the people who have very targeted expertise. Um, I really like, and maybe I just like want this to be true, but I really think this is like a great moment to be a generalist. And part of that is revenge. When I was applying for college back in the day, like everyone wanted you to be like the top 0.01 % in one thing versus like well-rounded and good at a lot of things.

25:45And it feels like something that like kind of swings back and forth. But I really do think that it's very hard to predict the future right now. It would be very hard to pick a very specific career path right now. And so actually having sort of a broader purview, having the ability to learn how to learn, which is supposedly what a liberal arts degree does, we can debate whether most schools actually do a good job of that. But I was reflecting on this question for myself. I double majored in government and English. And I didn't even think about applicability to the real world because I was going to go straight into law school and then had like a quarter life crisis and ended up moving to San Francisco and taking the first job I could get, which was at an entry level PR firm, which was my accidental pathway into tech.

26:38And the reason that worked out for me was because at the time I was doing that, there were lots of entry level roles available to people who had good enough grades who they believed they could train up to do this like very basic job right I I literally studied a printout of the Wikipedia page for PR on my way to that interview I knew nothing about the craft but they were used to taking people like me and sort of explaining how the job worked and and then had like you know a little ladder you could climb within that the challenge is like does that still exist for people like can you come out with a liberal arts degree but no applicable skills and I think like my my hunch is probably not and so I think I think like I'm still pro-liberal arts degree but I think you need to be like doing things alongside that right like whether that's like building like playing around with like building apps like launching your own podcast like I just think you can't go out into the real world anymore without any understanding of like what, like I really was like this, like naive little liberal arts baby and it was okay.

27:48Like there was a path for me. And I think people just aren't willing to do that work for, for people anymore. Like you need to have some tangible skills, um, to be, to be relevant in these, in these like, you know, dwindling entry-level roles. Oh, interesting. So I, I would take the counter on that. I still think liberal arts is relevant and valuable as long as you're like this smart, charismatic generalist who can still, you know, provide examples of how you're good at learning, right? And so I was, I saw a Sam Altman interview where I think somebody asked him like, you know, AGI is coming, how, what skills are you teaching your son, right, to prepare for this world?

28:33And he, you know, he said kind of what you just said, Ashley, like getting really good at learning how to learn. And I think that whether it's you're building something on the side or, you know, like showing, you know, initiative, right, demonstrating an ability to learn quickly and well, and just having like tangible proof points, right, that you can do it, I think, is almost just as good as actually, you know, developing like another sort of like entirely separate path and parallel so I think it's less mutually exclusive but I'm still for like oh I know there's a psychology okay okay the psychology of like selling yeah the psychology of like trusting someone and making a bet on them I think that like those things like human behavior and psychology like take a lot longer to really change even though technology is moving so quickly and everything's moving so quickly I think the time it takes for someone to trust another person or to trust a company and, you know, buy something or trial something for the first time, like that's not changing that quickly.

29:42And so I think as long as that is true, people who are, you know, just like magnetic, you know, like those people you meet where you're like, oh my gosh, like that person, you know, they just exude charisma and people are gravitating towards them or that person is their time to trust is shorter than the average. There's somebody, there's an emerging manager I just gave a reference to. And I said that I was like, oh, you know, like the time to trust is very short for this person. Like they, they, they, that's a really interesting and amazing quality about them. Um, I think those things actually become much more valuable.