23 AI Trends keeping me up at night

0:00What's up everyone? Today I'm going to talk about all the things in AI that's keeping me up at night. I've got a giant list of just things that I can't stop thinking about all the opportunities, some things that scare me, some ideas that you can take and if you stick around to this whole episode uh maybe you'll be an insomniac like me, maybe it'll get those creative juices flowing, maybe it'll just have a you'll have a better sense of why I'm so excited about, you know, where we're at right now and some things that also freak me out. Um so I just figured I'd go up go on here and

0:31sort of reflect on a bunch of the things that are that are keeping me up at night, that are just making me motivated, that are that are interesting to me. And maybe it'll interest you, too. And, you know, if you're listening to this, I think I think it probably will. I think, you know, you're probably one of those people that, you know, see a lot of opportunity, might be, you know, 90% see a lot of opportunity, 10% a little scared, um but you're also looking for some ideas to help you move along and and

1:00make progress.

1:09The first thing that's really, you know, I'm thinking a lot about right now is like the one-hour company stack. So, you know, you grab an idea, you vibe code something, you build a landing page, you you know, you you you add a stripe and you can get first customers. I mean, the fact that you can do this, just the fact that this exists, that you can go to ideabrowser.com, get a validated idea, and just start vibe coding things with whatever vibe coding tool you want, is mind-blowing. The fact that you can create a company in a day. So, I think that like from my perspective, I'm

1:41trying to think about how can I, you know, throw a lot more experiments against the wall, you know? Um what are you know, I don't want to just build one company, try it for 6 months. I want to just create a culture of and a machine that I'm creating multiple companies trying different things. Same audience or multiple audiences. We're going to talk about audiences later, but this whole idea around the one-hour company stock, I mean, can't stop thinking about

2:08it. Second thing is, you know, if you think about the old timeline, this relates to the first the first point, the old timeline of building a company was you had an idea, you hired devs, that took a few a few months if you can find them. You built an MVP, if you're lucky that, you know, you can do that by month three. You launch it, maybe you go to

2:27Product Hunt. And then, you know, eventually you get to, you know, first revenue by month 12. You know, in 2026, you know, you can have an idea or grab an idea from Idea Browser by 9:00 a.m., have something built by 9:15 a.m., have a product built by, you know, 9:45, you get the first customer by 10:00, and you iterate by lunch. Someone is going to respond and be like, how is that possible? That's

2:51vibe coded slop. There's a few reasons how this is possible. One is use a, you know, you're not you're using a vibe coding platform. I shouldn't even say vibe coding vibe agent agent engineering platform. You're using something like Claude Code. Um or or or, you know, there's competitors, too. Um And, you know, you you you build something comprehensive. I mean, Codex has gotten pretty good, you know, Google AI Studio has gotten pretty good. Um Claude Code's gotten pretty good. So, just the fact that you can do this with one of those tools is awesome. The second thing is, you know, you do have to have an email list. You do have to

3:32have an audience. You do have to have some customers in order to actually get them, you know, cuz uh otherwise, you know, as you know, you know, finding the customers is is really hard, but if you have been building distribution, if you have been and that's another thing that's been keeping me up at night is just using using AI to

3:51build distribution. So, yeah, the old time versus new timeline definitely something I've been thinking about. The other thing that's keeping me up at night is this concept I'm calling ambient businesses. So, you know, ambient businesses that run with zero or very low daily human input. Input. So, having agents monitoring market, identifying opportunities, executing executing for you, you know, handling customer service, and basically setting up a business that you just check in once every few days,

4:27and just see what's going on. I think that we're going to get to a point where these ambient businesses or autonomous businesses are going to start doing seven, eight figures. So, this whole concept is just really, really cool to me. I think we're really early. I think that a lot of the autonomous company builder softwares end up building a like very AI slop stuff. But I do think that you know, this direction, I like to think about it as like the arrow of progress. The arrow of arrow of progress is moving us in this direction of you're going to be building ambient or autonomous businesses. You

5:06won't need to check in every single hour. You're going to have checks and balances that are going to steer your agents in the right places, and I think there's just a huge opportunity there. This timeline, the agent economy timeline is something that also keeps me up at night. So, between 2009 and 2005, sorry, 2009 and 2015, that really was the App Store era. You know, people downloaded apps and humans operated them. Then in 2015 to 2024, I think the

5:37API economy really started to take over. You know, developers were wiring APIs together. I believe that, you know, 2025 to 2030, the agent economy is is here. So, agents discovering and hiring other agents on the fly. So, fixed text acts dissolving. I actually think that there's like a huge startup idea for someone to build the glassdoor of AI agents. So, how do you, you know, create reputation

6:08around agents and who to hire? If someone could, you know, create a marketplace like sort of like a mult book that was a social network or is a social network for for agents that got acquired by Meta for allegedly $200 million. You know, what is the glassdoor of AI agents? I know this sounds like such a far-fetched idea, but this is going to happen. I read a stat I think that said I think it was by Gartner, 20% of commerce by 2030 will be agent to agent, machine to machine. So, how do you create startups that that recreate the you know, that look at what internet products

6:52and just create the agent version of it. So, I think um there's you know, this market is going to 52 billion in 2030. Today, there's 31,000 agent skills on marketplaces, but they're pretty bad, you know. Most of them are are actually garbage. So, I think there's a huge opportunity to build skills, build agents and you know, it's giving me bags under my eyes. So, this whole concept of agents hiring agents, CEO agent, sales agent, dev agents, marketing agents. I recently did a uh tutorial on how to use paperclip, which

7:29is a you know, speaks to this concept. It's an open-source technology. Go check that out if you haven't. But, basically creating an org chart like a serverless function where agents spin up subtasks and shut them down when they're them done. I mean, this this is just such an interesting idea. Instead of actually prompting via a jobs to be done framework, it's like with humans, how do you actually just hire agents who manage other agents to get the job done? Super super interesting and

7:59you know, a lot of opportunity there. YC predicts that there's going to be 300 plus unicorns in vertical AI in in in in you know, this decade. So, vertical software huge you know, huge you know, huge business Constellation Software I think they own like 500 plus companies or something like that. They've built you know, SAS for vertical you know, in the vertical software space. So, like super boring workflows in you know, education in defense in you know, all

8:34all sorts of like really boring stuff. Um and uh there's the same opportunity to sort of build your own Constellation Software but in vertical AI. So, I think that you know, if you're listening to this thinking about like what do you have an unfair advantage? What is your niche that you understand really well because I think that the people that you know, are building in the vertical agent map I think there's just a lot of opportunity. You know, the YCs of the world are going to focus on these big big categories insurance, real estate, logistics, elder care, legal, health care, sales. Um I'm not suggesting that

9:12you go and do that. I think that you should pick a wedge in a sub niche of one of these you know, vertical categories. You know, start there and then sort of expand from there. You know, there's going to be so much funding into these into these big categories. So, you kind of want to you know, just pick something that has a a little less uh

9:35competition. Um but I do think that there's a lot of opportunity here. So, you know, how do you think about vertical SaaS versus vertical AI? So, this is something I'm thinking a lot about. Vertical SaaS captures a fraction of IT spend. You're generally selling software licenses. Um, humans are operating the tool and you're looking at the 10 to 100 million dollar outcome usually. Of course,

10:00there's uh exceptions to that rule. Uh, vertical AI taps directly into labor P&L. So, what I mean by that is you're getting uh you're building basically an agent as a software. Because you're doing that, the thing that uh pe- uh companies are hiring you to do is what they would hire human beings to do. So, the actual market for vertical AI is bigger than vertical SaaS. You know, you're going to want to think about selling outcomes and results. Agents are doing the work and because of this, I think that the outcomes in vertical AI are on average going to be bigger than they are in vertical SaaS. So, unlike

10:37SaaS, uh which captures IT budget, vertical AI replaces headcount and that's just a 10x bigger total addressable market. So, yeah, if that's not keeping you up at night and all the opportunities here, I do not know what is. So, what are some boring gold mine verticals? Um, you want to look at stuff that just runs on phone calls and faxes and boring stuff. Uh, these are all big categories like insurance, but you know, insurance still uses 30-year actuary tables, legal, uh logistics, elder care, government accounting, construction. You know, you're going to want to pick stuff want to pick stuff with, you know, in these sub niche in these niches, but sub niche

11:20very, very sub niche down. And you know, if I were you, I'd try to pick something that doesn't have a lot of red tape. Obviously, selling to government is really hard. Um, uh so, you know, things like that is are probably less interesting, but you know, the more boring the better and the more niche the better to start. Uh, so SAS is really uh evolved in terms of like how we've been pricing. So, it used to be that you would do per seat licensing, $50 user a month or whatever. All these All these, you know, big SAS companies did this and that's why you're seeing a

11:57huge correction in the stock market with Well, two reasons why you're seeing a huge correction in the stock market with SAS companies. I mean, some of these stocks are down like 50, 60% huge companies that were once trading at like 12x revenue are now trading at 4x revenue on billions of dollars of revenue. And why is it happening? Well, it's happening because uh you know, there's going to be less seats, one, and then two, people are scared that or investors are scared that you can

12:25just vibe code these solutions. Um, so yeah, go you know, the evolution is going from went from seats to usage-based, so pay for what you consume, to now you're just seeing a lot more outcome-based stuff. So, pay per result delivered. And why are you seeing that? Because you're getting you're having agents actually do the work. Um, so yeah, Gartner says 40% of enterprise SAS shifts to outcome-based by 2030. So, seat-based is going to decline from 21 to 15%. So, you know, where is the opportunity? Why is this keeping me up at night? What are What are ways to go and create outcome-based pay per result businesses right now? Um,

13:06there's so many. There's so many. And if you can be the first to market, you have an advantage there cuz when you're reaching out to people, you know, via cold or even if you're just posting on your email newsletter or on your, you know, social media accounts, you know, there's there's it's interesting to people, right? So,

13:25uh you might be able to sell a lot more. So, this whole seat-based versus outcome-based, you know, shift is just fascinating. The old way of $100 a seat a month, pay when whether you use it or not, uh you know, you get you pay 10 seats, you get $1,000 a month, and like the value is unclear. I think we've all kind of felt this. You know, there's some I won't name names, but there's some software that uh my company my holding company like Checkout pays for, um and I'm just like do we Is it Do we Are we getting the value out of this? Um you know, we're

13:59paying thousands of dollars a month. So, this whole idea around, you know, maybe you pay $1.50 per resolved ticket, um pay only for results, um and you have companies like Zendesk, which is pretty big already doing this. Um you know, 83% of AI native SaaS has already switched. So, my you know, my big point with all this is someone's going to build a billion-dollar business doing nothing but converting converting legacy SaaS to uh outcome pricing. And uh I think there's just a lot of opportunity to help them do that. I also think there's a ton of opportunity to just go and build outcome-based startups on your own. Like, why even

14:40help them um when you could be uh when you could just be incubating yourself. Um I definitely think there there's something to be thinking about. So, uh I have I have to mention like I do think that there's going to be somewhat of a SaaS graveyard, but you know, I started to think about like what is the framework

15:00for thinking about what's going to die? So, I think generic CRMs are probably going to die. Um I'm not saying that Salesforce is going to die. I'm not saying that HubSpot is going to die. Those companies, you know, you can see see the writing on the wall, and they're moving towards that future. Um but I am saying that if you're generic, um and you're not moving to this future, you probably won't you probably won't

15:25work. So, agents are going to do it better. Basic analytics dashboard, not those businesses probably not working because AI generates insights on demands. Template marketplaces, very tough business to be in because AI is generating custom templates in instantly. Scheduling tools, I don't know, right? Like where's the future of that when agents handle calendars and just basic customer support, you know, chatbots are already replacing

15:54this. So, what is going to survive here? Well, basically, vertical workflow tools that pivot to agent companies, infrastructure and data modes. So, there's basically this scarcity flip that's happening. So, what is being uh commoditized by AI? Well, code, generic content, uh sorry, generic content, basic design, data entry and routine analysis. So, what's what is scarce and premium and what I keep thinking about and a lot of people talk about this on on Twitter is that, you know, the the value is going to migrate from execution to judgment. So, creative judgment, human-made crafts, physical experiences. I'm looking at incubating stuff here. It's really you know, a huge opportunity. Original

16:42weird thinking, just like being weird is going to sell uh in 2026 and beyond. Uh because a lot of these LLMs are just they're not good at being weird and you had a you know, you have a unique uh purview into life. You've experienced certain things. So, like leaning into your weirdness and then proprietary data. Uh what is the premium stack um in terms of you know, what are people going to value the most in an AI world is something that I'm thinking a lot about and I think that the most premium is going to be

17:21human-made. I don't know if you saw the Porsche campaign where they did a 100% human-made ad campaign. So it was like a race to establish an AI-free logo. I could see that luxury brands are going to lean into human-made and no AI involved. Think of like certification labels like organic for food right? No AI. So that's just something to think about in terms

17:48of like ideas for building stuff. Premium no AI involved. Most premium. The premium would be AI-assisted but human-led. So having some human in the loop I think is going to be seen as premium in the AI age. So you get human taste with AI speed. And then I think a commodity you know from a from a perception perspective will be like I'm buying just a fully AI service. And then I think there's sort of a race to zero pricing

18:19in some categories. So I you know I mentioned that I was interested in incubating and investing in you know basically IRL stuff. And the reason why is when digital is infinite and AI generated scarcity shifts to physical presence with other humans right? So I think things like you know karaoke bars you know escape rooms immersive theater you know co-working live music you know this whole experience economy is already here and accelerating. And there's just a ton of opportunity here. And that's keeping me

18:56up all all night. Another interesting concept is is this concept called founder agent fit. So you know when I was on the you know on the come up uh, to speak Um, it was all everyone just kept talking about when I moved to Silicon Valley. Everyone kept talking about founder founder market fit. Do you understand the customer and the market? Do you have Do you have some insight as you as the founder into the market? So, if you, uh, were building a social network for college students, you know, were you recently a college

19:30student or are you a college student? And I believe that, you know, where we're going is founder agent fit fit. So, can you orchestrate a fleet of agents, you know, towards your goal? So, uh, you know, the bigger shift here is thinking as yourself as a film director. So, a film director is not holding a

19:51camera. The film director is not acting. The film director is not writing the score. Uh, you know, they get performances for actors. They're trying to get the most out of their actors. Now, the actor is, you know, shifting from a person to a a machine essentially. So, I think that this that founder skill, that founder agent fit, is just an interesting, you know,

20:19shift that's happening. Um, and if you're really good at, you know, building agents for a particular niche, managing them, getting the most out of them, uh, then you have an unfair advantage. We talked a little about this, uh, with regards to paperclip and and, you know, zero human companies, but you know, this whole idea of a ghost team org chart. Like, the fact that, you know, in the future, you know, I imagine a team page, you know, you go to a website and you go to the about page and then you click the team section, maybe, and then you see all these people and and their and their

20:54big smiles and you really get to know the team. But, the future ghost team, you know, might be just a couple people and a bunch of AI agents, sales agents, content agents, customer support. Uh, and those are maybe you name them. Uh maybe they have personalities. Uh maybe you end up even creating images for who these people are. Um and eventually they're going to talk back to you, right? They're going to be able to video chat you. They're going to be able to send you voice notes. Like it's going to be exactly Well, not exactly, but it's going to be in the ballpark of working

21:29with a human being. Uh so, it's just, you know, this whole idea of ghost team or chart, I mean, is crazy. As a As a person who's building a holding company and incubating businesses, um I think there's going to be a lot more holding companies uh because you're going to own a bunch of AI native agent businesses

21:49uh in similar niches or the same niche. Um and you're going to have these ghost teams running it. So, uh I remember I Kevin Kelly uh talking about the 100 true or the 1,000 true fans. Uh and if you're listening to this, you've probably heard I've heard of that. Um but I think that, you know, it's not necessarily a 1,000 true fans anymore in the AI age. And this is something I've been thinking about. I think it's more like the 100 true true fans. Because agents are cutting your costs so dramatically that 100 people paying you is a real

22:30business. And because of the fact that you can build software and charge, you know, $1,000 a month or $500 a month because they are doing the work of potentially human beings, then, you know, there's a way to build a huge business there. Also, you can they don't even need to pay you that much, right? You because you can run it with agents. Um and you your team needs to be so small from a cost perspective, could only be

23:00to right? Then, I just think that there's this like world where you create these like micro monopoly maps. So, you know, I'll go through I'll go through what I mean by this. Let's say you have a 500 5,000 engaged niche audience. You build a custom app. Maybe it's in 48 hours. You know, if you have an audience or if you have a newsletter or something, you can get to 100 customers over $50 at $50 a month. But, then you're running the business with with agents. So, uh and then you're making about $60,000 of profit for one person, which is incredible. And then you can go and

23:33incubate multiple of these and expand this one. So, yes, you do need to get 100 customers at $50 a month. Um and that's why, you know, I believe uh you know, building media and building content and uh understanding how to build a machine that, you know, creates high-quality meta ads is is really helpful here. So, even if you don't have an audience, you know, you can pay for them. And yeah, it's going to cut into your profits, but so be it. Um one thing that, you know, we've I've I've been pretty optimistic uh during, you know,

24:07dur- up up until now. But, uh I will say one thing that kind of freaks me out is the the agent attack sur- surface. So, uh you know, I'm sure you've heard of prompt injections, things like poison context windows, malicious MCP service, a- uh agent-to-agent manipulation, permission escalation, compromised training Basically, because we're giving access to uh you know, so much through our AI agents, I would be lying to you if I said this didn't freak me out that bad things, you know, are going to happen. And I think bad things are going to happen. You know, I think that cybersecurity hasn't caught up to this how fast we're moving

24:50in this AI agent world. And because of that, I think uh you know, some bad things are going to happen, so of course it's going to keep me up at night. Um, you know, Palo Alto Networks uh, documented real-world agent injection attacks. And uh, you know, if Palo Alto Networks is saying there's going to be a bunch of real-world agent injection attacks, well

25:16I I definitely trust them. So, how should we think about agent injection versus fishing? So, like you know, in the in the in the past, if you think about fishing, uh, like say 2010, it was basically like how do you trick a human being into clicking a bad link. You know, targeting email inboxes, human judgment is the defense, right? So, if you're if you had a good eye for fishing, like chances are

25:42you'd be okay. Even with that, billions were lost per year. So, you know, agent injection, where we are today as of recording this, you know, you could trick an AI agent via hidden instructions. It targets context windows and web content. The agent autonomy is the vulnerability, and I believe that the potential is far bigger than fishing. So, where agent where agents have system access and make autonomous decisions, poisoning their context window, I guess is the new fishing, right? So, I think it can be a lot more dangerous. I think a lot of bad things are going to happen. I do think there's a ton of opportunity to build, you know,

26:21cybersecurity soft software that helps with this. Um, so that's a you know, a whole rabbit hole of startup ideas I can go down. Uh, but I I think it's something to be aware of. The other thing to be uh, thinking a little bit about is the agent permission stack, right? So, what can your agent access? You know, files, emails, calendars, bank accounts. People are giving bank account access to some of their agents, right? You're seeing them, you know, here's here's $5,000, go

26:51and trade for me. You know, what can your agent remember? Conversations, personal data, business data. What can your agent do? Sending emails, make purchases, modify code, delete data. What can your agent share? You know, for example, with other agents or with third parties. The point here is you're going to want to do some digital hygiene. So, quarterly agent cleanses, review review permissions like you review app access

27:18on the web. So, you know, sometimes I'll go into some of the some of the SaaS's I use and I'll be like, "Yeah, this particular, you know, app really doesn't need access to it." So, I I I dislo- unlink it. And I think we're going to do a similar thing with agent permissions as well. I believe that now in a in a in the in the AGI, the build cost is basically zero. Agents are doing a lot of the work. There's a ton of niches that are wide open and audiences are underpriced. I don't believe that this is going to last forever, but this is what's motivating me so

27:57much. I think that there's 12 months where competition starts catching up. Some of the best niches get claimed. Some of the tools get crowded. I think there's probably 24 months where the window narrows. The builders who get started now are going to start owning moats around data, network, brand, trust. So, people keep waiting for things to settle down. Things are not settling down. This is the new normal. I think

28:22that there's a so much opportunity here. And that's why every day matters so much. And, you know, you definitely can listen to the podcast Startup Ideas podcast, this podcast right here, because I'm just sharing things in real time. And and just trying to help help help you, you know, increase your probability of success with ideas, tactics that I see

28:44working in real time. Uh believe that this window is asymmetric. So, uh you know, what I mean by that is what you need is an API key, some prompts, a tweet, a niche audience of, you know, as we talked about today, like pretty small, 100 to 5,000, and there's this asymmetry. So, what you can get as a business that runs 24/7, you could

29:07create a business that has 95% margins. Um you could, you know, it's it's not crazy if it's if it's agent first. You know, okay, could it go to 70%, 80%, 60%? Yes, but these are incredible businesses that you can be creating uh with compounding distribution and zero or a few employees. Um so I believe that this is the most asymmetric time to be building a startup. What people are saying now is don't build in public, don't build in public. Um it used to be build in public, build in public, build in public. And I think that you, you know, it is helpful to build in public. And people say don't build in

29:45public because you're inviting competition. I think that the benefits outweigh the cons. Um specifically like building for, you know, if if your followers and your audience is actually your customers, then you could share what you're building, the community could vote on what they're on what you're building. And what's so cool about this AI age now is you can ship ship updates in a day or

30:082 days or 5 days. Users are basically becoming co-builders, and that just increases trust and distribution compounds. So, it creates this really cool uh flywheel when you're building with your audience. Um I also believe that, you know, forking a business, like how you fork fork a repo on GitHub, uh is going to be very common. So, you know, in a world where you can just copy other people's businesses really quickly, bringing in a community and and making them feel like they're a part of building what you're building, I think it's going to be absolutely get uh a huge moat and and really important. And there you have it. I mean, like I said,

30:50this is an incredible time to be building. There's just so much so many things happening all at the same time. Uh it does feel overwhelming in a lot of ways, but if you just get to work, start making progress every day, realize that you're not going to understand every single AI tool on the planet, understand how to use it perfectly on the planet, realize that you're just learning as you go, you're building as you go with momentum, every day better than the next. Um I mean,

31:22what an incredible time to be building. Um and uh let's do this together. I'll see you on the next episode and thank you for less listening. Hope this got your creative juices flowing and uh I'll see you soon.